London and Washington are set to announce a “close collaboration” on AI safety, according to officials from both countries. The collaboration aims to combine the United States’ new regulations on AI development with the United Kingdom’s existing work on AI safety through the Frontier AI Taskforce.
The announcement of bilateral collaboration between the UK and the US safety institute will be made during the AI safety summit at Bletchley Park, as stated by a UK official. The US Vice President, Kamala Harris, will also play a crucial role by announcing the creation of the United States AI Safety Institute, which will be based at the Department of Commerce.
The main objective of the US institute is to establish guidelines, standards, and best practices for evaluating and mitigating risks associated with AI. This includes catastrophic risks and societal harms like bias and misinformation. On the other hand, British Prime Minister Rishi Sunak plans to establish the AI Safety Institute in the UK, which will focus on evaluating and testing new types of AI.
To bolster their efforts in AI safety, the UK’s institute will build upon the work already done by the Frontier AI Taskforce. They have been granted access to technology models from leading AI companies such as Google DeepMind and OpenAI. This collaboration will not only involve information sharing and research collaboration, but both countries will also make separate announcements about their respective institutes.
The collaboration comes as the White House recently signed an executive order requiring developers of powerful AI systems to share safety test results with the US government before making them public. The UK’s Tech Secretary, Michelle Donelan, believes that it is easier for the US to lead the industry in transparency due to the dominance of American firms.
Due to the existing organization of the Frontier AI Taskforce and funding already in place, the UK aims to move quickly in creating its AI safety institute. They plan to share some AI safety information publicly, but will reserve more sensitive national security intelligence for a select group of like-minded governments.
The collaboration between the UK and the US on AI safety is a significant step towards ensuring the responsible and ethical development of AI technology. With both countries joining forces and leveraging their expertise, it is hoped that the risks associated with AI will be effectively addressed and mitigated, leading to a safer and more beneficial use of AI in various industries and sectors.
