Taking Australian diplomacy digital

What’s the problem?

Australia’s Department of Foreign Affairs and Trade (DFAT) now has a presence on the main digital platforms, but it is yet to master digital diplomacy: using these powerful new communication tools and platforms to better conduct its core mission of persuasion, influence and advocacy. There’s too much use of new media channels to transmit old media content, a tendency to duck rather than address difficult issues, and a failure to engage within the digital life cycle of a news story.

Data analytics and the integration of digital tools into mainstream diplomatic campaigns are both lacking. Beyond this, there’s a need to rethink how Australia does diplomacy in the digital age.

DFAT needs to find better ways to communicate with its stakeholders, using digital tools. It needs to recognise that increasingly statecraft is playing out in the cyber and information domains, and invest more in equipping itself to engage in those domains—even when such online engagement brings risk. 

DFAT must also reconceive its overseas presence and embrace some of the agility and nimbleness of the tech world in doing so.

What’s the solution?

DFAT needs to start treating digital diplomacy as core tradecraft, rather than optional add-on. It should provide compulsory digital training for all outgoing heads of mission and encourage healthy internal competition and innovation. It should pilot more sophisticated data analytics tools and integrate digital tools into regular diplomatic campaigns. It should develop and pilot a new stream of diplomatic reporting that’s punchier and timelier, and reaches a broader audience on hand-held devices.

DFAT should create new positions of ambassador to Silicon Valley and ambassador to the Chinese tech giants based in Beijing. It should experiment more with ‘pop-up’ diplomatic posts, pilot one-person posts and encourage innovation and experimentation in the conduct of digital diplomacy, conceiving of embassies as hubs and connectors for a broad set of interactions. 

Finally, DFAT needs to adopt some of the nimbleness and agility of the tech world in how it conducts Australia’s external policy. Failure to do so means the field is left to others.

Introduction

Australia’s DFAT has come a long way in a short time in its embrace of digital tools and technology.

DFAT, and most of our embassies around the world, now have a significant social media presence, often across several platforms (Figure 1). There has been an explosion of Twitter feeds, Facebook pages, Instagram accounts, and even blogs and YouTube channels,1 adding colour to what was (and remains) a rather lifeless website-only presence. In this, DFAT has been helped by political leaders who have embraced these tools as a means of modern-day communication.

After coming late to the game,2 DFAT now has a decent digital presence when benchmarked against other foreign ministries worldwide. It’s certainly not in the top 10, but it is credible.3

Figure 1: DFAT’s social media presence

Digital, but not yet doing digital diplomacy

However, in the rush to embrace digital media, there’s a danger that some of the bigger questions have gone begging, and that ends have been confused with means. Doing digital diplomacy well not only requires having the requisite digital platforms—it entails using them strategically and effectively to advance a diplomatic agenda.

This is where DFAT is struggling: it has gone digital, but it isn’t yet doing digital diplomacy. Having a large number of social media accounts and a growing crop of followers or friends isn’t sufficient. The test of success is whether those factors are being properly utilised to bring Australian diplomacy from the analogue into the digital age.

A changed operating environment

The essence of diplomacy hasn’t changed. Its main purpose remains the facilitation of communication between states and the exertion of influence (on other states or the international system) to protect and advance national interests. But what has changed vastly, almost beyond recognition, is the operating environment of diplomacy.

Even as recently as a decade or two ago—well within the professional lifespan of most of Australia’s senior diplomats—diplomacy as a profession, and hence DFAT as an institution, enjoyed several natural monopolies. First, there was the monopoly on information. It wasn’t that long ago that diplomats would fax press clippings or transcribe news articles and send them back to their capitals.

At a time when news on developments within other countries was scarce, and almost impossible to access remotely, diplomats stationed abroad were a vital—sometimes the only—source of information for capitals hungry for such intelligence.

Second, there was the monopoly on communication. In the era before modern modes of communication, the bulk of interactions between states took place through the medium of their diplomats. Leaders would meet or talk occasionally, but usually the challenge of making direct contact meant most communication was, of necessity, passed through ambassadors or envoys.

Third, there was the monopoly on representation. When communication with capitals was slow and difficult, and it could take several weeks to get an answer, diplomats abroad were expected to make decisions and improvise within a wide area of policy discretion.

These natural monopolies guaranteed relevance for foreign ministries, DFAT included, and sheltered them from competition. A government simply couldn’t run a foreign policy without a foreign ministry and its overseas diplomatic missions. Modern-day technology, however, has eroded most of these natural monopolies.

Diplomats no longer enjoy a monopoly on information. Leaders and decision-makers in capitals can readily access and follow most news from abroad, usually on demand, and from a variety of sources. Nor do diplomats enjoy a monopoly on communication. Today, leaders and senior officials are just as likely to communicate directly with their counterparts in another country—by phone, email, text message or, increasingly, an encrypted chat service—rather than through their diplomats.4

Finally, the monopoly on representation has ended. Diplomats are now expected to check nearly everything of significance with their capitals first, and modern communications mean they can (and are expected to) obtain revised instructions on how to handle an issue almost instantaneously.

Disruption, disintermediation and the digital pivot

The end result is that diplomacy has become a much more competitive space. Diplomats are being disintermediated by new technology and communication advances. States are increasingly able to understand, communicate and negotiate directly with other states, without the need for the intermediating service of diplomats. With the disruption of much of the traditional role of diplomats, the challenge for foreign ministries today is to pivot: to find new ways to generate value and ensure relevance in a much more contested field. And this is where digital tools can prove so important.

One of the main purposes of national security agencies is to deliver a strategic effect: to shape the behaviour and decision-making of foreign countries and their leaderships. Defence forces do this through alliances and partnerships, their force posture, deployments, joint exercises and military diplomacy (and, in extremis, through the threat or use of force). Development agencies do it through the direction and composition of their aid spending. Intelligence agencies do it through the collection of sensitive information, espionage and disruption.

In diplomacy, words are the bullets. A strategic effect is delivered through persuasion, influence, argument and advocacy directed towards a foreign population, nation or group of key actors or decision-makers. For this task, new communication tools—and especially social media—are a potential boon for diplomats.5 They allow diplomats to engage directly with the public or segments of the public in their country of posting, often in a targeted fashion. They provide the tools to deliver a message or engage in debate directly, rather than through traditional platforms.6 And they allow real-time interaction with a rapidly evolving media cycle, including the ability to rebut falsehoods, contest narratives, correct mistakes and provide the public with additional context to media reporting.

This is especially important now that political power is highly dispersed (partly the result of digital media giving each person a loudspeaker). To be an effective diplomat today requires more than just the formal engagement of your host government. If you want to be effective and shape the course of decision-making, then you need to be monitoring and engaging with those who shape the decision-making environment of political leaders within a society. That might include the media, business and industry groups, civil society, pressure and lobby groups, religious organisations, politically active diasporas and social media ‘influencers’. While this may be less true in autocratic countries, even there—thanks to social media and digital platforms—civil society has a voice that it previously lacked, and a means with which it can be directly engaged.7 Knowing and understanding the terrain of local opinion, and how to engage and shape it—the ‘last three feet’ of diplomacy8—is the unique value proposition of today’s diplomat and something that only a local, informed and networked presence can provide.

A credible but flawed digital presence

DFAT and the Australian network of embassies and high commissions abroad now have, on the whole, a credible digital presence—the tools needed to conduct those last three feet of diplomacy. This is necessary but not sufficient. The challenge is to fully utilise these platforms to conduct DFAT’s core business, which is diplomacy. And here, there’s still quite some way to go. There’s not yet a wholesale recognition and appreciation of how the advocacy landscape has changed. As a result, and with a few stand-out exceptions, most of DFAT’s digital channels suffer from the same three ailments.

First, there’s too much use of new media channels to transmit old media content. Digital media are a different format; they speak to a different audience, and require different—and more engaging—content. Good digital content is pithy, impactful and tailored, but too little of DFAT’s digital content meets that test. Using new media channels to transmit old media content (press releases and the like) ruins both.

Second, there’s a pronounced tendency for DFAT’s digital platforms to duck the difficult issues. There’s a place for building brand Australia, promoting tourism and spruiking soft news stories about Australia on digital platforms, but public and cultural diplomacy can’t be the sum total of our digital effort, or else we risk being (in the words of one insightful commentator) ‘all gums, no teeth’.9 Tempting as it is, there’s no point in running dead or lying low when a controversial issue is unfolding. This is exactly when digital platforms come to the fore and the credibility of your digital presence is tested. Too often, when a storm of controversy is raging all around them, DFAT’s digital channels bury their heads in the sand, go radio-silent, or promulgate the Panglossian fiction that all is well. If Australian nationals are set to be executed in a foreign country, or there are suggestions that the Chinese are building a military base in the southwest Pacific, or if a candidate for the Philippines presidency jokes about the sexual assault and murder of an Australian missionary, then we should expect that the relevant Australian digital diplomatic platform will have something worthwhile to say about it— to articulate our views and interests on an important issue.10 Likewise for major world events. The message must obviously reflect diplomatic realities, but to say nothing in such scenarios is simply not credible. It also lacks a prized trait of the digital age—authenticity—and so diminishes the value of the platform and treats readers as fools.

Figure 2: Twitter feed from selected foreign ministries on 12 June 2018, date of the US – North Korea summit in Singapore

Closely linked to this is a frequent failure to respond within the digital life cycle of a news story. Time differences and clearances may make this challenging, but our senior diplomats abroad have enough judgement and common sense to be trusted—indeed encouraged—to speak publicly on most issues within their patch without having every word approved by Canberra.11

Third, there’s a lack of personality in much of DFAT’s digital content. Part of the appeal of social media is its authenticity and directness—the idea that you get to know the person behind the message and can interact with them directly. But most of DFAT’s digital media content attempts to uphold the traditional division between public and private spheres. It’s stiff and aloof, and frequently non-responsive to attempts to engage. That’s an approach that may remain suitable to traditional diplomatic settings, but it jars in the flat, non-hierarchical, informal world of digital.

Operating in a new information domain: opportunities and threats

If used as part of a comprehensive strategy, the new digital world provides many opportunities to reinforce traditional diplomacy. The UK used digital tools to complement traditional diplomacy in its successful assembly of a broad coalition to respond to Russia’s apparent use of chemical weapons on UK territory, in Salisbury (Figure 3). Canada deployed a multifaceted digital campaign to support its objectives as G7 chair (notably, its initiative to tackle the problem of ocean plastics). Russia is an adept practitioner, frequently taking to digital channels to muddy the waters, promote alternative theories and create distractions when under international pressure (Figure 4). These countries have each integrated digital platforms into the prosecution of mainstream diplomatic priorities and campaigns, realising that digital tools can have a potentiating effect in support of a diplomatic campaign. In Australia, we’re yet to do this properly: we maintain an unhelpful separation between the digital realm and the mainstream diplomatic realm.

Figure 3: Part of the UK’s digital diplomatic effort to hold Russia accountable for Salisbury

Figure 4: Twitter feed from the Russian Embassy in London

Professional data analytics can be a powerful tool for this new diplomacy. Big data and network analyses can help identify online influencers and force amplifiers; track how narratives spread among online publics, and thus help to shape or combat them; allow communications that are tailored to the preferences and attributes of specific online communities; and support the rollout of sophisticated, multiphase campaigns. Most major corporate outfits use such tools, as do the diplomatic services of many foreign countries. The UK Foreign Office even has an internal ‘Head of Data Science’ position.12

Australia needs to get similarly professional and move beyond the simple counting of ‘likes’ and ‘followers’ as the metrics of digital impact.

Just as digital tools bring new opportunities to diplomacy, so they also bring new threats. They are changing the nature of statecraft, and the information domain is growing in importance as a theatre for contest between states. ‘Control of the narrative’—about what happened, about who’s at fault, about where justice lies, about what’s ‘real’ and what’s ‘fake’—is at the heart of this contest (Figure 5).

Diplomats have an important role to play here, in combating misrepresentations, squashing rumours and misinformation, and promoting their own country’s analysis and policy. Effective digital tools and good data analytics will be vital to this effort.13

Figure 5: The information domain is becoming a new theatre of state competition: textbook ‘trolling’ by two of its most capable practitioners

Similarly, today’s digital age means that disinformation, propaganda and rumours designed to influence or destabilise another country’s political system can be launched almost instantaneously, from across the globe, timed for maximum impact, and targeted towards a narrow audience (Figure 6). Unlike overt steps or traditional covert action, such measures are low-cost, low-risk and highly deniable. Russian state interference in the 2016 US presidential elections is likely to be just the tip of this iceberg.14 Although defending against such attacks is primarily the work of intelligence and cybersecurity agencies, we should expect our diplomats to be alert to the risk of such attacks and attuned to the tell-tale fingerprints. But they need to have the tools and the digital literacy to recognise, understand and engage with such information-warfare and ‘active measures’ campaigns.

Figure 6: Content identified by Twitter as originating from and spread by the Russian Internet Research Agency during the 2016 US presidential election.
Source: Update on Twitter’s review of the 2016 US election, 31 January 2018, Twitter, online.

Moving beyond social media

DFAT’s use of digital tools needs to extend far beyond social media, however. In the consular sphere, the department now does a good job in engaging with the travelling public through the digital Smartraveller platforms, but it is yet to modernise how it communicates with some of its main clients within the government. 

The Australian diplomatic network’s main form of communication remains the classified diplomatic cable or telegram. This was once one of the best—indeed one of the only—ways of communicating information and analysis from abroad in a timely and secure fashion. But while modern technology has since moved on, and the pace of events with it, the cable system has remained frozen in time. For the demands of the modern ship of state, it’s too slow, too cumbersome and too difficult to access to be of much operational use. It’s thoroughly analogue, is largely internally focused and has a steadily shrinking readership and impact.

DFAT’s continued reliance on this system as its primary means of communication needlessly restricts its audience and increasingly deals it out of policy influence in Canberra, where many of the national security agencies don’t access or don’t bother to read DFAT’s cables. The department is completely out of sync with the working habits and preferences of today’s governing class, and how they wish to receive information. It doesn’t connect easily or widely to other agencies. Consequently, DFAT’s analysis and advice from its overseas network—one of its main value propositions—is underutilised and undervalued, with implications for policy influence, credibility and the contest for finite government resources.

DFAT must create and foster new methods of communication that are timelier, more accessible and more relevant. There should be different information products for different purposes and different audiences, and the cable system should be only one of several ways in which our diplomats convey information and analysis. As just one suggestion, why not create the equivalent of an encrypted Telegram group or closed Twitter feed that allows non-sensitive but time-critical reporting from across the diplomatic network, with a smattering of judgement and analysis, to be accessed by decision-makers in news-feed style from their handheld devices? Figure 7 shows what it could look like: daily headline take-outs from across our diplomatic network, designed for decision-makers without the time, ability or appetite to wade through the cable system (but with links to more comprehensive analysis). There would still be a place for more detailed reporting and analysis (perhaps accessed via links to a secure cloud-based site), but that, too, should be in a form that reflects the habits and preferences of the readership. Newspapers have made the painful transition away from print and towards new media. DFAT should walk the same path.

Figure 7: Illustrative example of a sample DiploFeed from 2018 (fictional infographic only—does not represent the views of DFAT or its posts)

Rethinking diplomacy

We need to rethink how we do diplomacy in the digital age. A diplomatic presence shouldn’t always have to mean an embassy or a chancery, with all the expense and infrastructure and security overlay that entails. Modern-day communication tools are so powerful that we should rightly expect our diplomats to operate more self-sufficiently, just as foreign correspondents do. There are many parts of the world where Australia would benefit from greater diplomatic representation—we have one of the smallest diplomatic footprints of any country in the OECD, after all 15 —but where we have none because the entry costs to establish a full embassy are so high. Digital tools have brought those barriers to entry down. There should no longer be a minimum viable size for an embassy. We should consider an ‘embassy-lite’ or one-person post in countries where we could do with a presence but can’t justify a fully fledged embassy. With DFAT’s ‘pop-up embassy’ in Estonia, Australia has made a small start down this path. We should continue.16

Similarly, we must assess whether states and international organisations are the only external actors that are worthy of a dedicated diplomatic presence. We should look at creating dedicated ambassadors to the tech giants of Silicon Valley, as France and Denmark have done.17 The FAANGs— Facebook, Apple, Amazon, Netflix and Google—are now immensely important international actors in their own right. Together, their market capitalisation is US$3 trillion, but it’s their business model and ubiquity as much as their size that make them key actors for states. We have issues at stake with each of them—from privacy to taxation, from counterterrorism to cyber-interference and national security capabilities. Similarly for the major Chinese tech giants, the BATs (Baidu, Alibaba and Tencent), whose enduring influence might prove to be greater and about which we know and understand far too little.

Why not have ambassadors dedicated to building and managing these critical relationships, which are surely as important as our relationships with some of the smaller countries where we maintain a diplomatic presence?

