Bletchley Park, a former codebreaking spy base near London, where the first international AI Safety Summit was held. Delegates from 28 nations, including the U.S. and China, came together to address the potential risks associated with the rapid advancements in artificial intelligence. The summit focused on cutting-edge “frontier” AI that some scientists warn could pose existential threats to humanity.

The British Prime Minister, Rishi Sunak, hailed the declaration at the summit as a "landmark achievement" that highlights the urgency in understanding the risks of AI for the long-term future of our children and grandchildren. However, U.S. Vice President Kamala Harris urged countries to go further and faster in addressing the risks of AI. She stressed that the world needs to act now to address not only existential threats but also other risks that are causing harm and feel existential to many individuals, such as faulty AI algorithms and deep fake photos used for abusive purposes.

The summit was a significant initiative for Sunak, who aims to make the U.K. a hub for computing innovation. He framed the summit as the beginning of a global conversation on the safe development of AI. Harris was scheduled to attend the summit the following day, joining government officials from over two dozen countries, including Canada, France, Germany, India, Japan, Saudi Arabia, and China, despite protests from some members of Sunak's Conservative Party.

Although the agreement, named the Bletchley Declaration, lacked specific details and regulatory proposals for AI development, it was considered an achievement to have 28 nations pledge to work towards shared agreement and responsibility regarding AI risks. The countries committed to holding a series of further meetings, with South Korea planning to host a mini virtual AI summit in six months, followed by an in-person one in France a year from now.

During the summit, China's Vice Minister of Science and Technology, Wu Zhaohui, emphasized the uncertainty, lack of transparency, and emerging complexity of AI technology. He called for global collaboration and the sharing of knowledge to make AI technologies available to the public under open source terms. Tesla CEO Elon Musk was also expected to discuss AI with Sunak, as he had previously signed a statement expressing concerns about the perils AI poses to humanity.

Prominent figures attending the meeting included European Commission President Ursula von der Leyen, United Nations Secretary-General Antonio Guterres, executives from U.S. artificial intelligence companies like Anthropic, Google's DeepMind, and OpenAI, as well as influential computer scientists such as Yoshua Bengio. The closed-door meeting format facilitated healthy debate, with informal networking sessions building trust among attendees.

Mustafa Suleyman, CEO of Inflection AI, highlighted the significant disagreements between countries on open source AI systems. While such systems allow for quick problem discovery and resolution, they also raise concerns about malicious use. Bengio, one of the “godfathers” of AI, pointed out the incompatibility between open source and security, questioning how to address this issue.

Regarding regulation, Sunak stated that only governments, not companies, can ensure people's safety from AI's dangers. However, he cautioned against rushing into regulation and emphasized the need for a full understanding of AI technology. In contrast, Harris emphasized the importance of addressing current societal harms caused by AI, such as bias, discrimination, and the spread of misinformation. She highlighted President Joe Biden's executive order on AI safeguards as evidence of the U.S. leading by example in developing rules that prioritize the public interest.

The news item highlighted the ongoing discussions at the summit and the differing perspectives on AI risks and regulation. It underscored the global collaboration needed to address these risks while balancing the potential benefits of AI technology.