Policy paper

AI Safety Summit 2023: Roundtable Chairs' Summaries, 2 November

Published 3 November 2023

The discussions that took place at the AI Safety Summit at Bletchley Park on 1-2 November 2023 brought together international stakeholders from governments, leading AI companies, civil society and academia.

The following points, discussed with the Chairs of each roundtable, summarise the key discussion points, themes and emerging issues from the discussions on the second day of the Summit.

The UK is publishing this as Chair of the Summit, not as a UK government policy document.

Roundtable: Priorities for international attention on AI over the next 5 years to 2028

Chair: Oliver Dowden, Deputy Prime Minister, Government of the United Kingdom

A strategic discussion of the next 5 years to 2028 with a focus on the priorities for international collaboration, the key choices and challenges.

• It is vital to acknowledge the enormous opportunities AI has the potential to bring. Addressing risks will help unlock the opportunities in a way that all communities can benefit.
• In particular, innovative use of AI has huge potential to create new ways to address challenges. It is already having tangible impacts on everything from facilitating access to public services, to medical advancements, to enabling faster and more effective planning and response to climate emergencies.
• Building trust and understanding is crucial for unlocking opportunity and addressing risk. This requires reassurance that these technologies are being used ethically and responsibly.
• In order to unleash these opportunities, we need to invest in skills. This means supporting skills development for the wider public alongside enhancing technical capability within governments. Institutes like the AI Safety Institute – with academia and business working side-by-side - have an important role to play.
• A multistakeholder approach, bringing in academia and civil society, will enable us to reap the benefits that AI has to offer.

Roundtable: Creating actions and next steps for future collaboration

Chair: Michelle Donelan, Secretary of State for Science, Innovation and Technology, Government of the United Kingdom

A practical discussion of international collaboration following the Summit, including developing a shared understanding of model capabilities and safety risks, and risks from disinformation and deepfakes in the context of elections.

• Working quickly together to identify and evaluate the risks associated with frontier AI is of vital importance in mitigating the risks and reaping the benefits that AI can bring.
• It is critical that such an approach is inclusive – so that we do not leave anyone behind – and that it should draw on existing initiatives, such as the Global Partnership on AI, and the work within multilateral organisations in undertaking these tasks.
• The ‘State of the Science’ Report is an important first step and countries should nominate representatives onto the Expert Advisory Panel.
• Frontier AI poses significant risks to the integrity of the information environment and to political processes, lowering the bar to entry for disinformation actors whilst increasing the volume and realism of disinformation content.
• Some ministers noted concerns about the ability for AI-powered disinformation and deepfakes to upcoming impact elections.
• With these risks already materialising, and a high number of elections across the world scheduled for the next year, this is an area which requires rapid collaboration and action, with such action only being successful if it involves a coordinated effort between governments and industry.

Roundtable: Creating actions and next steps for future collaboration

Chairs: James Cleverly, Foreign Secretary, Government of the United Kingdom, and Michelle Donelan, Secretary of State for Science, Innovation and Technology, Government of the United Kingdom

A strategic discussion to explore where AI is creating the greatest opportunities now, and the specific areas most fruitful for further international collaboration.

• AI’s positive impact can already be seen across key areas of our lives. For example, AI is helping to:
  • develop better public services and better citizen experiences
  • tackle fraud, working closely with the financial services sector
  • support language translation applications, breaking down barriers and enabling global trade
  • transform our climate change adaptation and mitigation approach, including dealing with severe fire and flood risks
  • achieve universal healthcare coverage and support in the diagnosis of diseases such as cancer; and
  • support greater energy efficiency and cut costs.
• There are also exciting opportunities for rapid skills and educational programmes to be developed.
• AI can help accelerate progress towards achieving the Sustainable Development Goals. It can help us address poverty, predict and respond to humanitarian crises, and support inclusive growth.
• This is especially important given global trends such as population growth. AI can help improve the efficiency of food supply chains.
• It is essential that countries work together and learn from each other, collaboration is needed to ensure that all parts of the world realise the truly transformative potential of AI for Good.

See a summary of the discussions from day 1 of the Summit.