In order to modernise diplomacy, Australia needs to begin envisaging the diplomatic network in a different way. Whereas in the past the government provided the network and infrastructure for traditional diplomatic interactions, the erosion of that monopoly means this network is at risk of becoming an underutilised asset. The flag and the chancery, the titles and the flummery, still count for a lot, as do the local networks, contacts and expertise, but how do we get more out of those assets?

The answer lies in broadening our conception of an embassy. We should be using our overseas presence as a platform and enabler to advance our interests across a much broader spectrum, and for a much broader set of stakeholders. Trade, economic and commercial diplomacy have always been traditional partners in this respect, but we need to look much further afield. How can we use the overseas network to support collaboration in innovation and research? How can we use our embassies to keep Australia on the cutting edge of public policy? What value or perspectives from overseas can be brought to bear on some of the major challenges in Australian domestic policy? These areas will depend on the complementarities and opportunities that exist, but they shouldn’t be treated as the poor cousins of traditional diplomatic work. The challenge is to conceive of the embassy as a facilitator of productive interaction and a broker of relationships—a creative hub of networks—and to find creative, non-traditional ways to use the overseas network to advance Australian national interests across the full spectrum.

Finally, DFAT needs to adopt some of the nimbleness and agility of the tech world. The bureaucracy is still far too slow to adopt reform and changes, partly because it insists on any changes happening wholesale, only after painstaking deliberation, and in a culture that focuses debilitatingly on downside risk and punishes failure. Why not encourage internal innovation, meaning different ways of delivering the same product? Promote experimentation and differential approaches. Test new platforms and business models. Run some pilots, iterate and adjust, gather the evidence, and see what works best. Don’t insist on homogeneity. Tolerate some screw-ups and failures and learn from them.18 This is the secret to innovation and continuous improvement, and it’s essential if our diplomatic services are to keep pace with the modern world.

Recommendations

  1. Commission an independent review of DFAT’s digital diplomacy efforts.19 The review should examine the department’s digital capabilities, assess the digital operating environment for Australian diplomacy, and make recommendations to improve Australia’s digital diplomacy effort.
  2. Treat digital diplomacy as core tradecraft, rather than optional add-on. Provide compulsory digital platform training for all outgoing heads of mission.
  3. Encourage healthy internal competition and innovation. Generate a monthly scorecard highlighting the best digital performers and posts. Promote and celebrate the successes.
  4. Pilot more sophisticated data analytics tools to analyse and measure impact, reach and engagement—and adjust tactics accordingly. Appoint a Chief Data Scientist to harness and employ data in the service of diplomacy.
  5. Develop and pilot a new stream of diplomatic reporting that’s punchier and timelier and reaches a broader audience on hand-held devices.
  6. Create new positions of ambassador to Silicon Valley (based in San Francisco) and ambassador to China’s tech giants (based in Beijing).
  7. Increase avenues to engage the Chinese public via Chinese social media platforms. This expansion should include dedicated Weibo accounts for the positions of Prime Minister and Foreign Minister.20
  8. Run a pilot of ‘embassy-lite’ or one-person posts. They’ll be more substantial and enduring than the ‘pop-up embassy’ in Estonia but still substantially lighter in footprint than a fully fledged diplomatic mission.
  9. Encourage innovation and experimentation in the conduct of digital diplomacy. Highlight and champion successes. Learn from (but don’t punish) the inevitable failures. Use DFAT’s Innovation XChange in this task, but broaden its focus beyond the aid program and extend its remit into mainstream diplomacy.
  10. Recognise that our overseas network is an underutilised asset. Find creative but non-traditional ways to use it to advance Australian national interests. Conceive of embassies as hubs and connectors for a broad set of interactions. Highlight and promote the strong performers (sending the cultural signal to others).
  11. Create a Twitter account for the Secretary of DFAT to internally signal the importance of digital diplomacy, to provide a further mouthpiece for Australian interests, and to give the public insight into the important work that Australia’s diplomatic service does every day.

What is ASPI?

The Australian Strategic Policy Institute was formed in 2001 as an independent, non‑partisan think tank. Its core aim is to provide the Australian Government with fresh ideas on Australia’s defence, security and strategic policy choices. ASPI is responsible for informing the public on a range of strategic issues, generating new thinking for government and harnessing strategic thinking internationally.

ASPI International Cyber Policy Centre

The ASPI International Cyber Policy Centre’s mission is to shape debate, policy and understanding on cyber issues, informed by original research and close consultation with government, business and civil society.

It seeks to improve debate, policy and understanding on cyber issues by:

  1. conducting applied, original empirical research
  2. linking government, business and civil society
  3. leading debates and influencing policy in Australia and the Asia–Pacific.

We thank all of those who contribute to the ICPC with their time, intellect and passion for the subject matter. The work of the ICPC would be impossible without the financial support of our various sponsors.

Important disclaimer

This publication is designed to provide accurate and authoritative information in relation to the subject matter covered. It is provided with the understanding that the publisher is not engaged in rendering any form of professional or other advice or services. No person should rely on the contents of this publication without first obtaining advice from a qualified professional person.

© The Australian Strategic Policy Institute Limited 2019

This publication is subject to copyright. Except as permitted under the Copyright Act 1968, no part of it may in any form or by any means (electronic, mechanical, microcopying, photocopying, recording or otherwise) be reproduced, stored in a retrieval system or transmitted without prior written permission. Enquiries should be addressed to the publishers. Notwithstanding the above, educational institutions (including schools, independent colleges, universities and TAFEs) are granted permission to make copies of copyrighted works strictly for educational purposes without explicit permission from ASPI and free of charge.

  1. Department of Foreign Affairs and Trade (DFAT), Social media, Australian Government, no date, online. ↩︎
  2. See, for instance, Fergus Hanson, ‘DFAT the dinosaur needs to find Facebook friends’, The Australian, 23 November 2010, online. ↩︎
  3. See twiplomacy, online, for rankings across a number of dimensions. ↩︎

Top US China specialist Peter Mattis announced as ASPI distinguished fellow

The Australian Strategic Policy Institute (ASPI) is pleased to announce Peter Mattis – Research Fellow in China Studies at the Victims of Communism Memorial Foundation – as a distinguished ASPI fellow for 2019.

Peter will be spending the next month at ASPI working with the Institute’s different China specialists on a range of research projects.

Peter will also be a keynote speaker at ASPI’s inaugural China masterclass being held on 15 April in Canberra (almost sold out) and 17 April in Melbourne (tickets still available).

Executive Director Peter Jennings says: “ASPI is delighted to attract someone of Peter’s analytical calibre to spend a full month at our institute. Peter has made an enormous contribution to building the world’s knowledge of how the Chinese Communist Party, the People’s Liberation Army (PLA) and China’s intelligence systems operate – both at home and around the world. Peter’s work spans across a broad range of topics, from better understanding PLA activity in cyberspace, to Chinese party-state influence operations, espionage and military modernisation. ASPI is looking forward to hosting Peter so that he can continue some of this important work over the next month”

Peter was formerly a Fellow in the China Program at The Jamestown Foundation where he edited China Brief and was an international affairs analyst for the US Government. He received his M.A. in Security Studies from the Georgetown University School of Foreign Service and earned his B.A. in Political Science and Asian Studies from the University of Washington in Seattle. He also previously worked as a Research Associate at the National Bureau of Asian Research in its Strategic Asia and Northeast Asian Studies programs.

Peter is in Australia from now until 18 April. For media enquiries please contact reneejones@aspi.org.au / 0400 424 323

Agenda for change – 2019

In 2018, many commentators pronounced the rules-based global order to be out for the count. This presents serious challenges for a country such as Australia, which has been an active contributor and clear beneficiary of that order. The government that we elect in 2019’s federal election will be faced with difficult strategic policy choices unlike any we’ve confronted in the past 50 years.

This volume contains 30 short essays that cover a vast range of subjects, from the big geostrategic challenges of our times, through to defence strategy; border, cyber and human security; and key emergent technologies.

The essays provide busy policymakers with policy recommendations to navigate this new world, including proposals that ‘break the rules’ of traditional policy settings. Each of the essays is easily readable in one sitting—but their insightful and ambitious policy recommendations may take a little longer to digest.

Previous Agenda for change publications are also available here: 2016 and 2013.

Launch Event

Building a Safer Internet – Advocate, Validate, Educate

5 February is Safer Internet Day, a global initiative in some 140 countries to raise awareness of emerging online issues. At ASPI’s International Cyber Policy Centre we are pleased to support this initiative.

A safer internet is at the core of what we do. We engage with international and national media on unfolding incidents, events and developments. We regularly organise public events on pressing issues in the online environment that shape strategic policy direction. And we have become very active in the area of capacity-building and exercises: in the Asia-Pacific region and in Australia.

On the occasion of Safer Internet Day 2019, ASPI’s International Cyber Policy Centre is introducing a 3-year project that looks at improving the security of the Australian internet through the adoption of international security standards. These standards are technical means to ensure a secure exchange of information over the Internet. Adoption is voluntary and non-binding and is reliant on goodwill and incentives.

“The Internet Society (ISOC) promotes an open, globally-connected, secure and trustworthy Internet. The use of open standards developed by open processes such as that of the Internet Engineering Task Force (IETF), the premier Internet standards body, plays an important role in achieving this. We welcome initiatives such as these which take a multi-stakeholder approach and aim to strengthen everyday users’ ability to be safe and secure online”, said Rajnesh Singh, Chief, Asia-Pacific Regional Bureau at the Internet Society.

With support from auDA, the policy authority and self-regulatory body for the .au domain space, the International Cyber Policy Centre will develop a public test tool. This tool will validate websites, email accounts and connections against standards that are considered international good practice.

Cameron Boardman, CEO of auDA said: “We are really excited about this cooperation with ASPI. It allows us to build bridges between strategic policy makers, businesses and end-users, and our stakeholders – domain name registrars and operators in the IT industry”.

This initiative draws on examples that International Cyber Policy Centre experts have observed elsewhere and from the Centre’s membership of the Global Forum on Cyber Expertise.

“This initiative by ASPI is a great example of expertise and experience being shared across the global community. The GFCE Internet Infrastructure Initiative aims to share global good practices making best use of state-of-the-art global open internet standards, with support from the Internet.nl portal”, says Maarten Botterman, GFCE project leader.

A safe internet is a community-wide and multi-stakeholder effort. Whether its government, industry, the tech community, civil society or the user-consumer, we all have our own rights and duties. This project will enhance opportunities for proper due diligence by users and consumers, small businesses as well as internet service providers.

Stakeholder consultations meetings are scheduled for February and March 2019; as follows:

MELBOURNE CONSULTATION
Date: Friday 22nd February 2019
Time: 11:00 hrs – 13:00 hrs
Venue: Joint Cyber Security Centre
Level 32, 600 Bourke Street, Melbourne

CANBERRA CONSULTATION
Date: Friday March 1st 2019
Time: 11.00hrs – 13.00hrs
Venue: ASPI Offices
40 Macquarie St, Barton
ACT 2600

SYDNEY CONSULTATION
Date: Monday 4th March 2019
Time: 10:00 hrs – 12:30 hrs
Venue: Joint Cyber Security Centre
Level 25, Tower 2, Darling Park
201 Sussex Street, Sydney

If you would like to participate in building a safer internet, please contact ASPI here.

Australia’s cybersecurity future(s)

It’s January 2024. Does Australia still have the internet?

Introduction

Australia wants to create a future for cyberspace that’s open, free and secure, but that future is not assured. According to Dr Tobias Feakin, the Ambassador for Cyber Affairs, ‘Australia’s vision … and our ambitions across the broad spectrum of cyber affairs are impossible to achieve alone.’1 Key drivers are outside of the country’s control. The government can—and should—advance a positive vision, but Australia might not get its way.

What if the future of cybersecurity looks different from what we hope or expect? This is a hard question to answer. Day-to-day concerns demand our immediate attention, and, when we think about the future, we tend to extrapolate from current trends. As a result, we’re shocked or surprised by discontinuous change, and woefully unprepared to face new realities. The risk is particularly acute in cybersecurity, in which rapidly changing technologies combine with diverse social and political forces to create unexpected consequences. Therefore, as difficult as it is to rethink our assumptions about the future, failing to do so could be dangerous.

This report uses scenario analysis to examine one such future: a world where cyberspace is fragmented in the year 2024. Contrary to the ambition of Australia’s International Cyber Engagement Strategy, cyberspace is neither open nor free in this scenario. We analyse what that implies for cybersecurity. In particular, we examine the challenges and opportunities that Australian policymakers may face in the future and wish they had planned for in our present.

We conclude that Australia will be caught in the fray if the internet breaks apart. While this scenario isn’t all bad, Australia could be forced to fend for itself in an increasingly dangerous neighbourhood. The scenario isn’t a forecast or prediction. It’s a compelling narrative to provoke new thinking and critical discussion about what Australia must do now to prepare for different cybersecurity futures.

Our approach is as follows. First, we explain the methodology. Second, we identify the forces of change that drive this scenario. Third, we interact these drivers to describe one possible world in 2024. Finally, we highlight the strategic choices and challenges that this scenario raises for Australia.

Scenario analysis

Scenario analysis is a methodology for critical thinking about alternative futures. It was pioneered at RAND in the 1950s by Herman Kahn in his attempt to ‘think the unthinkable’ about thermonuclear war. The method was further developed by Pierre Wack and Ted Newland at Royal Dutch Shell, where scenario analysis was credited with anticipating the possibility of oil shocks during the 1970s.2 It’s now commonly used in industry and government. For instance, scenario analysis informs the US National Intelligence Council’s quadrennial Global trends report.3 It’s also applied by the Center for Long-Term Cybersecurity at the University of California, Berkeley, in reports on Cybersecurity futures 2020 and Asian cybersecurity futures.4

The goal of scenario analysis is to ask and, ideally, answer ‘what if’ questions about how different drivers of change—social, political, economic, technological—could combine to produce discontinuities and thus different possible worlds. This approach is forward looking. We apply it to imagine Australia’s cybersecurity environment circa 2024. It may be unsettling. Following best practice, we sought to simplify and then exaggerate the drivers of change in order to throw an alternative and perhaps undesirable future into sharp relief. Nevertheless, scenario analysis is still rooted in reality.

The propositions behind this qualitative analysis are plausible, the narrative is internally consistent, and the results reflect expert consultation.

This report breaks from the norm of scenario analysis by focusing on one of many possible futures.

Our focus is not predictive, however. We do not argue that internet fragmentation is probable or likely to play out as per this scenario. We do suggest that this kind of future is significant because it challenges Australia’s preferred vision for an open, free and secure cyberspace. Fragmentation is also a significant concern in internet policy.5 Furthermore, while it may be a single scenario, a fragmented world contains different environments or ecosystems, and analysing that diversity helps compensate for our focus on only one potential future. The challenges and opportunities of such a future therefore warrant special consideration (just as other scenarios warrant further research). Rather than fight the scenario, we encourage you to ask: What would Australia need to decide and do differently for cybersecurity if it confronts this world in 2024?

Drivers of change

Our scenario depicts the interplay or interaction effects of three hypothetical drivers for change: Asia online, tech giants, and great-power conflict. While none is certain, each premise is plausible. More importantly, the resulting scenario is not a linear extrapolation or forecast based on any single trend. It’s the combination of drivers that could contribute to internet fragmentation and result in a cybersecurity environment markedly different from today’s.

Asia online

First, the number of users, devices and applications in Asia grows substantially over the next five years. We imagine that internet penetration in the region grows faster than expected, jumping from less than 50% today to more than 80%, so that more than 3.5 billion people are online in Asia. As a result, there are as many people online in this region come 2024 as the total number of internet users around the world in 2019. By 2024, Asia is also home to more than 15 billion connected devices.

We assume that this rapid expansion of connectivity is unrivalled in other regions. It roughly correlates to Asia’s youthful and growing population, as well as its economic power as the new centre of the global economy. However, economic and political opportunities remain unevenly distributed over the next five years, as is the region’s digital transformation. Most web traffic in Asia is mobile, but connection speeds vary greatly across the urban–rural divide, and economic growth hasn’t reduced economic inequality.

Tech giants

Second, we posit large and locked-in technology platforms as another driver for change. Although new applications flourish over the next five years, we assume that the underlying technology stacks, layers or platforms upon which those applications are built resemble a few large tectonic plates. And those platforms are increasingly dominated by a handful of huge corporations.

Tech giants dominate the user experience, software development and hardware. For most people in 2024, ‘cyberspace’ is difficult to distinguish from megabrands such as Google, Apple, Facebook, Amazon and Microsoft, or, similarly, Alibaba, Tencent, Baidu, Sina Weibo and Huawei. These companies also dominate the marketplace for talent. Regardless of where they work, most software developers work with toolkits and application program interfaces that plug into a dominant platform. Proprietary software developed by tech giants enjoys a home-field advantage over apps built by third-party providers. Industry concentration shapes hardware and telecommunications infrastructure as well, including the ‘internet of things’ (IoT). On the one hand, we imagine that connected devices are ubiquitous and produced by a plethora of manufacturers in 2024. On the other hand, in many markets, many of these connections are mediated by platforms, hubs and bridges dominated by the ‘Big 10’ tech giants.

