With Big Tech's ever-expanding presence in the State, we need new analytical frameworks to understand its impact and future direction. Antonio Weiss and Joel Lindop suggest two concepts can help: Hobbes’ Leviathan and the AI ‘singularity’.
Recent American and British elections have placed an unprecedented focus on the role for technology to improve the state. Elon Musk and Vivek Ramaswamy, leading a Department of Government Efficiency, was the first (of many) eye-catching Trump appointments, while Poppy Gustafsson, founder of the cybersecurity firm Darktrace, was appointed UK Minister for Investment in October 2024.
Prime Minister Starmer said at the International Investment Summit 2024 in October, that AI promises to be “a game changer when it comes to the delivery of public services”. Musk and Ramaswamy’s comments regarding the American state have been more explosive, promising “mass headcount reductions across the federal bureaucracy”.
The US and UK have previously embraced technology, with Bill Clinton’s administration pioneering federal internet use and David Cameron’s Government Digital Service (GDS) streamlining government platforms. Both countries are now challenged by state budget deficits expanding to record peacetime levels. As technology offers previously unimagined answers for how better public services can be delivered, how the challenge and opportunity are navigated will be pivotal to our future.
The singularity as an opportunity
Coined by the technologist John von Neumann in the mid-twentieth century, and elaborated by many others since, the term (technology) ‘singularity’ refers to the concept of exponential acceleration of technology, including AI, in some senses surpassing human intelligence. It sounds extreme but in our view, the singularity is a practical and useful concept today.
The most familiar example of a (physical) singularity is a black hole, whose immense density warps spacetime to such an extent that even light cannot escape. However, we propose a different type of physical singularity as a more practical analogy: the acceleration of flight beyond the sound barrier. At this threshold, the laws governing aerodynamics shift, requiring entirely new designs like the swept-wing supersonic aircraft.
When software engineers have become the frontline of technology development across diverse sectors and their own productivity is hugely enhanced by generative AI, we are in a period of significant technological acceleration. And with AI chatbots now sufficiently comparable to human interactions that they are now capable of inciting major crimes, we have undoubtedly long-passed the Turing test.
The characteristics of technological acceleration today mean we will require new patterns of government, like those new types of supersonic aircraft wings. In the past, emerging technologies were adopted progressively and absorbed over time, becoming gradually familiar to the state and society. Now, technology available for practical use next year may not yet even be science fiction today. The risk of harm from surprising consequences of rapid advances outside our understanding is real.
How Leviathan meets the singularity
Famously, Thomas Hobbes described the state as a Leviathan, a giant and complex creature that people accept as an authority to protect them from existence being “nasty, brutish and short”. The Leviathan creates conditions that serve the common good, which includes imposing constraints. The relationship of other influences to Leviathan is important, such as the churches in Hobbes’ time, and technology in ours. Personifying the state as a Leviathan helps us reflect on its ability to embody decision-making and deliver judgments to produce positive outcomes for society as a whole.
The incentives of technology firms of course do not align with those of the common good. Inherently, enterprises aim to create wealth and power for their owners, who are often their founders and leaders. They may sometimes also be driven by social missions. But if social outcomes conflict with profits, the latter will win. The big technology companies involve concentrations of the brightest minds when it comes to solving technical problems. But there is no reason to suppose these same people are the most motivated towards solving the problems of government. Technology can potentially be a means of achieving ends or outcomes that the state chooses. However, as Marietje Schaake, author of The Tech Coup has warned, the state should allow technology to serve, but not let it take control. She argues: “We need to bring the same level of legal clarity, accountability mechanisms, and transparency measures to the digital realm that we expect around other innovations such as medicines, chemicals, foods, cars, or even processes such as the decision to engage in foreign conflict.”
Putting the singularity into public service
Simply making today’s services more efficient may save money, but it risks missing changes that could deliver far greater productivity gains and real improvements in services. Prioritising key demands of the electorate—such as better health, shorter waiting lists, less crime, reduced loneliness, and more entrepreneurship—should be the focus of the technology-enabled state.
To date, necessity rather than ambition has tended to be the mother of adoption in the public sector. The pandemic accelerated digital adoption, with the NHS App becoming a viable ‘patient passport‘ and the justice system shifting to electronic files, paving the way for broader technology integration.
Companies in the private sector have become familiar with tech disruption. Successful, enduring corporations are alert to the potential in technology, whether they operate in banking, retail, publishing or leisure. They know that if they don’t act on new technology applications, their market positions will be threatened.
The state is different, its goals and context more complex. Nevertheless, embracing technology is key to the government’s ability to improve lives and society. It will need to seek overtly radical ideas and test them out. The UK’s Government Digital Service has rightly placed great emphasis on ‘agile’ management: embedding testing into programmes of work, allowing course correction so that approaches are refined before new systems are rolled out in full.
The state needs to own the opportunities that come from technology, considering that the technology-enabled answer may look considerably different to previous approaches. The lights must be kept on in the services operating today, while at the same time radically different approaches for future services are tested. Enabled by technology, the government must contemplate how to change.
We propose a thought experiment that may help. Artificial General Intelligence (AGI) refers to the concept of superhuman intelligence of a truly general form able to find and communicate the best answer to any question. Arguably, AGI may technically never exist (although the trillion dollar investment prospectus for generative AI suggests it might); but conceptually, AGI could nevertheless point to approaches we can implement using the technology and people available to us today. In particular, AGI would not be constrained by the departmental and divisional structures of the past. Presumably superhuman intelligence would contemplate initiatives that span different portfolios such as health, crime and education. “What would AGI do?” is a useful question to pose when thinking about redesigning public services.
What radical opportunities are possible?
Radical opportunities can improve public services and lives, but success will require collaboration across academics, policymakers, and citizens. We invite your thoughts and will share follow-up proposals later in 2025.
Image: By Abraham Bosse – This file has been extracted from another file, Public Domain, https://commons.wikimedia.org/w/index.php?curid=154883426
The views and opinions expressed in this post are those of the author(s) and not necessarily those of the Bennett Institute for Public Policy.
