Rishi Sunak Confirms the AI Safety Institute at Bletchley but Postpones Regulations for Another Day

The U.K. established a Frontier AI taskforce in June to get ready for the AI Safety Summit happening this week. It is set to become a permanent part of the U.K.’s efforts to lead in AI policy. Today, U.K. Prime Minister Rishi Sunak officially introduced the AI Safety Institute, a “global hub based in the U.K.” responsible for testing the safety of emerging AI technologies.

Rishi Sunak Confirms the AI Safety Institute at Bletchley
Rishi Sunak Confirms the AI Safety Institute at Bletchley

The institute was informally announced last week ahead of this week’s summit. Now, the government has confirmed that Ian Hogarth, an investor, founder, and engineer who also chaired the taskforce, will lead it. Additionally, Yoshio Bengio, a prominent figure in AI, will be in charge of producing its first report.

Rishi Sunak Confirms the AI Safety Institute at Bletchley

The amount of funding the government will allocate to the AI Safety Institute is unclear, and it’s uncertain if industry players will be expected to contribute. The institute, situated under the Department of Science, Innovation, and Technology, is described as “backed by leading AI companies,” although this likely refers to endorsement rather than financial support. We’ve reached out to the DSIT for further information and will update accordingly.

The exact amount of government funding for the AI Safety Institute and whether industry players will be expected to contribute remains uncertain. While the institute falls under the Department of Science, Innovation, and Technology, it’s described as being “backed by leading AI companies,” which may primarily refer to endorsement rather than financial support. We’ve contacted the DSIT for clarification and will provide updates as we receive more information.

This news coincides with yesterday’s announcement of the Bletchley Declaration, a new agreement signed by all the countries participating in the summit. It commits them to collaborative testing and other obligations related to assessing the risks of “frontier AI” technologies, such as large language models.

Up until now, only the companies developing new AI models have been conducting safety tests,” Sunak stated during an evening meeting with journalists. He also mentioned the work being done by other countries, the UN, and the G7 to address AI issues. The new plan is to “collaboratively test the safety of new AI models before their release.

Key Insights and Absences at the AI Summit

However, it’s essential to note that all of this is still in its early stages. The U.K. has been hesitant to explore AI technology regulations, both at the platform level and in specific applications. Some argue that without effective enforcement, the concepts of safety and risk assessment may lack substance.

Sunak argued that it’s premature to impose regulations.

“In response to the accusation of being too lenient on legislation while emphasizing big ideas, Sunak explained, ‘The technology is evolving so rapidly that governments must ensure we can keep pace. Before imposing mandates and legislation, you need a clear understanding of what you’re regulating.

Although transparency is a significant goal in the long-term efforts related to this rapidly evolving technology landscape, today’s meetings at Bletchley, on the second day of the summit, did not align with that principle.

In addition to one-on-one meetings with European Commission President Ursula von der Leyen and United Nations Secretary-General António Guterres, today’s summit featured two main sessions. These sessions were closed to journalists, with only select groups present to observe the gatherings. Attendees included CEOs from DeepMind, OpenAI, Antrhopic, InflectionAI, Salesforce, and Mistral, as well as the President of Microsoft and the leader of AWS. Government representatives included Sunak, U.S. Vice President Kamala Harris, Giorgia Meloni of Italy, and French Finance Minister Bruno Le Maire.

It’s worth noting that despite being prominently mentioned on the first day, China did not participate in the closed plenary sessions on the second day.

Another notable absence from today’s sessions appears to be Elon Musk, the owner of X.ai (formerly known as Twitter). Sunak is scheduled to have a fireside chat with him this evening on Musk’s social platform. Interestingly, this chat is not anticipated to be broadcast live.

Check These Out


Please enter your comment!
Please enter your name here