Great-power conflict

The third driver is strategic competition and conflict between great powers. We posit a multipolar world in 2024. No great-power concert has emerged to manage territorial conflicts or the myriad state and non-state cyber operations. The US remains the only superpower with global reach, but that reach is rivalled by China’s, especially in the Pacific and Indian oceans. US power projection into the region is further limited by budget constraints (accentuated by an ongoing recession), as well as costly commitments to fighting in the Middle East and deterring a weak but assertive Russia. While NATO endures, nationalism and populism have fuelled extreme swings in American and European politics, fraying the alliance. ANZUS endures as well, but the US lacks a coherent strategy towards Asia in 2024. As a result, the US military posture isn’t supported by consistent political and economic policies.

Meanwhile, China has continued to rise. The Middle Kingdom is a middle-income country in 2024, with a nearly $15 trillion economy. Its One Belt, One Road and Digital Silk Road initiatives have established Chinese infrastructure, standards and platforms in several neighbouring economies. However, this economic and strategic agenda is resisted by India in the south and Russia in the north, along with European and American interests in Africa and Oceania. We posit that the Chinese economy has not dipped into recession, although its officially reported growth rate of 3% in the last quarter of 2023 is viewed with considerable scepticism. In China, as elsewhere, economic angst and nationalism have increased variability in foreign policy and contributed to competition and conflict in the region.

2024: Fragmented world, fragmented internet

In this scenario, Asia comes online but cyberspace fragments by 2024. Years of mounting tensions between the US, China, Russia and Western Europe have combined with entrenched platform technologies to result in a world where the internet—singular—is a thing of the past. The ‘World Wide Web’ is anachronistic. Instead, there are several weakly connected internets, each of which contains content and services that are largely inaccessible from outside the same country, region or bloc. There are tunnels through these walled gardens, but few users beyond specialists, spies and criminals have the skill or inclination to use them. Most users’ online access and experience is mediated and monitored by whichever tech giants enjoy official sanction in their local market. In most places, ‘social media’ are just media, and the IoT is just things.

The world’s largest internets are American and Chinese. Access to each correlates with physical proximity to the US or China, coupled with the broader user base of their respective tech giants. In particular, the American internet is accessible in most of the Western Hemisphere (corresponding to the American and Latin American regional internet registries). It’s also accessible in Western Europe, but tensions across the Atlantic have combined with divergent data protection and antitrust regulations, fuelling the emergence of a continental internet in the remnants of the European Union. Russia’s national internet is effectively cordoned off by internal information controls (heightened following the death of Vladimir Putin), combined with external blocking of untrusted traffic (Russian IP addresses being equated with criminal or intelligence operations and rejected by most border routers). National networks have also emerged in North Korea, Saudi Arabia and Venezuela. In addition to indigenous applications, the governments that regulate these and similar shards of cyberspace typically contract with Chinese or American firms to build platforms that are closed and customised for local censorship and surveillance.

Figure 1: Internets of the region, 2024

Enter the dragon

Like the Belt and Road Initiative, or the Nine-Dash Line, geography is a notable feature of the Chinese internet in 2024, which is portrayed as several concentric circles. Domestic services and content sit at the centre, behind the Great Firewall. China’s ‘Social Credit’ system hasn’t proved particularly effective in regulating behaviour offline; a goth-like fashion trend dubbed ‘false negative’ has even emerged to frustrate facial recognition. Nevertheless, China has become a nearly cashless society, and both big data and artificial intelligence are used to effectively monitor most online activity. The incidence of malware has decreased dramatically, and domestic cyber incident response is well coordinated.

Some cybersecurity experts worry that foreign intelligence services are exploiting the backdoor access required by China’s regulation of commercial encryption, yet the government denies any such allegation.

Outside the Great Firewall, similar services and content are available to those individuals, organisations and countries that use the platforms provided by China’s tech giants (or their local affiliates). Many do, particularly in Asia. By default, users in this second ring give their data to Chinese service providers.

Most of that information is stored on servers inside China. The outermost ring consists of custom networks that China has built but for which—purportedly—it has handed information controls over to the client, such as for the heavily restricted mobile apps recently launched in North Korea.

The Western Front

For many users in the US, the American internet in 2024 appears similar to the World Wide Web in 2019. A similar set of tech giants from Silicon Valley and Seattle dominate the market. Their proprietary platforms seem to seamlessly integrate users’ digital lives. Toddlers are frequently reported to perceive voices such as Google Home and Amazon Echo as disembodied members of their families. Data breaches of personally identifiable information are so common as to rarely make news; occasionally, car fleets and wired housing developments that have been bricked by cyberattacks make headlines. Net neutrality remains contentious and partisan. Demands from law enforcement for data collected by bystanders’ wearable tech during the Denver bombing in 2022 have ignited another round of debate over encryption (a debate joined by lobbyists for fintech and cryptocurrencies).

Lobbying by tech giants, fractious domestic politics and anti-statist ideology limit US federal regulations on cybersecurity. One exception is wireless broadband. A government-sponsored, industry-led consortium has rolled out a mobile network called US5G. Chinese companies are banned from building this infrastructure. Likewise, Chinese and Russian cybersecurity software is banned from use on US Government computers. The Security and Exchange Commission has also imposed reporting requirements on cryptocurrencies and initial coin offerings. Domestic information sharing has improved modestly after years of concerted attacks against critical infrastructure, but individual users still have little recourse, and the quality of cyber insurance is variable. US diplomats pay lip service to ideas such as ‘internet freedom’ and ‘cyber norms’ when they criticise authoritarian regimes, but the promotion and practice of the American internet abroad is largely determined by the commercial strategies of its tech giants.

Figure 2: The US5G logo

Fault lines

Asia is a contested zone in 2024. The US and China vie for power in the region while Chinese and American firms compete for market share. Unfortunately, the US and China appear caught in the ‘Thucydides trap’, as the rising and ruling powers jostle near the brink of armed conflict.6 War was narrowly averted in 2022 following a naval skirmish in the South China Sea that killed 65 sailors and marines aboard American and Chinese warships. Patriotic hacking—both state-sanctioned and self-radicalised—during this incident was intense and occasionally destructive. Since then, submarines have been reported patrolling undersea cables in the Pacific. In addition, real and imagined instances of Chinese and American firms facilitating offensive cyber operations by military and intelligence agencies have driven yet another wedge between their rival internets.

On the one hand, countries in the Indo-Pacific enjoy more choice than those in the Western Hemisphere, since the American and Chinese internets are both viable options in this region. Some countries are choosing to bandwagon with China. In 2024, Alibaba, Tencent, Baidu, Sina Weibo and Huawei are providing a bundle of telecommunication, media, IoT and financial services called WeConnect. This bundle has proved remarkably popular in Malaysia, for instance, and among the Chinese diaspora across Asia. WeConnect has also increased internet access in Myanmar and Cambodia by an order of magnitude: millions of their people have leapfrogged from having no phones to using Chinese smartphones overnight. In contrast, Japan uses the American internet as a matter of policy, and most users in Indonesia and the Philippines remain locked into Facebook and Google. India is non-aligned (despite the prevalence of American platforms), and Pakistan is hedging its bets (despite widespread adoption of WeConnect). Competition and choice between American and Chinese internets are fuelling digital innovation across the region.

On the other hand, innovation in this scenario is not improving global integration. Choosing one internet increasingly means forgoing access to others. Chinese and American cybersecurity standards are not compatible. Nor is compatibility of much interest to the tech giants. Years of national tariffs, investment restrictions, divergent regulations and export controls have limited their sales in the others’ domestic markets. Combined with the US5G network, these policies have forced American firms to shift away from Chinese suppliers. Similarly, the ‘Made in China 2025’ initiative has made Chinese tech giants more self-sufficient. The US–China skirmish in 2022 accelerated the disintegration of once highly integrated supply lines and manufacturing. When competing for customers in Asia, the tech giants are incentivised to collude within their own internet and exclude foreign rivals.

Moreover, the range of choice in this region comes at considerable cost. While some aspects of cybersecurity have improved inside Chinese and American internets, those improvements are lost in the mixing zones between them. Cheap, outdated and counterfeit technologies are most vulnerable, enabling cybercrime in 2024 to cost Asia as much as $3 trillion per year. Ransomware, DDoS by IoT botnets, cryptocurrency fraud, industrial espionage, election interference—all are common, especially at the local level. Diverse technology limits the spread or scale of most attacks, but it also provides criminals with many smaller targets of opportunity outside the Great Firewall. Jumbled laws across different jurisdictions also provide safe haven for state and non-state actors to launch attacks and hide ill-gotten gains. In this scenario, data protection isn’t imagined to be a top priority for hundreds of
millions of people who are coming online for the first time. Even more than the American internet, the Chinese internet in 2024 owes its success to users willing to forgo privacy in exchange for access and convenience. The appetite for adopting digital technologies in this contested environment is a recipe for legal and illegal innovation alike.

Moving forward: strategic choices and challenges for Australia

The world that we describe would have serious implications for Australian cybersecurity. At least three lessons stand out in our analysis.

Australia will be caught in the fray

In this scenario, China remains the primary pillar of the Australian economy and the US remains Australia’s security guarantor. Australia won’t want to take sides, and with good reason. But the digital economy may prove more sensitive to geopolitical tension than other markets, in which case Australia could face tough choices in cyberspace sooner rather than later.

The costs of choosing either an American or a Chinese internet could be significant, though not equal. Not choosing could be costly as well. While a mediating, brokering or hedging strategy may prove the lesser evil, it may also make Australia the target of intense pressure. Domestic affairs could become a microcosm of fierce regional competition. Potential outcomes include foreign surveillance, censorship and the manipulation of Australian markets, networks and politics. Chinese platforms are particularly suspect, but American technologies aren’t above reproach. How will federal, state and local governments respond in March 2024, for example, if mass student protests in Melbourne are manipulated through WeConnect? How much more difficult will whole-of-government policies and operations be, even at the federal level, if the tensions between cybersecurity and economics become increasingly pronounced?

29 November 2023

Australian Fintech Firm Shuttered:
US Alleges Data Manipulated by China

The Sydney-based cryptocurrency exchange TransPacific Ledger (TPL) was forced to shut down last night, less than a day after the discovery of data irregularities in trading worth more than $1.5 billion.

TPL suspended operations after the firm was implicated in the crash of blockchain backed indexes in the United States. Trading data brokered by TPL may have been manipulated in high-speed transactions between the US and China.

A darling of the Sydney start-up scene, TPL had been seen as a trusted and profitable intermediary between American and Chinese financial markets. ‘We have a sales office in Hong Kong, we’re fully licensed in Australia, and we comply with all US regulations,’ said TransPacific CEO Ed Jones in an interview last month.

However, US cryptocurrency exchanges crashed on Monday when irreconcilable discrepancies were reported across several ledgers. ‘TPL appears to be the common link,’ according to the White House press secretary, ‘but China is behind the bad data.’ US intelligence officials point to recent advancements in Chinese quantum computing, claiming that these computers could hack the authentication protocols behind blockchain. ‘Maybe this was an experiment that got out of hand,’ said one anonymous source.

Beijing brusquely rejected these claims. ‘False accusations accomplish nothing,’ according to one government spokeswoman. Prominent voices in Chinese media are now blaming unnamed criminals in Australia and demanded their immediate extradition.

The Australian Securities and Investments Commission is working with the Australian Signals Directorate in its investigation. Neither agency was available for comment. The ASX lost 5% after news about TPL broke on Tuesday.

Please note: the above is a fictional article created by the authors for the purpose of this report.

By straddling both internets, both networks could be used to push and pull divisions in Australian government and society. Moreover, even if Australia tries to straddle the US and China, other countries in Oceania may decide differently. For instance, how will Canberra respond if Papua New Guinea, Bougainville and Solomon Islands bargain to adopt the Chinese internet in 2024 unless Australia increases development assistance to expand and maintain their undersea cables? In this scenario, Australia will have to decide how much it’s willing to pay for its preferred strategy, both at home and around the neighbourhood.

Internet fragmentation isn’t all bad everywhere

As costly as straddling or choosing between American and Chinese internets would be for Australia, this isn’t a doomsday scenario. Some aspects of cybersecurity stand to improve inside each network. Harmonised standards and coordination across like-minded jurisdictions could improve incident response, information sharing (including vulnerability disclosure), patching and attribution. Technological diversity may increase at the regional and global levels, limiting the scale of any given platform and thus the extent to which attacks spread beyond any given country, region or bloc. Trust inside these networks may improve as well. For example, this scenario imagines that the average American in 2024 is relatively confident about US5G (despite expert debate about whether this network is demonstrably more secure than the Chinese alternative). Real or imagined, these security gains may make joining one club or another an attractive prospect for Australia.

Granted, the security gains inside each network are offset by friction between them. Australian policymakers will also bristle at claims by China, Russia and other authoritarian regimes that strict censorship and surveillance improve the security of their respective internets. Nevertheless, fragmentation or disintegration need be neither chaotic nor absolute. For better or worse, cross-fertilisation and ideological hypocrisy will occur as well, with American companies mirroring some of the practices used by their Chinese counterparts and vice versa.

Thursday, January 4, 2024

Mastercard and Walmart introduce a Social Credit System

Dismissing comparison to China, Walmart claims new system will help its consumers “live better” and “save money” during the US recession.

Please note: the above is a fictional article created by the authors for the purpose of this report.

Australia lives in a dangerous neighbourhood

The concurrent great-power transition and digital transformation of the region could be more turbulent than in any period in recent history. Tech giants will shape this transformation, but their commercial interests diverge from the public interest in Australian cybersecurity. In contrast to powerful corporations, international organisations such as the International Telecommunication Union appear even less impactful than usual in this scenario. Even multi-stakeholder organisations such as ICANN could be coopted or captured by commercial and geopolitical interests.

Tough Choices

Australia isn’t helpless in this environment, but it should prepare to help itself. Looking back, policymakers in 2024 may wish that preparation had started in 2019. Options include redoubling Australian efforts to champion an open, free and secure cyberspace in order to avoid the future imagined here. Advancing regional leadership, investing in capacity building and taking assertive action on shared interests may prove helpful. At the same time, however, policymakers should consider tough choices about cybersecurity in a less benign environment: 

  • Is Australia prepared to play hardball, not only with the US and China, but also with commercial tech giants, in order to advance its national interest?
  • If forced to take sides or straddle the great powers, how should Australia choose, and how can it mitigate the costs of doing so?
  • Even if there’s no defining moment (for example, President Trump or President Xi declaring ‘You’re either with us, or against us’), is muddling through on issues such as encryption in Australia’s national interest, especially if incremental decisions aggregate into a decisive choice?
  • What, if anything, can Australia do to help the next billion users in Asia come online in ways that improve rather than undermine critical aspects of cybersecurity?
  • And will a laissez-faire or, alternatively, compliance-driven approach to domestic cybersecurity suffice or prove lamentable in the years ahead?

These are important questions to answer, regardless of whether or not the scenario that we describe comes to pass. Scenario analysis doesn’t need to provide accurate predictions in order to provoke strategic thinking about the future of Australian cybersecurity.

Acknowledgements

This report was produced in collaboration between the Sydney Cyber Security Network and ASPI’s International Cyber Policy Centre. It was made possible thanks to a research grant provided by the Sydney Policy Lab. We also thank our research assistant Bryce Pereira, as well as the other experts and visionaries who provided helpful comments and feedback.

@SydneyCyber – https://sydney.edu.au/arts/our-research/centres-institutes-and-groups/sydney-cybersecurity-network.html

ASPI International Cyber Policy Centre

The ASPI International Cyber Policy Centre’s mission is to shape debate, policy and understanding on cyber issues, informed by original research and close consultation with government, business and civil society. It seeks to improve debate, policy and understanding on cyber issues by:

  1. conducting applied, original empirical research
  2. linking government, business and civil society
  3. leading debates and influencing policy in Australia and the Asia–Pacific.

We thank all of those who contribute to the ICPC with their time, intellect and passion for the subject matter. The work of the ICPC would be impossible without the financial support of our various sponsors.

Important disclaimer

This publication is designed to provide accurate and authoritative information in relation to the subject matter covered. It is provided with the understanding that the publisher is not engaged in rendering any form of professional or other advice or services. No person should rely on the contents of this publication without first obtaining advice from a qualified professional person.

