Speech delivered by Michelle Donelan on day 2 of the AI Seoul Summit, on Wednesday 22 May.

The Rt Hon Michelle Donelan MP

Good afternoon.

I’d like to once again by expressing my gratitude to our co-hosts the Republic of Korea and for Minister Lee, for bringing us together at this pivotal event.

Your dedication and leadership has made this gathering possible- just six months after we met at Bletchley Park last November.

It is truly inspiring to see so many familiar faces again around the table – and to reflect on the extraordinary progress we have made since the last meeting.

We have a joint mission to grip the risks so that we can together seize the incredible opportunities across our public services, our businesses and our societies at large.

When I spoke at the UK’s AI Safety Summit, I shared the story of how governments swiftly collaborated to ratify the Montreal Protocol, and tackle the existential threat of the hole in the ozone layer. I hoped that this example would serve as the benchmark for what we could achieve together through these Summits. And I’m happy to see that we have acted even faster for this global challenge of AI Safety.

At Bletchley, we announced the UK’s AI Safety Institute – the world’s first government-backed organisation dedicated to advanced AI safety for the public good.

Since then, the Institute has made remarkable strides using state-of-the-art tools to conduct pre and post deployment testing on advanced frontier models.

Yesterday, leaders agreed to bring these Institutes into a global network, showcasing the ‘Bletchley effect’ in action.

Last night marked a historic moment as leading companies from around the globe signed the Frontier AI Safety Commitments, pledging to improve AI safety and refraining from releasing new models if the risks are too high.

Additionally, last week, we published the Interim International AI Safety Report on the safety of advanced AI, bringing together the latest scientific research to inform robust, evidence-based policymaking.

Together, we achieved all of this in less than 6 months, demonstrating the power of global collaboration and action-orientated focus.

However, my message today is that we should not rest on our laurels. As the pace of AI development accelerates, we must match that speed with our own effort and action if we are to grip the risks and seize the opportunities.

In phase 2 from Seoul to France, we will push the boundaries of nascent science of frontier AI testing and evaluation. Simultaneously we must turn our attention to risk mitigation outside these models. Ensuring that society as a whole becomes resilient to the risks posed by AI.

This is where the concept of systemic safety really comes into play, embedding AI safety mechanisms into society’s systems, not just AI systems themselves.

The UK has already begun investing in systemic safety research, and we are eager to deepen our global collaboration on this critical task.

As Isaac Asimov said in 1988, the year in fact that Seoul hosted the Olympics “Science gathers knowledge faster than society gathers wisdom.”

It is our responsibility to ensure that human wisdom keeps pace with human knowledge as we navigate the most complex technology humanity has ever produced.

I look forward to engaging in detailed discussions and agreeing on concrete, tangible actions that will propel us into the next phase of global AI safety policy.

International collaboration between AI safety institutes and their equivalents will be absolutely mission critical to our shared success, and today’s showcase provides an opportunity to learn from one another and lay the very foundations for sustained and productive collaboration.

So with that, I would like to turn back to you Minister Lee.

Published 22 May 2024