This correspondence is featured in Australian Foreign Affairs 10: Friends, Allies and Enemies.
To read the full issue, log in, subscribe or buy the issue.
I wrote the bulk of my essay “Data Driven” before work, with a grumpy six-month-old either snoring or screaming nearby. There was sleep deprivation, too much caffeine and plenty of anxiety about what to include – and what not to include – given the word limit. Engaging with these thoughtful responses to my essay is helpful – it forces me to remember why, and what, I deliberately pushed aside or shaved off along the way.
Lesley Seebeck, Olivia Shen and Peter Rogers reveal how much more there is to unpack when it comes to the world’s intelligence communities and how they are engaging with, and being reshaped by, cyberspace, data and, increasingly, technology. Seebeck in particular makes some excellent points about the dangers of a technology-driven intelligence process. In this, she essentially outlines an entirely different essay that could, and should, be written.
Both Seebeck (“It’s not more data we need, but more seasoned analysts”) and Chen (“Spycraft will never be solely data-driven”) make the case that people are more vital than ever in assessing intelligence. I couldn’t agree more. Data is useless if you don’t have the right combination of specialists to draw insights from it. Intelligence communities will forever have to balance their investment in the collection process and in the analysts needed to draw insights and make judgements from that collection.
My own centre at the Australian Strategic Policy Institute (ASPI) provides a good example of just how important people are in the process of information collection and assessment, although the information we collect is far more public than that of intelligence agencies. A report published by ASPI’s International Cyber Policy Centre can now attract a readership of hundreds of thousands. Most of these reports are the result of small teams of analysts working together for months to slowly, manually build up datasets that can be used to underpin analysis and policy recommendations. These datasets don’t yet exist in the open-source environment (and often, we’re told, can’t be found in classified environments, either). Complicated datasets are hard to build, and even harder to assess, requiring the right combination of specialists who possess language skills and deep country and/or thematic expertise. Few advanced machines or artificial intelligence systems have proven useful here. While investments in new technologies are essential, this doesn’t diminish the importance of individuals. As Chen points out, “Having the right human in the driver’s seat is more important than ever.”
But intelligence communities need to invest in more than “seasoned analysts”. I’d argue that it’s equally vital to invest in the unseasoned (and great things can come from pairing the two). To draw on another example from my day job, some of the institute’s most-read and impactful reports and blog posts are written by three employees – all with very different backgrounds and a combined age of less than seventy-five. These three analysts are many things, but “seasoned” they are not!
Rogers makes some interesting points, especially on data privacy and ethics. The Australian public should expect, and demand, that ethics be infused into all facets of how their government engages the world – across diplomacy, defence and intelligence. Some intelligence communities, for example, now employ ethics advisers to provide guidance on different operational aspects to their work. One can see such advisers and debates about ethics becoming more central in today’s intelligence agencies.
But for the most part, Rogers misunderstands my essay. It is not about the targeting of Australians and the collection of their data by our government or the private sector – as deserving as these topics may be of further attention. Rather, my piece explores how governments around the world engage in foreign intelligence collection, and in particular how cyberspace, the proliferation of data and emerging technologies are affecting that collection, and the work of the analysts who write it up.
I certainly don’t believe and did not say, as Rogers implies, that it should be “business as usual” to collect the private data of Australians during a crisis. But I do believe that open-source intelligence collection, which all governments engage in, can provide useful and timely insights. It’s worth noting that academics and journalists also engage in open-source intelligence collection. It just has a different name: research and investigation. My original point, about the analysis of Chinese social media data during the early days of what became the COVID-19 pandemic, was based on work published by Xi’an Jiaotong University’s Department of Infectious Diseases. And it is universities, across the world, that have built, and continue to host, some of the best Chinese social media analytical tools.
Seebeck quotes US intelligence analyst Zachery Tyson Brown, who said that “consumers of intelligence are drowning in data, but thirsting for insight”. It’s also important to note that this thirst exists outside the intelligence community. Never has the world had access to so much information.
But have we invested enough, and in the right ways, to ensure we have sufficient people analysing this information in useful ways? I’d argue no, given that so much of the research produced by countries emerges from universities and is siphoned off into the world of academic publishing. Journal articles and monographs aren’t especially accessible to the general public; nor are they always useful to those outside of narrow academic fields. Cordoning new and valuable knowledge remains an enormous opportunity cost. Seebeck’s argument that “informed intelligence is needed more than ever in a disruptive and increasingly contested global environment” can be extended. Informed analysis – whether classified or unclassified – is needed more than ever.
My essay is thoroughly referenced, though the journal does not publish these references in its pages, and the overwhelming majority of my sources were from the United States – investigative journalists, think-tankers, academics and material from the intelligence agencies themselves. This served as a reminder about how little public discussion, analysis and reporting there is in Australia on intelligence issues. If it wasn’t for the plethora of public reporting and official commentary continuously emerging from the United States, and the sprinkling from the United Kingdom (where spy chiefs increasingly give media interviews outlining their priorities), there would have been very little for me to write.
But things are changing. Some Australian spy chiefs, and their agencies, are slowly coming out of the shadows, led by Mike Burgess, former director-general of the Australian Signals Directorate (ASD) and now head of ASIO, who has given a number of public speeches and oversaw the launch of Twitter accounts at both agencies. In September 2020, the new director-general of the ASD, Rachel Noble, gave her first major public speech, at the Australian National University. Days before, her US counterpart, Paul M. Nakasone, commander of US Cyber Command and director of the National Security Agency, co-authored an essay in US journal Foreign Affairs titled “How to Compete in Cyberspace”.
There is a gap in Australian public discourse when it comes to intelligence issues, and Australia would benefit from more journalists, think-tankers and academics reporting on, writing about and debating these issues. And perhaps Australian Foreign Affairs could pull together some of our spy chiefs for a special edition that would help fill this void.
Danielle Cave is deputy director of the International Cyber Policy Centre at the Australian Strategic Policy Institute.