© The Australian Strategic Policy Institute Limited 2018

This publication is subject to copyright. Except as permitted under the Copyright Act 1968, no part of it may in any form or by any means (electronic, mechanical, microcopying, photocopying, recording or otherwise) be reproduced, stored in a retrieval system or transmitted without prior written permission. Enquiries should be addressed to the publishers. Notwithstanding the above, educational institutions (including schools, independent colleges, universities and TAFEs) are granted permission to make copies of copyrighted works strictly for educational purposes without explicit permission from ASPI and free of charge.

  1. Department of Foreign Affairs and Trade, Australia’s International Cyber Engagement Strategy, Australian Government, October 2017, 7. ↩︎
  2. For background, see Pierre Wack, ‘Scenarios: Shooting the Rapids – How Medium-Term Analysis Illuminated the Power of Scenarios for Shell Management,’ Harvard Business Review (1985), 139-150; Peter Schwartz, The Art of the Long View: Planning for the Future in an Uncertain World, Doubleday, New Your 1991; Naazneen H. Barma, Brent Durbin, Eric Lorber, and Rachel E. Whitlark, ‘“Imagine a World in Which”: Using Scenarios in Political Science’, International Studies Perspectives 17 (2016), 117-135. ↩︎
  3. For example, see National Intelligence Council, Global trends: paradox of progress, January 2017 ↩︎
  4. Center for Long-Term Cybersecurity, Cybersecurity futures 2020, online; Jonathan Reiber, Arun M Sukumar, Asian cybersecurity futures: opportunities and risk in the rising digital world, Center for Long-term Cybersecurity ↩︎
  5. Among others, see William J Drake, Vinton G Cerf, Wolfgang Kleinwachter, Internet fragmentation: an overview, Future of the Internet Initiative White Paper, World Economic Forum, January 2016, online; Scott Malcomson, Splinternet: how geopolitics and commerce are fragmenting the World Wide Web, OR Books, New York, 2016; Davey Alba, ‘The world may be heading for a fragmented “splinternet”’, WIRED, 7 June 2017 ↩︎
  6. Graham Allison, ‘The Thucydides trap: are the US and China headed for war?’, The Atlantic, 24 September 2015 ↩︎

Online Influence and Hostile Narratives in Eastern Asia – Report

ASPI’s International Cyber Policy Centre wrote a report for the NATO Strategic Communications Centre of Excellence that examined online influence and hostile narratives in Asia.

Eastern Asia — which we define as including East and Southeast Asia — is a region of increasing geopolitical competition with many racial, cultural and societal fractures. With the rapid expansion of inexpensive internet access, these fractures and tensions mean that many states in the region are both vulnerable to, and a source of, hostile information activities that are being used to achieve strategic goals both inside and outside the region.

This report documents examples of hostile information activities that have originated in Eastern Asia and have been targeted in the following countries:

  • Taiwan
  • The Hong Kong-based protest movement
  • West Papua
  • The Philippines

Because these activities often target social media, they have been difficult for law enforcement and national security organizations to police. Across the globe, countries are pursuing different methods of tackling the spread of hostile information activities with differing degrees of success. These approaches can range from law enforcement, temporary internet shutdowns, and attempts to legislate against ‘fake news’ or disinformation, through to wider societal media literacy initiatives.

Read this report, authored by ASPI International Cyber Policy Centre researcher Hannah Smith, here.

Introducing integrated E-Government in Australia

Foreword

With the 2016 distributed denial of service attack on Australia’s first fully digital Census and Centrelink’s 2017 automated debt-recovery system glitches still fresh in our minds, it would be easy to pause in the pursuit of digitising government services.

The reality, however, is that there are compelling benefits to expediting government digital transformation, and the case for change is not simply one of customer convenience.

Deloitte Access Economics has estimated that the federal and state governments conduct 811 million citizen transactions each year. It calculated that lifting the share of transactions performed digitally from 60% to 80% over a 10-year period would lead to government productivity benefits worth $17.9 billion, plus a further $8.7 billion in benefits to citizens. 

But the benefits of integrated digital government services extend even beyond time and resources saved. Data is the fuel for many new business models and, according to OECD measures, right now Australia performs only moderately well compared to international peers, particularly in relation to the availability of open government data.

The OECD has estimated that adopting more data driven decision-making in government has potential output and productivity benefits of 5% to 6% in the US, while improving data quality and access by 10% could increase labour productivity by an average of 14%. That can have additional flow-on effects across the economy. Almost 2 million people are employed in the three levels of government in Australia, meaning that 16% of the country’s 12.5-million-strong workforce is employed in the public sector.

This represents a strategic capability, enabling knowledge and skills transfer across the broader economy. Based on previous productivity gains from technology take-up, that can have significant benefits for Australia’s output. Further adoption of digital technologies across the economy has the potential to add an extra $66 billion to Australia’s GDP over the next five years alone.

So the case for change is clear; the question is really about how to do it. How do we maximise the opportunities, while best protecting citizens’ data and privacy? This policy brief is intended to start that conversation.

Yohan Ramasundara
President, Australian Computer Society

What’s the problem?

Australia was an early leader in the digitalisation of government services, and some Australian Government departments and state governments have continued to innovate and deliver enhanced services online. However, in the global context, Australia has now fallen behind and has so far failed to adopt an integrated approach to e-government that joins up all government services across all three tiers of government. For citizens, this makes life harder than it needs to be and consumes time that could be spent on other things.

For businesses, it increases transaction costs. Although existing user interfaces are logical and user-friendly, there’s still a limited amount of third-stage e-services enabling two-way interactions between citizens and governmental institutions.1 Critical missing pieces inhibiting the flourishing of e-services are a properly functioning digital identity ecosystem and a digital signature.2

What’s the solution?

The Australian Government should launch a consultation with the states and local governments to develop an integrated approach to e-government that joins up all services from all three tiers of government. The model will need to be customised to Australia’s unique circumstances but should be designed to reduce business transaction costs, allow citizens to engage seamlessly with the federal, state and local governments and prioritise citizens’ control and ownership of their data.

A decentralised architecture should be used to ensure there’s no single point of failure and to allow easy and secure integration with existing digital government platforms. The federal government should provide essential enabling systems: 

  • a digital identity (eID)—one has already been developed by Australia Post, and a second is being built, but significant work is needed to allow eID to take root
  • the legal, organisational and technical preconditions for a digital signature—legislation should ensure that the digital signature has equal legal weight to a traditional handwritten signature
  • secure data exchanges between different government IT systems.

Introduction

Integrated Australian e-government would mean that less of citizens’ and businesses’ time would be wasted engaging with government. A digital signature would make official transactions simple: signing contracts or submitting applications could be done in moments. Mindless hassles when moving between jurisdictions (such as swapping licences from one state to another) would evaporate overnight; there would be no need to conduct 100-point identity checks in person, and time-consuming visits to physical government offices would become a thing of the past. In Estonia, where e-government is a national passion, officials estimate that these efficiencies lift annual GDP by 2%.3

While many government departments already have user-friendly online portals, and some states have begun integrating several services within single online platforms (such as Service NSW and Service Victoria4), Australia has yet to attempt a citizen-centric approach that makes citizen and business engagement with all three tiers of government seamless. It also lacks critical enabling systems. The major building blocks needed to achieve an integrated approach to e-government are an integrated government back office and a simple, easy-to-use and secure eID and digital signature. 

That isn’t to downplay the practical challenges of joining up three tiers of government that have historically resisted cooperation or the attention to detail needed to address cybersecurity challenges. Joined-up e-government is nonetheless essential to a high-functioning 21st-century economy and should be attempted.

E-government in Australia

Australia was initially quick to join the global e-government trend, and even developed an international reputation as an early leader in this area (peaking around 1999).5 However, a joined-up approach to e-government wasn’t achieved.6 The success of some large departments, such as the Australian Taxation Office and Centrelink, has depended more on a joined-up ‘front end’ rather than an integrated back end that allows citizens to engage with government seamlessly.7

A national identification scheme (the Australia Card) was proposed in the 1980s. However, the Australia Card Bill generated significant public concerns about privacy and was defeated in the Senate.8 In 2006, Prime Minister John Howard made another attempt with the Access Card,9 before it too was shut down by the Rudd government in 2007.

The Electronic Transactions Act 1999 meant that when entities were required under federal law to give information in writing, provide a signature or produce a document, they could do it electronically.

However, the Australian Government and state and territory governments exempted a large volume of legislation from the operation of the Act. While the Act was an enabler, it didn’t create a ‘unique and un-forgeable identifier that can be checked by the receiver to verify authenticity and integrity and provide for non-repudiation’.10

At the end of the 1990s, the Department of Communications, Information Technology and the Arts was a central player in the coordination of e-government. Two units were created within the department: the Office for Government Online and the National Office for the Information Economy (NOIE), which provided advice and support to the government on internet-specific matters.11 Some of the functions of the NOIE were subsequently taken over by the Australian Government Information Management Office, which was established in April 2004.

However, government departments and agencies had variable reputations, and innovative cross-government projects usually originated from the biggest departments.12 To an extent, that’s still the case, but with more coordination. In general, the major electronic players (such as the Tax Office and Centrelink) and innovative state governments were leading the field, advising central agencies and driving central initiatives.13

In 2016, the federal government established a new agency to manage the government’s digital and ICT agendas: the Digital Transformation Agency (the successor to the Digital Transformation Office, launched in 2015). The agency aims to integrate digital delivery across the federal government and also enhance the transparency of the government’s ICT and digital projects. It covers strategic and policy leadership on whole-of-government and shared ICT and digital service delivery, including ICT procurement policy.14 The Digital Transformation Agenda, coordinated by the agency, foresees agencies and departments delivering ‘a range of initiatives that will provide benefits to all users and improve their digital experience’, including Single Touch Payroll; My Health Record; health payments; trusted digital authentication and verification; whole-of-government platforms; grants administration; and a streamlined online business registration service.15

The Trusted Digital Identity Framework outlines a consistent approach to digital identity in Australia and will be an important component of any integrated approach to e-government.16 Some $92.4 million in funding was secured in the 2018–19 federal budget17 to create the infrastructure that will underpin an eID (Govpass), and the government is aiming to roll out pilot services to half a million users by the end of June 2019.18 This will largely duplicate an eID recently launched by Australia Post called Digital iD. The challenges to the widespread rollout and adoption of eID in Australia are dealt with in a previous Policy Brief.19

States and local councils also deliver a range of services online. A leading actor is the New South Wales Government, which offers a single sign-on service for secure access to government transactions; more than 1.5 million customers have already signed up.20 Victoria is another leader. In May 2016, it released the Victorian Government Information Technology Strategy, which outlines steps the government is taking to improve the security of information and infrastructure critical to the proper functioning of e-government.

At the local government level, the City of Sydney is contributing to the open data movement by making accessible to the public an ever-growing range of data in a number of formats. The datasets provide information on environmental sustainability, transport, arts and culture, facilities, parks and more.21 Opening up data facilitates the creation and management of open services for the private and community sectors, increases transparency and stimulates the economy. It also decreases the number of information requests and reduces administrative workload.

An integrated approach to e-government in Australia

An integrated approach to e-government in Australia would require detailed consultations across all three tiers of government, and with business and the public. However, several principles derived from the experience of others can help frame the approach. 

The once-only principle

The once-only principle (OOP) is central to joined-up government. The EU addressed this in its eGovernment Action Plan 2016–2020, where the foundations for the EU Digital OOP are laid out.22 The OOP requires that individuals and businesses shouldn’t have to supply the same information more than once to public entities (for example, when notifying a change of address). This requires the existence of public-sector interoperability at different levels: organisational, legal and technical. The conceptual model of the new European Interoperability Framework foresees interoperability levels as integral parts of integrated public service governance, meaning that different public administrations work together to meet citizens’ needs and provide public services in a seamless way.23

A decentralised approach

Facilitating secure data exchanges and interoperability between different government agencies doesn’t require the creation of a single database (a so-called superdatabase) that consolidates all data from other databases. In fact, doing that poses serious security risks. A decentralised approach enables different databases and IT solutions in the three tiers of government to ‘talk’ to each other securely and solves the problem of how to integrate the myriad different government databases and systems that already exist. Four key elements underpin this secure exchange:

  • the identification of both the sender and the receiver of the data
  • the encryption of data exchanged to ensure the data is unreadable in case someone intercepts it
  • the time stamping of data transactions
  • a legal audit trail via archiving and logging of electronic records.

In Estonia, X-Road (Figure 1) is a distributed information exchange platform that makes it possible for different systems to communicate across the entire governmental sector.24

Figure 1: Estonia’s X-Road

A digital identity

Digital identity is central to e-government. It serves two main functions: proving one’s identity in the virtual space and verifying virtual transactions. Given the administrative division of Australia into six states and two territories, specific cross-border solutions promise added efficiencies. The EU has taken steps in the direction of cross-border electronic identification and trust services. Its eIDAS Regulation (no. 910/2014) ensures that people and businesses are able to use their own national eID schemes to access public services in other EU countries where such schemes are available. It also ensures the legal validity of digital interactions; that is, they have the same legal status as traditional paper-based transactions. The EU case highlights the need to provide a predictable regulatory environment to enable secure and seamless electronic interactions between businesses, citizens and public authorities. With Australia Post’s Digital iD and Govpass, Australia is laying the foundation for a national eID, although some major questions remain to be addressed.

Privacy

Addressing privacy concerns through a citizen-driven e-government model is important in winning public support for integrated e-government, especially given the history of the failed Australia Card and scandals such as eCensus. Mutual trust is the key to interactions in which the government collects information about citizens and citizens provide their own data to the government. The principles of confidentiality, integrity and accessibility of data are all critical. Building trust between citizens and authorities is at the core of a working e-government model, so considerable emphasis should be put on communicating with citizens about how and for what reason their data will be processed by the government.

One lesson learned from abroad is the value of placing citizens in the driving seat. In Estonia, for example, every time a citizen’s personal data is accessed by a government agency, the individual user can see that access via a log and contest it if they believe it to be improper. Another example from Estonia is related to the right to choose whether to use digital identity or not. Those who do not want to use their digital identity can still use a physical service centre. Australia is also planning an opt-in approach to its new digital identity; however, it may become de facto compulsory if private-sector organisations are able to insist as a condition of service that it’s used (for example, to use online banking). Were that to eventuate, it would raise concerns about anonymity and the ability to not share information.

A joined-up back office

In order to provide easily accessible e-government services across all tiers of government, a joined-up back office is central. So far, the success of some major agencies, such as the Tax Office and Centrelink, depends more on a joined-up ‘front-end’ (the interface between the user and the back office). As Catherine Garner has noted: ‘Improving Australia’s cross-agency collaboration and integration will provide efficient, dynamic systems with greater personalisation and support Australia on its journey to become an e-government leader’.25

Evaluating outcomes from government-funded services

The ability to evaluate outcomes of publicly funded services is an important means of measuring the effectiveness of the government services being provided to citizens. Applying strict privacy and information security practices, there would be value in evaluating outcomes from government spending at the population level, rather than on a simple agency-by-agency basis. There would be community benefits in having the secure, de-identified evidence base made available for approved service improvement and evaluation of government-funded programs and policies.

Other issues

In addition to these guiding principles, Australia will need to resolve a number of other important issues. In summary, they include the need to:

  • ensure secure data exchange and security of data
  • manage the integration process and metadata related to systems and services (a clearly defined and regulated approval process, for example via the Office of the Australian Information Commissioner, is needed for adding new components or new services to ensure smooth integration and the maintenance of security and privacy standards)
  • ensure the right of all citizens using e-government services to easily access information about how government is using their data
  • ensure the right of citizens to decide who can access their data
  • ensure the right of citizens to decide whether or not to use their eID.

Lessons learned from abroad

To implement integrated e-government in Australia, work is needed at several organisational, legislative and technical levels. A few conceptual questions were important when Estonia was developing integrated e-government:

  • The question of how to identify people, businesses and real estate had to be addressed. In order to enable trustable and secure data exchanges between different databases and information systems, some identifiers for people, businesses and cadastral units are needed. In Estonia, ID numbers of people and businesses and also cadastral numbers are regulated by law and implemented in all databases and information systems. This is the precondition for secure and trustable data exchanges between different systems.
  • The digital ID and digital signature are issued by the same process.26 Private keys (for use by the public key infrastructure) are generated by crypto-processor (chip) and aren’t downloadable.27 The eID and digital signature constitute a part of the government-issued and guaranteed infrastructure, which is used by both the private and the public sectors.
  • While an eID is obligatory if a citizen wants to use e-government services, the citizen isn’t obliged to use their digital identity (they can use non-eID-based systems if they prefer).
  • Finally, the citizen is the owner of their own data.28 They can control the use of the data managed by the government. The use of personal data is strictly regulated by law. Everyone can restrict the use of their data by blocking access to it if the law doesn’t specify otherwise.

