The UK government's AI Safety Institute will open its first overseas office in San Francisco this summer, in a move confirmed by Technology Secretary Michelle Donelan.
As announced in a press release, the first foreign office of the AI Safety Institute (AISI) will recruit an initial team of technical staff led by a research director, complementing the Institute's growing London headquarters, which already houses more than 30 experts.
The strategic expansion aims to benefit from technology talent in the San Francisco Bay Area, where many of the world's leading technology and artificial intelligence companies are based.
British AI Safety Institute to go global
The London office will continue to expand risk assessments of advanced artificial intelligence systems, while the new San Francisco office will facilitate close collaboration between the two nations.
Both the United Kingdom and the United States have committed to signing similar AI safety agreements, including signing the Bletchley Declaration at a summit hosted by Britain along with 25 other countries and the European Union.
Donelan commented: “[The expansion] It is a pivotal moment in the UK's ability to study both the risks and potential of AI from a global perspective, strengthening our partnership with the US and paving the way for other countries to draw on our expertise as we continue to lead the way. world in AI security. .”
The UK and Canada have also committed to a partnership aimed at improving research into AI safety.
The announcement coincides with the publication of recent results of AI security tests conducted by AISI. In addition to highlighting some of the technical limitations of large language models, the study also noted that “All models tested remain highly vulnerable to basic 'jailbreaks', and some will produce harmful results even without dedicated attempts to bypass safeguards.”
Ian Hogarth, President of AISI, commented: “AI safety is still a very young and emerging field… Our ambition is to continue to push the frontier of this field by developing cutting-edge assessments, with an emphasis on risk. related to national security. .”
The announcement also precedes the 2024 Seoul AI Summit, which is seen as a successor of sorts to the British Bletchley Park summit held in November 2023. International governments, AI companies, academia and the civil society come together to continue discussions on AI safety. .