The Global AI Safety Summit: Notable Takeaways

• Agreement that, to realise the opportunities of AI, there must be an international approach to increase our understanding of the technology and how we can mitigate risks.
• A shared responsibility for AI development where transparency and accountability are of increased importance.
• The establishment of the AI Safety Institute – a UK-based but internationally facing expert panel backed by industry and many nation states. This will evaluate frontier systems, carry out advanced research on AI safety and increase the level of information sharing between governments, the private sector and civil society. A similar Institute was recently announced in the US as part of President Biden’s Executive Order on AI.
• A continuance in dialogue, with another plan to come together twice in 2024 to discuss the careful and beneficial use of AI.

We must look ahead at what is next. Adarga looks forward to continuing these conversations in the UK, US, Australia, and other allied countries. These agreements are a welcome step forward in the international journey towards regulating AI and follow the UK’s AI regulation white paper, which correctly identified a context-driven, regulator-led approach as being suitable to encourage development and adoption of AI.

Sarah Beddoes, Adarga’s General Counsel and Co-Chair of Adarga’s Committee for Responsible AI commented: “Adarga welcomes further conversation on AI policy and the establishment of the AI Safety Institute, dedicated to driving forward frictionless innovation. Adarga’s Committee for Responsible AI recognises the need for an approach centred around the responsible creation, good operation, and human-centred governance and oversight of AI. Going forwards, it is crucial that governments strike the correct balance between safety and innovation, to ensure faster adoption and deployment of AI and to allow us to realise the benefits of this technological revolution.”

Further AI safety takeaways from Adarga

Adarga's Dr. David Green detailed three focus areas to enhance AI safety in the world of generative AI earlier this month. Company CEO, Rob Bassett Cross, gave his thoughts following the Global AI Safety Summit, explaining that to maintain public trust, there must be reassurance on existing, as well as future risks. He also explained the need to provide SMEs a voice in all AI debates, given that companies within this vertical represent by far the largest community of innovators, entrepreneurs, data scientists, developers and builders. SMEs are also responsible for developing and deploying some of the most exciting, new AI applications. Given their market presence, those companies will be equally responsible for ensuring AI safety as the large global companies.