Another lesson from Estonia concerns back-office integration. Several conceptual agreements underpinned the design of the country’s e-government architecture:

  • Decentralisation: The system is decentralised. There’s no single point of failure, and the central management of the system doesn’t ‘see’ the data, but only whether the system is working.
  • Ease of implementation: The system should be easy to implement. Government institutions shouldn’t need to change their existing systems and processes. Training on the integration of the systems should be offered to all technical experts working in e-government back offices.
  • Neutrality of technology platforms: The integration of systems doesn’t mean that all technical systems use the same platform. Usually, governments use a range of proprietary software platforms as well as open-source solutions and technologies developed by different vendors. Integrated e-government should accommodate those variances.29
  • Security of transactions: Integrity, confidentiality and non-repudiation (the assurance that a party to a contract or a communication can’t deny the authenticity of their signature on a document or the sending of a message that originated from them) should be guaranteed.30
  • Security of data and services: Data and services should be secured so they can be transferred via public networks. The use of the public internet should be enabled, and the development of separate (usually very expensive) government data networks should be avoided.
  • Agile planning and implementation: It’s necessary to avoid large, complex projects and instead develop a comprehensive general architecture that can be divided into small components, while still giving due consideration to security requirements.

Recommendations

We make the following recommendations for the further development of e-government in Australia.

  • Avoid large e-government projects. Agile development can minimise risks, enable faster results and avoid implementation challenges.
  • Establish a properly functioning secure eID and digital signature for each citizen. The eID should be simple and user-friendly, issued by government (similarly to passports) and guaranteed by law. It should be used for both e-government services and business e-services.
  • Back-office integration should be coordinated centrally but done in a decentralised way, enabling secure data exchange between systems connected via the internet. The integration platform should enable the integration of different technical platforms in different locations, in different legal environments and with different organisational set-ups. The integration platform should be as simple as possible and not require changes to existing back-office processes and systems. Process redesign can be done step by step.
  • A citizen-centric model is important to win public support for integrated e-government. It should allow people to control their private data and provide legal guarantees, supported by organisational and technical frameworks. Building trust takes time, so carefully planned communication between the government and citizens is critical, including building up and publicising a track record of competent and secure service delivery. This can be assisted by following basic design concepts and data protection principles when designing the eID and the back-office integration of IT systems.

Integrated e-government offers major benefits to businesses and citizens. It reduces the time and costs associated with transacting with government and with each other and makes life easier. A thoughtful approach to designing integrated e-government (such as decentralisation) will also mean that the risks of a data breach won’t be increased. Australia’s geography and population size don’t present any technical obstacles to rolling out a world-class e-government system.

The move to create digital identities in Australia also suggests growing political momentum to take a more holistic approach to e-government. If it’s citizen-centric, it could help win public support, too.

Important disclaimer

This publication is designed to provide accurate and authoritative information in relation to the subject matter covered. It is provided with the understanding that the publisher is not engaged in rendering any form of professional or other advice or services. No person should rely on the contents of this publication without first obtaining advice from a qualified professional person.


© The Australian Strategic Policy Institute Limited 2018
This publication is subject to copyright. Except as permitted under the Copyright Act 1968, no part of it may in any form or by any means (electronic, mechanical, microcopying, photocopying, recording or otherwise) be reproduced, stored in a retrieval system or transmitted without prior written permission. Enquiries should be addressed to the publishers.

  1. The online sophistication ranking assesses service delivery against a five-stage maturity model: information; one-way interaction; two-way interaction; transaction; and targeting/automation. The fourth and fifth stages can be referred to as ‘full online availability’. For more information, see Capgemini, IDC, Rand Europe, Sogeti, DTi, Digitizing public services in Europe: putting ambition into action, 9th benchmark measurement, report for the European Commission, December 2010 ↩︎
  2. The release of the South Australian Government’s digital driver’s licence is a useful case study, highlighting what’s possible, but also the critical missing piece for nationally consistent electronic identity and digital signatures, which inhibits the flourishing of e-services. See Department of Planning, Transport and Infrastructure, South Australian driver’s licences to go digital, South Australian Government, 22 September 2017 ↩︎
  3. Charlemagne, ‘Estonia is trying to convert the EU to its digital creed’, The Economist, 6 July 2017 ↩︎
  4. Along with the Australian Computer Society, both the NSW and Victorian governments contributed funding towards this research and the visit to Australia by Dr Arvo Ott. ↩︎
  5. P Chen, RK Gibson, W Lusoli, SJ Ward, ‘Australian governments and online communication’, in S Young (ed.), Australian government communication, Cambridge University Press, Cambridge, 2007. ↩︎
  6. The Australian Management Advisory Committee’s 2004 Connecting government report defined the concept of whole-of-government in the Australian Public Service as follows: ‘Whole-of-government denotes public services agencies working across portfolio boundaries to achieve a shared goal and an integrated government response to particular issues. Approaches can be formal or informal. They can focus on policy development, program management, and service delivery.’ ↩︎
  7. P Dunleavy, H Margetts, S Bastow, J Tinkler, ‘Australian e-government in comparative perspective’, Australian Journal of Political Science, 2008, 43(1):13–26 ↩︎
  8. G Greenleaf, ‘The Australia Card: towards a national surveillance system’, Law Society Journal, 1987, 25(9), online; R Clarke, ‘Just another piece of plastic for your wallet: the “Australia Card” scheme’, Prometheus, 1987, 5(1):29–45. ↩︎
  9. Office of the Access Card, How will the card benefit you?, Australian Government, no date. ↩︎
  10. Attorney-General’s Department, The Electronic Transactions Act 1999, information sheet, no date. ↩︎
  11. Also, in 1997 the new Liberal–National government launched a major central government outsourcing initiative in order to improve private-sector involvement in government. The aim was to outsource IT across the whole federal government. All departments and agencies were forced to outsource their IT operations to one of the largest international IT corporations with an Australian presence. In 2001, following critical reports from the Australian National Audit Office, the initiative was replaced by more conventional procurement methods. However, the same contractors continued to be important players, consolidating the IT market and leaving little expertise within the government, except for the largest departments. See Dunleavy et al., ‘Australian e-government in comparative perspective’. ↩︎
  12. For instance, the Australian Taxation Office enables individual taxpayers and their agents to use the ‘e-Tax’ electronic tax return lodgement facility to prepopulate their tax returns with data provided through Medicare Australia and Centrelink. Dunleavy et al., ‘Australian e-government in comparative perspective’. ↩︎
  13. Dunleavy et al., ‘Australian e-government in comparative perspective’. ↩︎
  14. Eden Estopace, ‘Australia creates new digital agency to oversee government’s ICT projects’, EGov Innovation, 1 January 2016. ↩︎
  15. Digital Transformation Agency (DTA), ‘Whole-of-government transformation vision’, in Digital Transformation Agenda, Australian Government, no date ↩︎
  16. DTA, ‘Consultation’, in Trusted Digital Identity Framework ↩︎
  17. Australian Government, Budget 2018–19, Budget paper no. 1, 1–22 ↩︎
  18. Michael Keenan, ‘Delivering Australia’s digital future’, transcript, 13 June 2018 ↩︎
  19. Fergus Hanson, Preventing another Australia Card fail: unlocking the potential of digital identity, ASPI ICPC, October 2018 ↩︎
  20. Ping Identity, ‘More than 3 million sign up to NSW’s unified SSO portal’, 2018 ↩︎
  21. City of Sydney, City of Sydney open data portal ↩︎
  22. European Commission, EU-wide digital once-only principle for citizens and businesses: policy options and their impacts, 1 February 2017 ↩︎
  23. European Commission, The new European Interoperability Framework, 13 July 2018, online. The DTA also has a ‘tell us once’ principle; DTA, Digital Transformation Agenda ↩︎
  24. For more information about X-Road in Estonia, see Information System Authority, Data Exchange Layer X-Road, Republic of Estonia, 21 February 2017, online; and ‘X-Road’, Cybernetica, online. One video on e-Estonia is ‘Living in a digital society: e-Estonia’, YouTube, 21 May 2015 ↩︎
  25. Catherine Garner, ‘Can Australia lead the world in e-government?’, The Canberra Times, 27 September 2016 ↩︎
  26. More information on eID in Estonia is accessible at ID, online; and ‘Estonian e-identity corner stone: state issued national ID card’, YouTube, 10 July 2013 ↩︎
  27. Key generation is performed on the user’s card and not by a central facility. ↩︎
  28. Under the Archives Act, all data and information held by the government is owned by the government. Intellectual property may be owned by the originator of the data, but not the object within which it’s contained. Legislative changes are in train to expand the definition so that it isn’t just property based. Legal dilemmas beyond the scope of this paper include whether access approval can be separate from ownership and how far that extends. Another is what happens to and who owns personal data if someone dies. ↩︎
  29. Integrated e-government inherently presents a large and attractive target for attack. To mitigate this, the basic systems participating as servers in this environment must meet ASD EPL levels of security compliance, preferably at EAL4+ and OSLSPP. OSLSPP enables full separation of data and processes with high trust. ↩︎
  30. For some systems, such as those using Windows XP, this wouldn’t be possible to guarantee. ↩︎

Mapping Xinjiang’s ‘re-education’ camps

This report by ASPI’s International Cyber Policy Centre collates and adds to the current open-source research into China’s growing network of extrajudicial ‘re-education’ camps in Xinjiang province.

The report contributes new research, while also bringing together much of the existing research into a single database. This work has included cross-referencing multiple points of evidence to corroborate claims that the listed facilities are punitive in nature and more akin to prison camps than what the Chinese authorities call ‘transformation through education centres’.

By matching various pieces of documentary evidence with satellite imagery of the precise locations of various camps, this report helps consolidate, confirm and add to evidence already compiled by other researchers.

Key takeaways

  • This ASPI ICPC report covers 28 locations, a small sample of the total network of re-education camps in Xinjiang. Estimates of the total number vary, but recent media reports have identified roughly 180 facilities and some estimates range as high as 1,200 across the region.
  • Since early 2016 there has been a 465% growth in the size of the 28 camps identified in this report.1 2
  • As of late September 2018—across the 28 camps analysed—this report has measured a total of 2,700,000 m2 of floor space, which is the equivalent of 43 Melbourne Cricket Ground stadiums.
  • The greatest growth over this period occurred across the most recent quarter analysed (July, August and September 2018), which saw 700,000 m2 of floor space being added across the 28 camps.
  • Some individual facilities have experienced exponential growth in size since they were repurposed and/or constructed. For example, a facility in Hotan that the New York Times reported on in September 20183 expanded from 7,000 m2 in early 2016 to 172,850 m2 by September 2018—a 2469.29% increase over an approximately 18-month period.
  • The growth in construction has increased at a considerably faster pace in the summer months, with a spike in construction during the third quarters of both 2017 and 2018.

Introduction

China’s censors have been expunging evidence of the country’s vast network of extrajudicial ‘re-education’ camps in Xinjiang province from the internet just as fast as researchers have been finding it.

From first-hand testimony to satellite imagery, researchers have now provided empirical data that authoritatively paints a picture of the extent of China’s biggest human rights abuse since the 1989 post-Tiananmen purge.

Word of this rapidly growing network of ‘re-education’ camps first started to spread with interviews of the relatives of detainees.4 Further research drew on information in public construction and service tenders which documented and detailed the sizes and security features of these re-education camps.5

Other documents such as public recruitment notices, government budget reports, government work reports and Chinese articles in local media and social media have helped to reveal details of how Chinese authorities are rapidly expanding this network of camps.

The cumulative effect of this onslaught of evidence, as well as condemnation from US lawmakers6 and the UN,7 has forced Chinese authorities to move from outright denial of the camps’ existence to a public relations offensive in which they present the camps as places for ‘free vocational training’8 rather than anything punitive.

This ASPI ICPC report contributes new research, while also bringing together much of the existing research into a single database. This work has included cross-referencing multiple points of evidence to corroborate claims that the listed facilities are punitive in nature and more akin to prison camps than what the CCP calls ‘transformation through education centres’.

The report matches the plethora of documentary evidence already uncovered with satellite imagery of this sprawling network of camps. The report takes a conservative approach in deciding what the likely use of each facility is. Each potential camp is assigned a red, orange or green tag representing our level of confidence based on the available open-source data.

The data

This report collects and collates a huge amount of data and it attempted to include as much of that as possible into a database. Some subsets of the database are new—for example, our data on the growth in the size of these 28 facilities. Others have been identified by other researchers, NGOs or media outlets. Where possible, data from these sources has been included in the database, with citations and hyperlinks to the original work.

Brief summaries of the collected data are presented and tabulated in this report; however, using the accompanying database, it is possible to explore all data points in more depth and draw individual conclusions. 

The database is by no means an exhaustive list and it will continue to develop and grow as additional datasets are added.9 It is hoped it will provide media outlets, researchers and governments with current and useful information, and become a resource to which they can potentially contribute.

Camps that have multiple points of strong evidence are deemed to be internment camps and were marked green using the traffic light system. These points of evidence include, for example, facilities that are described as ‘transformation through education’ facilities in official documents, that this research has geo-located from tender documents, or that contain physical features captured in satellite imagery such as barbed wire, reinforced walls and watchtowers. 

Orange tags on other camps denote a comparatively smaller amount of publicly available evidence necessary to conclude the ultimate use of the facilities. Red camps denote minimal or incomplete evidence. Because of that lack of evidence, they have not been included in the public database.

This is not meant to suggest that the scope and scale of the system is small. Agence France-Presse (AFP) estimates there are at least 181 such facilities in Xinjiang,10 while research by German-based academic Adrian Zenz suggests there may be as many as 1,200 facilities.11

Instead, this report and its underlying database aim to create a repository of existing research into the Xinjiang camps in order to save for posterity the information that China’s censors are rapidly deleting from the public record.

Figure 1: Heat map showing the distribution and size of the 28 camps across Xinjiang province. The larger the combined size of facilities in an area, the darker the shade on the map. Kashgar City and its surrounds feature the highest density of facility floor space and are therefore likely where the greatest numbers of re-education detainees are held.

Figure 2: The cumulative floor area in the analysed facilities. Following the second quarter of 2017, many already-constructed buildings were converted into re-education facilities (separated into camps tagged green and orange). 

Figure 3: The rate of quarterly additional construction. Spikes can be seen during the summer months (third quarters) of 2017 and 2018. Growth so far in 2018 (1.169 million square metres) has already outpaced growth in the entirety of 2017 (918,000 m2).

Case studies

The devil is in the detail: The Kashgar City Vocational Technical Education Training Center12

Coordinates: 39°27’9.59″N, 76°6’34.24″E

Last month, Global Times editor Hu Xijin visited what he referred to as a ‘vocational training center’ in Kashgar. He posted a two-minute video of the trip on his Twitter account.13

Hu visited Middle School No. 4 located to the east of Kashgar City. This school, as well as Middle Schools 5 and 6, were under construction across the first half of 2017. Over the summer break, ovals at Middle Schools 5 and 6 were turfed with grass. These schools were being built adjacent to two other schools—the Kashgar City High School and the Huka Experimental Middle School (沪喀实验中学).

But by July 2017, when construction was complete, every ‘school’ building in the southwest of the facility (previously Middle School No. 5) was surrounded by tall fencing that had been painted green and topped with razor wire. By August, much of School No. 6 was enclosed with similar fencing. Upon completion in around November 2017, School No. 4 was also highly securitised and a tender was released calling for bidders to oversee and install new equipment, including a new surveillance camera system.14

In March 2018, one of the previously turfed sports ovals was demolished and replaced by four large six-storey buildings, totalling roughly 50,000 m2 of floor space. Each was surrounded by six 10-by-18 m fenced yards for detainees.

Kashgar City High School and Huka Experimental Middle School, only 50 m to the north of Kashgar Middle School No. 4, paint a dramatically different picture. Basketball courts are filled with students playing outside, and people can be seen in satellite imagery walking between buildings in the schools and on the large sports fields. 

The video posted by Hu Xijin of Middle School No. 4 on 24 October shows detainees dancing and playing table-tennis and basketball. However, this visit—and the footage shared on social media—may not reflect the regular daily experiences of the detainees.

Through satellite and imagery analysis—including imagery updated daily—we can determine that these courts are coloured mats that are recent additions to the camp. The mats were placed on a concrete-covered area that is normally bare and appears inaccessible to detainees.

Lifted edge of the basketball mat suggests that these courts are likely not permanent.

Across 25 satellite images between August 2017 and August 2018, which show the facility since its construction, not a single image featured these outdoor courts. But these coloured mats do appear in satellite imagery available from 10 October. Global Times editor Hu Xijin posted about his visit to these facilities on Twitter and Weibo on 24 October.15

The location filmed by Hu Xijin in Kashgar City Vocational Technical Education Training Center. Features outlined in the panorama produced from Global Times reporting correspond to outlines in the same colour in the satellite imagery.

Checking in with the Shule County Chengnan Training Center since the Economist’s May 2018 coverage16

Coordinates: 39°21’27.64″N, 76°3’2.39″E

On 31 May 2018 the Economist included satellite footage of the ‘Shule County Chengnan Training Center’ in a lengthy article it published on China’s ‘apartheid with Chinese characteristics’.17

We have tracked this camp’s enormous growth since the Economist article featured satellite imagery of the camp. Since March 2018—which was the date the satellite image was taken from—the facility has more than doubled in size.

