The UK government launches an AI Safety Institute with $100 million in funding to evaluate and mitigate risks from advanced AI systems.
The UK government has established an AI Safety Institute with $100 million in initial funding, positioning Britain as a global leader in AI safety research and governance. The institute will evaluate advanced AI systems for potential risks and develop mitigation strategies.
The institute will work with AI developers to conduct pre-deployment safety evaluations, focusing on systems that could pose risks to society if improperly designed. It will also research alignment techniques to ensure AI systems remain beneficial as they become more capable.
"AI is the most transformative technology of our time, and the UK intends to lead in making it safe," said the UK Prime Minister. "This institute will ensure we understand the risks and can respond appropriately."
The institute will collaborate with international partners including the US AI Safety Institute and the EU's AI Office. It will also advise the UK government on AI regulation and policy, feeding into the country's broader regulatory framework.
Academic institutions across the UK have expressed strong support, with several announcing new AI safety research programs in partnership with the institute.
