Benedict Macon-Cooney, Chief Strategist, Tony Blair Institute for Global Change
State of Open: The UK in 2023
Phase Two “Show us the Money”
Part 2: “AI Openness”
Benedict underscores the transformative role of AI in the UK’s technological landscape. Acknowledging existing strengths in AI, he proposes an ambitious plan, including a tenfold increase in compute capacity, innovative laboratories, and a procurement agency to enhance the nation’s AI capabilities. Emphasising the need for agile regulation in the face of rapid technological advancements, the passage advocates for simultaneous research and regulatory efforts. It stresses the importance of the UK taking a clear stance on the Open Source Community to avoid constraining AI to a few entities. Benedict suggests providing cloud access to API models as a means of supporting Open Source Software while mitigating misuse risks. Ultimately, he underscores the critical role of AI in shaping the future of economic progress and growth, calling for leadership in building a diverse and open AI industry.
Thought Leadership: A New National Purpose
Benedict Macon-Cooney, Chief Strategist, Tony Blair Institute for Global Change
We are in a new era. The progress in AI will almost certainly be the most important technological development of our lifetime. It will help with relatively simple things such as process automation while opening up profound possibilities for discovery in biology and material science, as well as for industries such as energy that are essential to our physical world.
As we set out in our recent paper, “A New National Purpose”, the pace at which AI Is developing presents both significant opportunities and risks for the UK. The country has some strength in depth in this technology, with companies such as DeepMind, Faculty and Exscientia contributing to accelerate nuclear fusion science, precision medicine and improve the machinery of government. Through the AI Taskforce and appointment of Ian Hogarth, the UK government has also signalled its intent to bring deep expertise into decision-making.
But the scale of ambition needs to step up if the UK is to maintain an edge in the critical strategic technology of the 21st Century. The plan we put forward includes how to build AI-era infrastructure, improve talent, encourage innovation as well as how AI can change the shape and role of the state today. Our ideas included increasing compute capacity tenfold, new Disruptive Innovation Laboratories and an Advanced Procurement Agency so that the government can stimulate the UK ecosystem.
Given the pace of change and the uncertainty of the situation, we also believe it is too soon to lay out a definitive plan of action for regulation of AI. Instead, through a new national effort focused on researching and testing safe AI, we believe that research and regulation should proceed in tandem and in an agile way.
If the UK is to be a global leader in AI, it is also important that the UK takes a clear position on the Open Source Community. The software that has been built by this community underpins the modern internet, from infrastructure to operating systems to algorithms. As a result of recent progress, Open Source is again enabling cutting-edge development in AI. Some of this will be driven by the reduction in compute costs, while the ability to run advanced models such as Stable Diffusion on a home computer puts powerful tools in the hands of artists, creators and small businesses. It will be integral to the democratisation of AI.
Meanwhile, Meta’s LLaMA allowed Stanford University to build a GPT-3-like model of their own for just $600. The Emirate of Abu Dhabi has also made its AI model, Falcon 40B, available Open Source for research and commercial use.
Such availability is important for students, researchers and builders around the world. “Open Source AI” speeds adoption, increases fair and trustworthy AI and advances the sciences that use AI too. But openness does also present risks, as it also provides bad actors with potential powerful tools.
Some have taken pre-emptive steps to react to risks that are yet to unfold. The European Union, which tends to take a position of regulator of resort, has decided to weaken the position of the Open Source community through the AI Act. This would be a mistake and the UK should not follow suit. Instead, we should use the opportunity to offer a different model.
Providing cloud access to frontier API models would be one way of supporting Open Source Software in a way that mitigates the risk of misuse, and one we argued for in a New National Purpose. But there are other approaches too, such as the one pursued by the French National Centre for Scientific Research.
The Taskforce’s role on this will also be critical.
And just as building in the open provides opportunities, building Open Source safety would confer significant benefits to the AI community and wider world.
The future of economic progress and growth is going to be dependent on AI.
The UK has to show leadership in building its industry, including Open Source. This requires taking a clear position on the value that openness brings. There is a risk that this technology is confined to the hands of just a few actors, with the potential dividends being too narrow.
This has happened before with technology.
It cannot happen with the most transformative technology that exists today.