Across the 2.5-year time period covered in this report,18 the facility has grown from 5 to 24 buildings or wings. Its total floor size has increased during that period from 12,200 m2 to 129,600 m2. This represents an increase in size of 1062.3%.

The camp is described in official documents as a ‘transformation through education’ facility, and a tender shows the involvement of the Shule County Justice Bureau.19 Through satellite and imagery analysis, the camp’s physical features—including barricaded facilities, watchtowers, and enclosures surrounded by barbed-wire fencing—can be clearly seen.

But the evidence base for this facility goes beyond satellite imagery, tenders and floor sizes. In addition, we have matched our satellite images to the first-hand accounts, street-view imagery and video footage published by religious freedom advocacy group Bitter Winter in September 2018.20

Bitter Winter’s evidence highlights several key features of the facility. Footage from newly constructed buildings shows the scale of the camp. The reporting detailed the structure of these facilities. Each floor consists of 28 rooms, and each room is monitored by two security cameras.

Footage acquired by Bitter Winter of the Chengnan Training Centre. Features outlined in the photos correspond to outlines in the same colour in the satellite imagery.

Methodology

This report provides a quantifiable picture of the spread and growth of China’s large network of camps throughout the Xinjiang region. These camps were located through various means, including via unique satellite signatures and physical features; official construction bidding tenders from the Chinese government; and media collected from official sources, local and international NGOs, academics and digital activists. Considerable information was drawn from the analysis of freely available or commercial satellite imagery. 

Satellite imagery of these camps shows highly securitised facilities with features such as significant fencing to heavily restrict the movement of individuals, consistent coverage by watchtowers, and strategic barricades with only small numbers of entry points. Often the perimeter around these camps is multi-layered and consists of large walls with tall razor-wire fencing on both the inside and outside. These features allowed us to pinpoint the location of camps mentioned in official construction tenders. 

Locating camps was aided significantly by engaging and sharing information with Shawn Zhang, a student at the University of British Columbia.21 In addition, official media and reporting by NGOs and activists were vital. These sources provided media from some facilities which allowed us to match the features shown—such as buildings and fencing—with the available satellite imagery.

The floor area of every facility was measured. 

The growth in floor area of these facilities was calculated for every quarter from the beginning of 2016 to September 2018. In most cases, this process involved measuring the roof area of every building using Google Earth imagery and other commercial satellite imagery collected by Digital Globe. Floor area was then calculated by multiplying roof area by the number of storeys in each building. The number of storeys was estimated from satellite imagery by either counting the externally visible windows when the building’s facade was shown or, when the facade was not prominently featured, by analysing the length of the shadows cast by the building. Where footage of these buildings from the ground existed, this was used as the primary source for the number of storeys. 

Some facilities contained additional buildings that were constructed after the most recently available Digital Globe imagery. For these cases, the floor area was calculated from lower resolution (3 m pixels as opposed to 30–50 cm pixels) imagery provided by Planet Labs.

No attempt was made in this analysis to differentiate between buildings used for different purposes, and the total area of each facility includes teaching buildings, administrative buildings and dormitories that house detainees. 

In addition, no attempt was made to determine the date of a facility’s first use as a re-education facility. For facilities such as schools or government-built residential housing that have been converted to re-education centres, our measurements represent the total building area within the current facility’s boundaries. 

These measurements were translated into chronological growth by cross-referencing building measurements with monthly satellite imagery accessed through Planet Labs’ Explorer portal to determine the period of time over which each building was constructed or completed. Some buildings that were too small to register in Planet Lab’s lower resolution imagery, such as single-storey utility buildings or sheds, were not included in this analysis. This data can be found in the database accompanying this report.

Facilities were then matched to publicly available construction tenders released by local governments using Chinese-language web-searching and links collected by other researchers (chiefly, Adrian Zenz, a China security expert at Germany’s European School of Culture and Theology). Saving this information often involved a race against time to gather the data before the documents were removed by those censoring China’s cyberspace. Every important document discovered and included in our database was permanently archived online.

Finally, the report drew on media reporting in local, national and international outlets. This media collection—including photographs, videos and geographical data—was used to further confirm key details such as the location, use or purpose, and physical features of each facility.

Conclusion

The speed with which China has built its sprawling network of indoctrination centres in Xinjiang is reminiscent of Beijing’s efforts in the South China Sea. Similar to the pace with which it has created new ‘islands’ where none existed before, the Chinese state has changed the facts on the ground in Xinjiang so dramatically that it has allowed little time for other countries to meaningfully react.

This report clearly shows the speed with which this build-out of internment camps is taking place. Moreover, the structures being built appear intended for permanent use. Chillingly, stories of detainees being released from these camps are few and far between.

Without any concerted international pressure, it seems likely the Chinese state will continue to perpetrate these human rights violations on a massive scale with impunity.
 

Acknowledgments

ASPI ICPC would like to thank Dr Samantha Hoffman and Alex Joske for their contributions to this research.

This project would not have been possible without the crucial ongoing work of Shawn Zhang, Adrian Zenz, journalists and civil society groups.

Important disclaimer

This publication is designed to provide accurate and authoritative information in relation to the subject matter covered. It is provided with the understanding that the publisher is not engaged in rendering any form of professional or other advice or services. No person should rely on the contents of this publication without first obtaining advice from a qualified professional person.


© The Australian Strategic Policy Institute Limited 2018

This publication is subject to copyright. Except as permitted under the Copyright Act 1968, no part of it may in any form or by any means (electronic, mechanical, microcopying, photocopying, recording or otherwise) be reproduced, stored in a retrieval system or transmitted without prior written permission. Enquiries should be addressed to the publishers. Notwithstanding the above, educational institutions (including schools, independent colleges, universities and TAFEs) are granted permission to make copies of copyrighted works strictly for educational purposes without explicit permission from ASPI and free of charge.

  1. The centre featured on state broadcaster CCTV last week is one of at least 181 such facilities in Xinjiang, according to data collected by AFP, online. ↩︎
  2. tandfonline.com ↩︎
  3. Listed as Camp 5 in the ICPC public database, online. ↩︎
  4. hrw.org ↩︎
  5. jamestown.org ↩︎
  6. cecc.gov ↩︎
  7. theguardian.com ↩︎
  8. globaltimes.cn ↩︎
  9. If you would like to highlight new or missing information that you think should be added to the database, please contact icpc@aspi.org.au ↩︎
  10. hongkongfp.com ↩︎
  11. washingtonpost.com ↩︎
  12. Camp 15 in the ICPC public database. ↩︎
  13. twitter.com ↩︎
  14. jzbnet.com ↩︎
  15. twitter.com ↩︎
  16. Camp 3 in the ICPC public database. ↩︎
  17. economist.com ↩︎
  18. January 2016 to September 2018. ↩︎
  19. archive.org ↩︎
  20. bitterwinter.org ↩︎
  21. Shawn Zhang’s Medium blog can be found here: medium.com ↩︎

Picking flowers, making honey

The Chinese military’s collaboration with foreign universities.

What’s the problem?

China’s People’s Liberation Army (PLA) is expanding its research collaboration with universities outside of China. Since 2007, the PLA has sponsored more than 2,500 military scientists and engineers to study abroad and has developed relationships with researchers and institutions across the globe.1

This collaboration is highest in the Five Eyes countries, Germany and Singapore, and is often unintentionally supported by taxpayer funds.2 Australia has been engaged in the highest level of PLA collaboration among Five Eyes countries per capita, at six times the level in the US. Nearly all PLA scientists sent abroad are Chinese Communist Party (CCP) members who return to China on time.

Dozens of PLA scientists have obscured their military affiliations to travel to Five Eyes countries and the European Union, including at least 17 to Australia, where they work in areas such as hypersonic missiles and navigation technology. Those countries don’t count China as a security ally but rather treat it as one of their main intelligence adversaries.3

The activities discussed in this paper, described by the PLA as a process of ‘picking flowers in foreign lands to make honey in China’ (异国采花，中华酿蜜), risk harming the West’s strategic advantage.4

Helping a rival military develop its expertise and technology isn’t in the national interest, yet it’s not clear that Western universities and governments are fully aware of this phenomenon.5 Some universities have failed to respond to legitimate security concerns in their engagement with China. Current policies by governments and universities have not fully addressed issues like the transfer of knowledge and technology through collaboration with the PLA. Clear government policy towards universities working with the PLA is also lacking.6

What’s the solution?

Understanding and responding to PLA collaboration will require closer engagement between governments and universities. While universities haven’t self-regulated on this issue and haven’t controlled the associated security risks, universities and researchers will not effectively limit the risks of PLA collaboration on their own until governments develop clear policies on it.

Governments need to explore a wider range of tools for limiting technology transfer, including better scrutiny of visa applications by Chinese military scientists and further legislation targeting military end users.

Governments should also consider increasing funding to strategic science and technology fields, while actively limiting problematic foreign investment in those fields. Universities must recognise the risks of such collaboration and seek to learn the extent and nature of their collaboration with the PLA by actively working with government, civil society and security professionals.

Introduction

In 2017, the head of the American Association for the Advancement of Science said that ‘Scientific progress depends on openness, transparency and the free flow of ideas.’7 This collaborative and open spirit, including collaboration with Chinese scientists, has led to some of the great scientific achievements of recent times.8

While countries such as Australia and the US pride themselves on their scientific achievements, their universities and research institutes face limited or declining domestic funding.9 To address these issues, many universities have turned to China—an emerging scientific powerhouse that has sought to build ties to scientific communities around the world.10 This collaboration has generally been a productive and welcome part of the Australia–China relationship. 

The Chinese military has also ridden this wave of research collaboration, sponsoring more than 2,500 scientists to travel to universities in technologically advanced countries such as Australia as students or visiting scholars over the past decade.11 The volume of peer-reviewed literature produced by PLA scientists in collaboration with foreign scientists each year has grown steadily since 2008, following increases in the number of PLA scientists sent abroad (Figure 1).12 Those scientists work in strategic and emerging technology sectors such as quantum physics, signal processing, cryptography, navigation technology and autonomous vehicles.

The PLA’s program of sending scientists abroad is different from standard military exchanges, in which military officers visit each other’s institutions. Those open exchanges build understanding, communication and relationships between militaries.

Figure 1: PLA collaboration, as measured by the number of peer-reviewed articles co-authored by PLA scientists with overseas scientists, 2006 to 2017

In contrast, the PLA National University of Defense Technology (NUDT, 解放军国防科学技术大学) appears to conceive of its military exchanges separately from its international research ties, which are concentrated in foreign universities and not military institutions.13 Scientists sent abroad by the PLA have minimal or no interaction with military personnel in their host countries. Some of those travelling overseas have actively used cover to disguise their military affiliations, claiming to be from non-existent academic institutions.

Around half of those sent abroad are PhD scholars who either complete their doctorates overseas or spend up to two years as visiting PhD scholars and who can usually be identified by searching peer-reviewed literature. While most come from NUDT, the Army Engineering University is another major source.14 The remaining half are sent overseas for short-term trips, spending up to a year as visiting scholars. Few of those scientists have left online traces of their time overseas.

While foreign universities’ ties with the PLA have grown, it isn’t clear that universities have developed an understanding of the PLA and how their collaboration with it differs from familiar forms of scientific collaboration. To date, there’s been no significant public discussion on why universities should be directly contributing to the technology of a non-allied military. Importantly, there’s also little evidence that universities are making any meaningful distinction between collaboration with the Chinese military and the rest of their collaboration with China.

A handful of universities have strongly defended their collaboration with the PLA. Among universities in Five Eyes countries, the University of New South Wales (UNSW) has published the most peer-reviewed literature in collaboration with PLA scientists. After attracting scrutiny for this collaboration, the university’s deputy vice-chancellor wrote, ‘Any fears that our intellectual property or security is undermined through our work with international partners are entirely unfounded.’15

Australia’s Curtin University has described its collaboration with the PLA in similar terms, insisting that work by its scientists with PLA experts on explosions and projectiles doesn’t violate any laws and is civilian research.16

Government research agencies have also engaged in collaboration with the PLA. For example, researchers at the Australian Government’s Commonwealth Scientific and Industrial Research Organisation (CSIRO) have collaborated with NUDT scientists on cloud computing technology.

Those same NUDT scientists were using cloud computing technology for combat simulations.17 Large sums of government funds have been used for collaboration with PLA scientists. One professor at UNSW, for instance, worked with PLA scientists using Australian Research Council grants worth $2.3 million.18 Internationally, defence funding has also been used for research with PLA scientists; for example, a paper written by University of Manchester scientists with a visiting student from NUDT lists US Air Force and Navy grants as funding sources.19

International military–civil fusion

In China, the PLA’s overseas research collaboration is described in frank terms. The PLA Daily uses the saying ‘Picking flowers in foreign lands to make honey in China’ to explain how it seeks to leverage overseas expertise, research and training to develop better military technology.20

This is one aspect of what China calls ‘military–civil fusion’ (军民融合). The term refers to China’s efforts to improve its military’s ability to take advantage of the creativity of the civilian sector and develop its own indigenous military–industrial complex. Described by PLA experts as a ‘cornerstone of PRC national defense reform’, military–civil fusion is helping to drive the modernisation of the PLA.21

So important is military–civil fusion to President Xi Jinping’s military reforms that he described it earlier this year as a prerequisite for building strategic capabilities and a strong military.22

Illustrating the benefits that the PLA obtains from its overseas research collaboration, a publication run by China’s Ministry of Education stated that NUDT’s collaboration with the University of Cambridge to train visiting PLA students will ‘greatly raise the nation’s power in the fields of national defence, communications, anti-jamming for imaging and high-precision navigation’.23 Likewise, before travelling to Sweden for doctoral studies in quantum physics, an NUDT scientist was told by his supervisor, ‘Without breakthroughs in physics, how can there be rapid developments in weaponry?’24

Figure 2: Lieutenant General Yang Xuejun (2nd from right) and Xi Jinping, chairman of the Central Military Commission, in July 2017

Lieutenant-General Yang Xuejun (杨学军, Figure 2), who oversaw a substantial rise in NUDT’s overseas links when he was its president from 2011 to 2017, appears to be one of the key figures behind this phenomenon. NUDT, as the Chinese military’s largest science and technology university, can be seen as representative of broader initiatives in this area. The university is the main source of PLA scientists studying abroad and by 2013 had reportedly sent more than 1,600 scientists overseas as students or visiting scholars, including roughly a third of its PhD scholars.25 An article written by NUDT scholars claims that the university received 300m renminbi ($A60m) from the Chinese government to send 765 graduate students to study abroad.26 According to General Yang, who has implied that NUDT’s overseas ties are a form of military–civil fusion, the university ‘has already reaped great benefits from going down the open university path and the military–civil fusion road’.27

General Yang’s recent promotion to membership of the 205-member 19th CCP Central Committee and to leadership of the Academy of Military Sciences, the PLA’s premier research institution, reflects Xi Jinping’s emphasis on ‘rejuvenating the military with science and technology’.28 It was probably also a recognition of the success with which Yang developed NUDT’s international ties.

Yang, himself a supercomputer expert, has collaborated extensively with UNSW and ran the program to develop the Tianhe-1A supercomputer, once ranked as the world’s fastest supercomputer.29 The NUDT supercomputer program’s role in nuclear weapons testing led to NUDT being placed on the US Government’s Entity List in 2015, meaning that the university faces stricter export controls, yet substantial numbers of NUDT scientists continue to train outside China, including in the US, the UK and Australia.30

The PLA encourages scientists to work on areas of interest to the military while they’re overseas. For example, a 2016 article by NUDT specialists in graduate student education recommends that, in choosing where to study overseas, students’ first priority should be the relevance of the research direction of an overseas institution to their work in China, as they ‘must comprehensively consider the continuity of their research work when in China with that when they are studying overseas’.31 When students are overseas, the report adds, they should ‘fully take advantage of the cutting-edge research conditions and environment abroad’ and ‘map out the arrangements of their overseas research and their plans for research after returning to China’. This alignment of domestic and overseas work indicates that the cases of PLA scientists gaining skills while in Australia that they then use for military projects aren’t outliers; they’re representative examples.32

Sources of and destinations for PLA scientists

PLA scientists come from a wide range of institutions and disciplines within the Chinese military. Analysing peer-reviewed publications co-authored by PLA scientists and overseas scientists indicates that the US, the UK, Canada, Australia and Germany were, in that order, the top five countries engaged in research collaboration with the PLA in 2017 (Figure 3). Those countries appear to be the primary destinations for PLA scientists sent abroad.

Figure 3: The top 10 countries for PLA collaboration, as measured by peer-reviewed literature co-authored by PLA scientists, 2006 to 2017

PLA scientists sent abroad as visiting scholars came from institutions such as:

  • the Northwestern Institute of Nuclear Technology (西北核技术研究所), which works on nuclear and high-power microwave weapons
  • the Chemical Defense Institute of the Academy of Military Sciences (军事科学院防化研究院), which specialises in chemical weapons research and has sent a sarin gas expert overseas
  • the Navy Submarine Academy (海军潜艇学院) in Qingdao
  • the Armored Forces Engineering Academy (装甲兵工程学院) in Beijing, which works on tank technology
  • the China Aerodynamics Research and Development Center (中国空气动力研究与发展中心), which has sent scramjet researchers to study overseas
  • the Rocket Force Engineering University (火箭军工程大学), which conducts research for China’s missile programs
  • the Academy of Equipment Command and Technology (装备指挥技术学院), which in 2007 sent a specialist in antisatellite weaponry to the University of Michigan using civilian cover.33

The volume of peer-reviewed literature co-authored by PLA researchers and overseas researchers is a rough indicator of the level of PLA collaboration at each university. Figure 3 shows that the leading countries for PLA collaboration by this measure for 2017 were, in order, the US, the UK, Canada, Australia and Germany, indicating that they’re likely to be the main destinations for PLA scientists studying abroad. Singapore, Sweden and the Netherlands are other major destinations for PLA scientists. Over the past decade, Australia has been engaged in the highest level of this collaboration among the Five Eyes countries per capita, at six times the level in the US.

It’s also possible to estimate the number of PLA scientists sent to each country since 2007, based on the above findings.34 Approximately 500 Chinese military scientists were sent to each of the UK and the US, roughly 300 each to Australia and Canada and more than 100 each to Germany and Singapore. Hundreds more have been sent to other countries, including the Netherlands, Sweden, Japan and France.

Figure 4, using the same dataset, shows the top 10 universities outside China for PLA collaboration. Nanyang Technological University in Singapore has the highest level of PLA collaboration, followed closely by UNSW in Australia. Other universities in Canada, Australia, the UK and the Netherlands also engage in high levels of collaboration with the PLA.35

Figure 4: The top 10 universities outside of China for PLA collaboration, as measured by the number of peer-reviewed publications, 2006 to 2017

The PLA’s links to universities across the world go beyond student admissions. The Chinese military, through its own universities and research institutions, has worked to build relationships with overseas universities and leading overseas researchers. A 2014 document published by NUDT claimed that the university had recruited 20 foreign nationals as teachers and ‘established academic relationships with over 100 universities and research units in over 50 countries and regions’.36

Scientists from Australia, the UK and the US are listed as potential doctoral supervisors for NUDT students in 2018.37

NUDT has also built ties with overseas universities at the institutional level. For example, NUDT’s Quantum Information Interdisciplinary Talent Training Program cooperates with the University of Cambridge’s Cavendish Laboratory.38 The People’s Daily claimed that, in addition to agreements with Oxford and Cambridge, NUDT has established ‘overseas study bases’ at institutions including Harvard University.39 New Zealand’s Massey University also signed a memorandum of understanding with NUDT in 2008.40

Maintaining loyalty to the CCP

The PLA, as the armed wing of the CCP, insists that all overseas party members strictly abide by ‘external exchange discipline standards’.41 According to the PLA Daily, ‘the openness of internationally expanding talent cultivation does not represent a “relaxation”, and we certainly cannot “let go”.’42 General Yang Xuejun has also specifically warned of the need to carefully manage military secrets while increasing the university’s openness.43

Those permitted to study overseas go through intensive training prior to their departure and are ‘all budding shoots with good grades and strong potential for innovation’.44 Alongside academic credentials, political credentials are also of key importance for military scientists hoping to study abroad. The PLA Daily warns that, if students sent overseas ‘develop issues with their politics and ideology, the consequences would be inconceivable (后果不堪设想)’.45 NUDT therefore appears to sponsor only CCP members for overseas study and works hard to maintain their loyalty to the party and negate ‘all kinds of harmful ideologies’.46 Reportedly, all 200 students and researchers from NUDT who were studying or visiting overseas in 2013 were party members.47

The People’s Daily claimed in 2013 that students sent overseas by NUDT had established eight party branches overseas and organised events for party members, so that ‘personnel studying abroad would keep their convictions rock-solid’ (坚守信念如磐).48 Another report from 2015 claimed that NUDT’s College of Optoelectric Science and Engineering alone had established 10 overseas party branches.49 More recent reports hint that such branches are still being established. For example, party media reported in October 2017 that students from one of NUDT’s colleges had established a WeChat group for the college’s more than 30 students overseas to study the 19th Party Congress.50 ‘Their red hearts,’ the report concluded, ‘look to the party.’

Party branches have also been used to coerce overseas Chinese scholars. An investigation by Foreign Policy found that some visiting students from Chinese universities who formed party branches abroad were asked to report on any subversive opinions held by their classmates.51 It’s probable that similar kinds of pressure are exerted on overseas PLA researchers.

Online communication forms an important part of PLA efforts to maintain discipline among overseas personnel and is complemented by in-person contact. One report stated that students from NUDT’s College of Optoelectric Science and Engineering ‘regularly chat with College leaders by video call and exchange emails with NUDT academic supervisors and student cadres to discuss their thoughts, exchange ideas on academic matters, and clarify points of interest’.52 Regulations on the political education of overseas students by the same NUDT college include provisions for ‘overseas inspection’ and for students to return to China in the middle of their study for ‘remedial education’.

One NUDT professor used a trip to an overseas conference as an opportunity to meet eight NUDT scientists studying in the region to ‘pass on the greetings and requests of party organisations’. The regulations also include provisions for ‘joint education and interaction with families’, which may imply that pressure on the family members of overseas PLA scientists is used to maintain discipline.53

The close watch that the PLA keeps on its overseas scientists helps ensure that all those sent abroad return to the Chinese military. NUDT, for example, requires that those applying to study abroad show their intent to return to ‘serve the construction of the nation, national defence and the military’.54

The PLA Daily claimed in 2013 that all the students whom NUDT had sent abroad in recent years returned on time to ‘become key forces in their work units’.55

Institutes that don’t exist: deception by PLA scientists

While most scientists sent abroad by the PLA appear to be open about which institutions they come from, this report has identified two dozen new cases of PLA scientists travelling abroad using cover to obscure their military affiliations. In at least 17 of these cases, PLA scientists used cover to travel to Australia. These scientists use various kinds of cover, ranging from the use of misleading historical names for their institutions to the use of names of non-existent institutions.

Features of deception by the PLA

An article from 2002 on the website of a Chinese overseas study agency offers insights into the use of cover. In response to a question asking whether having graduated from a military institution would affect one’s ability to get an overseas visa, the company responded: 

Many military colleges and military units externally have common names (民间称呼) that don’t reveal their military characteristics. NUDT, for example, is externally known as Changsha Institute of Technology. This is the best way [to avoid having your visa application rejected].56

The Changsha Institute of Technology was a PLA institution subsumed by NUDT in 1975.57 While the quote above doesn’t come from an official source, it at least indicates how these unsophisticated but nonetheless effective covers are understood as tools for hiding one’s military background.

Besides using non-existent institutions with innocuous-sounding names as cover, PLA members also claim to be from real civilian institutions in the same regions as their military units. New Zealand MP Yang Jian, for example, who taught intelligence officers at the PLA Foreign Languages Institute in Luoyang, claimed in his New Zealand residency application to have worked at Luoyang University.58 Before moving to New Zealand in 1999, Yang received an Australian Government aid scholarship to study at the Australian National University, earning a master’s degree and doctorate in international relations. During that period, he interned at the Senate Standing Committee on Foreign Affairs, Defence and Trade, and headed the Canberra Chinese Students and Scholars Association, which retains intimate ties to the Chinese Embassy to this day.59 Yang told media, ‘the system asked me to use the partner university,’ referring to Luoyang University.60

A number of PLA scientists using cover to travel abroad have created LinkedIn profiles using their cover institutions, which may have been used to shore up their claimed affiliations while overseas.61

The use of cover appears to be managed differently by each institution, some of which use cover far more often than others.62 Cover is also not used consistently within each institution. As described below, PLA Information Engineering University (PLAIEU) researchers have both used cover and openly stated their affiliation at the same conferences. It’s unclear whether this indicates that the use of cover is up to the discretion of each researcher or perhaps that it relates to the sensitivity of a researcher’s work or position in the PLA.

NUDT appears to no longer use the ‘Changsha Institute of Technology’ as cover, but it engages in a different kind of deception. A document published by NUDT for students hoping to study abroad advises them that, when providing documentation in their applications to foreign institutions, ‘military and political courses can be excluded’ from their academic records.63 This appears designed to mislead overseas authorities, universities and researchers by downplaying the extent to which NUDT is a military institution and to which these students are military scientists.

The Xi’an Research Institute of High Technology

Scientists from the PLA Rocket Force Engineering University (RFEU, 火箭军工程大学)64, a key research base for the PLA Rocket Force, claim to be from the ‘Xi’an Research Institute of High Technology’ (西安高技术研究所), which appears to only exist on paper.

At least five RFEU scientists claiming to be from the Xi’an Research Institute have travelled overseas as visiting scholars, including one of the PLA’s leading missile experts, Major General Hu Changhua (胡昌华), and three of his close associates at RFEU. General Hu (Figure 5), who heads RFEU’s Missile Testing and Control Simulation Experimental Teaching Centre, visited the University of Duisburg–Essen in Germany for four months in 2008.65 It’s unclear what he worked on in Germany, as he didn’t publish any papers while there, but his work for the PLA focuses on flight control systems and fault diagnosis for missiles.66

Two RFEU scientists who frequently publish with Hu, Zhou Zhijie (周志杰)67 and Wang Zhaoqiang (王兆强),68 were visiting scholars at universities in England; they claim in their English publications to be from the Xi’an Research Institute.69

Figure 5: Major General Hu Changhua, profiled by China Central Television’s military affairs channel in 2016:

‘Right now I’m a professor at RFEU and head of the Military Key Lab on Missile Testing and Control Technology.’ 

Figure 5: Major General Hu Changhua, profiled by China Central Television’s military affairs channel in 2016: ‘Right now I’m a professor at RFEU and head of the Military Key Lab on Missile Testing and Control Technology.’


Source: CCTV, 28 October 2016, YouTube.

Hu Xiaoxiang: a case study

Identifying the Xi’an Research Institute of High Technology as a cover institute helps shed light on the January 2015 expulsion from Norway of a Chinese scientist and his supervisor, a dual citizen of Germany and Iran. The expulsion came after Norwegian authorities determined that the work of the Chinese scientist, later named in court as Hu Xiaoxiang (扈晓翔), could be used to develop hypersonic cruise missiles (Figure 6).70

Figure 6: Hu Xiaoxiang

Hu wrote five papers with his supervisor at the University of Agder, all of which listed the Xi’an Research Institute as his affiliation. The papers focused on air-breathing hypersonic vehicles, which travel at over five times the speed of sound and ‘can carry more payload than ordinary flight vehicles’.71 Hu’s work was supported by a Norwegian Government grant for offshore wind energy research.72

Besides his affiliation with the Xi’an Research Institute, there’s a large body of evidence tying Hu to RFEU. The website of RFEU’s missile research centre states that Hu Xiaoxiang won an award in 2014 for his PhD thesis on hypersonic aircraft, supervised by General Hu Changhua.73 The website also says that in 2014 he received 250,000 renminbi (A$50,000) from the Chinese Government for a three-year research project on hypersonic aircraft (Figure 7).74 In 2016, he was described as a lecturer at the centre, which received 14 awards for missile research between 2010 and 2014.75 In some publications, Hu also listed the Harbin Institute of Technology, a civilian university heavily engaged in military research, as a second affiliation.76

Relations between China and Norway were put on ice when the Nobel Peace Prize was awarded to Chinese democracy activist Liu Xiaobo in 2010, and the Chinese Government was quick to attack Norway for Hu’s expulsion.77 Only in December 2016 did the two countries ‘normalise’ diplomatic relations. Public statements by Norwegian authorities didn’t explain the Chinese scientist’s military affiliation or mention the Xi’an Research Institute, as the information was likely classified.

Figure 7: A paper published by Hu Xiaoxiang shortly after his expulsion from Norway, stating an affiliation with RFEU in the Chinese version of the abstract but the Xi’an Research Institute in the English version.

A few months later, in September 2015, a court overturned the expulsions. Hu’s lawyer stated after the trial that ‘there is no evidence in the case that my client is part of research collaboration on missiles and weapons with China.’78 The University of Agder lauded the decision as a win for academic freedom.

The Norwegian Government later successfully appealed the overturning of Hu’s supervisor’s expulsion. However, it’s unclear whether any appeal was made in Hu’s own case, which hasn’t been made publicly available.79 Neither the Xi’an Research Institute, Hu Changhua nor RFEU was mentioned in the judge’s ruling on the German-Iranian supervisor’s case or any coverage of the expulsions.

The Zhengzhou Institute of Surveying and Mapping

Among the 40 Chinese military scientists listed as presenting papers at the 9th International Symposium on Mobile Mapping Technology, nine claimed to be from an institution with no apparent military affiliation.80 Most of the other 30 military scientists at the conference, hosted by UNSW in December 2015, were openly from NUDT and a research institute of China North Industries Group Corporation (also known as Norinco Group), China’s largest arms manufacturer; the rest came mainly from the PLA Information Engineering University.

The nine claimed to be from the Zhengzhou Institute of Surveying and Mapping. This institute, which was officially known as the PLA Institute of Surveying and Mapping, no longer exists, having been subsumed in 1999 by PLAIEU—itself a major player in cyber operations and a key training ground for signals intelligence officers.81 The Zhengzhou Institute appears to live on as cover for PLA scientists interacting with foreigners. Nearly 300 peer-reviewed papers have been published by authors claiming to be from the institute.82

The use of the Zhengzhou Institute of Surveying and Mapping as cover doesn’t stop at international conferences. Numerous examples of visiting scholars claiming to be from there have been uncovered for this report. They include Zhu Xinhui (朱新慧), a lecturer at PLAIEU specialising in navigation technology, who visited UNSW from 2015 to 2016.83 In numerous journal articles and in the program of the mobile mapping conference mentioned above, however, she is described as being from the Zhengzhou Institute of Surveying and Mapping.84

Guo Jianfeng (郭建锋), an associate professor at PLAIEU, visited Curtin University for a year in 2014. A specialist on navigation system data processing, Guo was described on the website of Curtin University’s Global Navigation Satellite Systems Research Centre as being on ‘sabbatical leave from the Department of Geodesy of the Institute of Surveying and Mapping, Zhengzhou, China’.85

The Zhengzhou Information Science and Technology Institute

Another cover institute, the Zhengzhou Information Science and Technology Institute (ZISTI), which appears to exist only on paper, has also been widely used by PLAIEU scientists to publish research and travel overseas. More than 1,300 pieces of peer-reviewed literature have been authored by individuals claiming to be from ZISTI.86

One paper in a Chinese-language journal by a PLAIEU researcher, which includes an English version of the abstract and author information, clearly shows that ZISTI is a cover institute (Figure 8). The paper’s Chinese text describes the first author as affiliated with PLAIEU, but the English version describes the
same author as affiliated with ZISTI.87 Nearly all of the authors sampled who claimed an affiliation with ZISTI could be shown to be working at PLAIEU.

Figure 8: Chinese and English versions of a paper published by a PLAIEU scientist, demonstrating the use of the Zhengzhou Information Science and Technology Institute as cover.

Scientists claiming to be from ZISTI have attended international conferences both inside and outside China. For example, seven researchers affiliated with ZISTI are listed in the program of a conference on signal processing at the Gold Coast in Australia in 2014. Experts from American, Australian and Korean
defence research agencies were also in attendance.88

As with the Zhengzhou Institute of Surveying and Mapping, ZISTI has been used as cover for PLA scientists travelling overseas as visiting scholars. For example, Zhu Yijun (朱义君) is an associate professor at PLAIEU specialising in signals engineering.89 Claiming to be from ZISTI, in 2011 he visited Canada’s McMaster University, where he worked on wireless communications technology with wide-ranging military applications.90

PLAIEU scientists claiming to be from ZISTI have also travelled to the US as visiting scholars and for conferences.91

Espionage and intellectual property theft

In addition to their overt activities, PLA researchers, especially those who haven’t been forthcoming about their military affiliations, may engage in espionage or steal intellectual property while overseas. The PLA engages in such high levels of espionage that in 2014 the US Government took the unusual step of publicly indicting five Chinese military hackers.92 Military scientists abroad who regularly communicate with superiors in China, receive visits by superiors while overseas and return home in the middle of their time abroad for ‘remedial education’, as described in the examples outlined above, offer safe and convenient channels for Chinese intelligence agencies to access sensitive information from overseas.93

Amateur collectors with STEM expertise have been implicated in a high proportion of intellectual property theft and espionage cases involving China.94 Scientists and engineers involved in military research projects, while they might not have received formal training as spies, are uniquely qualified to identify and exfiltrate valuable information to overcome specific hurdles in the development of new technologies.

Should universities collaborate with the PLA?

Assessing the costs and benefits of research collaboration with the PLA shows that it comes with significant security risks while offering unclear benefits. It isn’t in the national interest of most of the countries examined in this report to help build the capabilities of a rival military. Other forms of cooperation with the Chinese military, such as joint exercises and exchanges that build understanding and communication, are largely beneficial but distinct from the kinds of research collaboration addressed in this report.

The benefits of research collaboration with the Chinese military are difficult to measure, but could include the following:

  • Training PLA scientists and working with them leads to scientific developments and published research while attracting some funding. 
  • A small proportion of collaboration with the PLA appears sufficiently transparent and falls into areas of fundamental research such that the benefits may outweigh security risks. One possible example is cooperation between the American and Chinese governments on the multinational Daya Bay Reactor Neutrino Experiment, which involves NUDT.

A number of benefits usually associated with research collaboration with militaries and foreign countries haven’t been observed in PLA collaboration:

  • PLA collaboration doesn’t lead to long-term improvement in the talent of institutions and countries accepting PLA scientists, as the PLA claims that 100 per cent of scientists sent abroad by NUDT in the years before 2013 returned to China on time.95 
  • The forms of PLA collaboration studied in this report don’t promote understanding and relationships between militaries, as they aren’t military exchanges and often aren’t overt.
  • While overseas, PLA scientists remain under the close watch of the CCP, which works to ensure that they remain loyal and aren’t influenced by their experience living in free societies. 
  • It’s improbable that PLA scientists working with overseas civilian researchers would share with or disclose to those researchers any significant research breakthroughs of military value.

There are many risks and costs associated with current approaches to training and collaborating with PLA scientists:

  • Training PLA scientists improves the scientific talent and knowledge of a military treated by many as a strategic competitor.96
  • PLA scientists often engage in deception in their interactions with foreign institutions and their staff, making it difficult for those collaborating with them to take appropriate security precautions.
  • PLA scientists could gather intelligence and steal technology while they’re overseas, especially if they’re hiding their military affiliations.
  • Failures to address concerns about PLA collaboration and to develop policies differentiating it from wider engagement with China risk tarring all research ties with China with the same brush.
  • Research collaboration with the PLA contributes to technology that may be used against Australia and its partners in a conflict or for intelligence collection.
  • Universities with ties to the PLA risk eroding trust between themselves and funders of research, such as defence research agencies, scientific agencies and industry.
  • Universities risk reputational damage by collaborating with a non-allied military.
  • Public funding worth millions of dollars is being used for collaboration with a non-allied military, with little to no input from taxpayers.

Current policy and legislation are inadequate

Export controls are the primary mechanism by which countries seek to manage the supply of sensitive technology and goods to overseas entities. However, the ability of export control laws to effectively manage the risks posed by PLA research collaboration is limited. In Australia, few cases of research or cooperation contrary to our national interests are believed to have been prevented through the Defence Trade Controls Act 2012.97 The current review of the Act offers an opportunity to address some of these limitations.

There are a few reasons for these difficulties. First, intangible transfer of technology—the primary form of technology transfer taking place through the kinds of collaboration studied in this paper— is extremely difficult to control in practice because it doesn’t involve the export of physical goods.98 Second, the Act doesn’t regulate the supply of controlled technology, which includes instruction and training, to individuals in Australia even if they’re PLA members. Third, some of this collaboration covers emerging technologies, such as quantum physics, that are important but not included in the Defence and Strategic Goods List, as their applications aren’t yet fully known. Export control lists tend to be slow to incorporate emerging technologies, so regulatory power can come well after issues become apparent. Fourth, the Act doesn’t regulate the supply of controlled technology by Australians when they’re outside of Australia, such as training given to PLA members by Australian academics visiting China.
 

Recommendations

The PLA’s collaboration with foreign universities is growing and the expansion of international ties remains one of NUDT’s priorities.99 The developments outlined in this report warrant more attention and different approaches from those currently employed by most governments and universities. Responses to PLA collaboration need to be informed by clear government policies and move beyond export controls, using the full range of tools available to governments and universities. The Australian Government, for example, can do more to work in partnership with our research sector to advance scientific progress while protecting national security and ensuring that relevant research doesn’t advance the Chinese military’s capabilities.

Based on the findings of this report, it is recommended that governments pursue the following measures:

Deepen discussions within government on PLA collaboration to determine how it relates to the national interest

  • Determine what kinds of collaboration with the PLA should be further controlled or even prohibited and establish clear policy on engagement with PLA research organisations and personnel.
  • Foster international discussions on PLA collaboration to develop multilateral responses.
  • Develop interagency responses to PLA collaboration to ensure better integration of efforts by defence and export control agencies, intelligence agencies and immigration agencies.
  • Share information about cases and trends in PLA collaboration, particularly cases of deception by PLA scientists, with partners across the globe.

Increase communication and outreach to universities, companies and publics

  • Establish a committee bringing together members of the national security community and university leaders. This committee could serve as a forum to share key information and foster a more cooperative working environment while also providing a space for the university sector and national security community to better understand each other’s perspectives. The US Federal Bureau of Investigation’s National Security Higher Education Advisory Board is a useful model to emulate.100
  • Ensure that companies funding research at universities are aware of any PLA collaboration and understand future measures to control such collaboration.
  • Politicians and senior public servants should better articulate what’s in the national interest and publicly explain why advancing China’s military capabilities isn’t in the national interest.101

Improve the scrutiny of visa applications by foreign military personnel

  • Enhance and better coordinate efforts by government agencies such as Australia’s Department of Home Affairs, Department of Defence and Australian Security Intelligence Organisation to ensure that military scientists applying for visas are identified and properly vetted.102
  • Create a list of Chinese and other non-allied military and military-linked research institutions, including civilian universities heavily engaged in military research, for use by immigration officials.

Re-examine export controls

  • The Australian Government should consider further controlling technology transfer to certain end users. Transfers of controlled technology to PLA members and civilians heavily engaged in military research should be restricted regardless of their geographical location.
  • The Australian Government should create a list of entities posing national security risks that are subject to special export licence requirements, modelled on the US’s Entity List.
  • The government should help universities train and provide resources for staff with export control compliance duties.
  • Work continuously with experienced scientists in emerging technology fields to determine whether and how emerging technologies should be controlled.
  • Ensure that universities are fully complying with controls relating to the intangible transfer of technology in their collaboration with the PLA.

Regulate scientific training given to foreign military personnel

  • Introduce legislation that draws on the US Code of Federal Regulations’ rules on defence services, which require those offering training to foreign military personnel to first receive a waiver from the US Department of Defense.103 This could take the form of an expansion of the Defence Trade Controls Act that restricts technology transfer to members of certain governments and organisations.

Regulate the use of government resources in collaboration with the Chinese military and other non-allied militaries

  • Update internal policies in government research institutions such as CSIRO to limit or ban collaboration with non-allied militaries, particularly in dual-use areas.
  • Funding bodies such as the Australian Research Council should prohibit funding in some areas from being used in collaboration with non-allied militaries.
  • Carefully evaluate any collaboration with PLA scientists on government-funded projects, particularly defence projects.

Increase government and other funding for research in strategic research areas

  • Fields such as artificial intelligence and quantum physics should receive more government funding to ensure that talent and ideas stay in Australia.
  • Universities working in strategic research areas should be encouraged to collaborate with allied military and defence countries rather than non-allied militaries.

Limit problematic forms of foreign investment in strategic research areas

  • Investment by Chinese defence companies such as China Electronics Technology Group Corporation into strategically important fields should be prohibited.104

Universities should also pursue the following measures:

Build understanding of PLA collaboration

  • Produce credible and thorough assessments of the extent of PLA collaboration on campuses.
  • Develop processes for managing PLA collaboration so that security risks can be identified and resolved

Raise awareness among employees

  • Ensure that those interacting with members of non-allied militaries take appropriate security precautions.

Exercise greater oversight of visiting scholar and student application

Develop internal policies on collaboration with foreign military personnel

  • Require employees to receive approval before collaborating with or training members of non-allied militaries.

Important disclaimer

This publication is designed to provide accurate and authoritative information in relation to the subject matter covered. It is provided with the understanding that the publisher is not engaged in rendering any form of professional or other advice or services. No person should rely on the contents of this publication without first obtaining advice from a qualified professional person.


© The Australian Strategic Policy Institute Limited 2018

This publication is subject to copyright. Except as permitted under the Copyright Act 1968, no part of it may in any form or by any means (electronic, mechanical, microcopying, photocopying, recording or otherwise) be reproduced, stored in a retrieval system or transmitted without prior written permission. Enquiries should be addressed to the publishers. Notwithstanding the above, educational institutions (including schools, independent colleges, universities and TAFEs) are granted permission to make copies of copyrighted works strictly for educational purposes without explicit permission from ASPI and free of charge.

First published October 2018

  1. This estimate has sought to exclude PLA medical scientists and doctors by not counting those affiliated with PLA medical institutions. Media reports, many of which are cited in this report, were one important source for determining the number of PLA scientists sent abroad. Feng Chunmei 冯春梅, Cai Weibin 蔡渭滨, Li Zhi 李治, ‘Guofang keji daxue shixiang weilai zhanzheng de rencai hangmu’ 国防科技大学 驶向未来 战争的人才航母 [NUDT—An aircraft carrier of talent steering towards future wars], Renmin Ribao 人民日报, 8 August 2013, online, claims that NUDT had sent 1,600 scientists overseas as students or visiting scholars ‘in recent years’. Assuming the 1,600 figure describes the number of NUDT scientists sent abroad between 2007, when the PLA substantially increased the number of scientists it sent overseas, and 2013, this gives roughly 230 NUDT scientists sent overseas each year. Conservatively, this indicates that well over 2,000 NUDT scientists have been sent abroad since 2007. Accounting for the fact that NUDT is responsible for approximately 80% of publications written by PLA scientists with overseas scientists and assuming that represents the proportion of PLA scientists overseas who are from NUDT, this means that more than 2,500 PLA scientists have been sent overseas since 2007. This estimate was also supported by a second set of open-source data which, to prevent the information from being removed, has not been revealed. ↩︎
  2. New Zealand is not counted here, despite being a Five Eyes country. It has high levels of PLA collaboration, especially relative to its population, but is not among the top countries for collaboration more generally. ↩︎
  3. C Uhlmann, ‘China an “extreme” threat to Australia: ASIO’, 9 News, 31 January 2018, online; Bill Gertz, ‘FBI director warns China is America’s most significant intelligence threat’, The Washington Free Beacon, 19 July 2018, online; ‘German intelligence unmasks alleged covert Chinese social media profiles’, Reuters, 10 December 2017. For a discussion of the case of Huang Jing in Singapore, see John Garnaut, ‘Australia’s China reset’, The Monthly, August 2018. ↩︎
  4. Wang Wowen 王握文, ‘Zouchu guomen, dang zuzhi shenghuo “bu diaoxian”’, 走出国门，党组织生活’不掉线’ [Exiting the country, they stay connected with the life of party organisations], Jiefangjunbao 解放军报, 1 July 2015, online. ↩︎
  5. One of the only papers to address research collaboration with the PLA is Elsa Kania, Technological entanglement, ASPI, Canberra, 28 June 2018, online. ↩︎
  6. Section 1286 of the John S. McCain National Defense Authorization Act for Fiscal Year 2019 offers an important starting point for policies on scientific engagement with China and the PLA, seeking to protect scientists from undue foreign influence, safeguard important information and support the growth of domestic talent. ↩︎
  7. Richard Holt, AAAS statement on White House proclamation on immigration and visas, American Association for the Advancement of Science, 25 September 2017, online. ↩︎
  8. See Yangyang Cheng, ‘The future of particle physics will live and die in China’, Foreign Policy, 2 November 2017, for an eye-opening discussion of the level of political involvement in China’s scientific research, even research into particle physics, online. ↩︎
  9. DJ Howard, FN Laird, ‘The new normal in funding university science’, Issues in Science and Technology, 2013, 30(1), online; M Clarke, ‘Federal government university budget leaves 10,000 places unfunded, Universities Australia says’, ABC News, 18 January 2018, online; N Whigham, ‘Medical and scientific research at a crossroads in Australia as funding stagnates’, News.com.au, 7 November 2016. ↩︎
  10. UNSW, for example, has partnered with the Chinese Government’s Torch Program, attracting tens of millions of dollars in R&D funding from Chinese companies. See ‘UNSW celebrates first anniversary of Torch partnership with China’, UNSW Media, 28 March 2017, online. ↩︎
  11. It appears that most of those sent abroad are PLA ‘civilian cadres’ (文职干部), rather than ranking military officers. While they’re counted as members of the PLA, civilian cadres aren’t combat personnel and often work in technical areas, such as scientific research. See information about civilian cadres at the following link. ↩︎
  12. Peer-reviewed literature is the most accessible but not the only measure of PLA collaboration. Other facets of PLA collaboration include visiting and lecturing at PLA institutions, supervising PLA students and visiting scholars, which are correlated with but distinct from the level of peer-reviewed literature. Findings on peer-reviewed literature by PLA scientists with foreign researchers are based on searches in Scopus, the largest database of peer-reviewed literature, covering 16 PLA institutions and aliases. Hong Kong wasn’t counted together with the PRC mainland. Note that publications by PLA scientists from medical institutions have been excluded. The following institutions and aliases were included in the search: National University of Defense Technology, National Key Laboratory for Parallel and Distributed Processing, PLA University of Science and Technology, PLA Information Engineering University, Zhengzhou Information Science and Technology Institute, Zhengzhou Institute of Surveying and Mapping, Air Force Engineering University, Second Artillery Engineering College, Xi’an Research Institute of High Technology, Academy of Armored Force Engineering, Academy of Equipment Command and Technology, National Digital Switching System Engineering and Technological Research Center, Northwest Institute of Nuclear Technology, China Aerodynamics Research and Development Center, Naval University of Engineering and PLA Electronic Engineering Institute. ↩︎
  13. See the section on international ties, which discusses sending students abroad and building academic ties separately from military exchanges, in Liu Hang (ed.), 2015 National University of Defence Technology admissions guide, online. ↩︎
  14. The Army Engineering University was formed in August 2017 through the merger of the PLA University of Science and Technology and a number of other army colleges. See Anonymous, ‘Lujun gongcheng daxue jiepai, you gongchengbing xueyuan deng 5 suo yuanxiao heping zujian’ 陆军工程大学揭牌，由工程兵学院等5所院校合并组建 [The Army Engineering University is unveiled, formed by the merger of the Engineering College and five other institutions], Pengpai 澎湃, 3 August 2017, online. ↩︎
  15. Brian Boyle, ‘Chinese partnerships are vital for universities and global research’, Financial Review, 29 October 2017, online. ↩︎
  16. Clive Hamilton, Alex Joske, ‘Australian universities are helping China’s military surpass the United States’, Sydney Morning Herald, 24 October 2017, online. ↩︎
  17. Clive Hamilton, Silent Invasion, Hardy Grant Books, 2018, 190–193. ↩︎
  18. Hamilton & Joske, ‘Australian universities are helping China’s military surpass the United States’. ↩︎
  19. Mengjian Zhu, Moshe Ben Shalom, Artem Mishchsenko, Vladimir Falko, Kostya Novoselov, Andre Geim, ‘Supercurrent and multiple Andreev reflections in micrometer-long ballistic graphene Josephson junctions’, Nanoscale, 2018, issue 6, online. ↩︎

Huawei and Australia’s 5G Network

Over the course of 2018, ASPI staff and writers for The Strategist participated in a dynamic public debate about the participation of Chinese telecommunications equipment manufacturer Huawei in Australia’s 5G network.

Australia’s 5G network is critical national infrastructure and this was one of the most important policy decisions the government had to make this year.

ASPI felt it was vital to stimulate and lead a frank and robust public discussion, in Australia and throughout the wider region, which analysed and debated the national security, cybersecurity and international implications of Huawei’s involvement in this infrastructure.

In this report, in chronological order, you’ll read a range of views written up in The Strategist, The Australian and The Financial Times.

These articles tackle a variety of issues surrounding the decision, including the cybersecurity dimension, the broader Australia–China relationship, other states’ experiences with Huawei, the Chinese Government’s approach to cyber espionage and intellectual property theft and, importantly, the Chinese party-state’s view of state security and intelligence work.

When it comes to important national security, cybersecurity and critical infrastructure decisions, ASPI will continue to stimulate Australian public discourse and fill gaps in global debates.

We also encourage the Australian Government to take a more forward-leaning approach to its participation in public discourse so that the public and key stakeholders are as informed as possible when hard and complicated policy decisions like this need to be made.