British banks will be given access in the next week to a powerful AI tool that was deemed too dangerous to be released to the public, as a series of senior finance figures warned over its impact.<\/p>\n
Anthropic, which has so far limited the release<\/a> of the new model to a small clutch of primarily US businesses, including Amazon, Apple and Microsoft, said it would expand that to UK financial institutions.<\/p>\n \u201cThat is in the very near term, in the next week,\u201d Pip White, Anthropic\u2019s head of UK, Ireland and northern Europe operations, said in a Bloomberg TV interview. \u201cAs you would expect, the engagement I have had from UK CEOs in the last week has been significant.\u201d<\/p>\n Anthropic, which is the company behind the Claude family of AI tools, has said that its latest model, Mythos, poses an unprecedented risk because of its ability to expose flaws in IT systems.<\/p>\n \u201cAI models have reached a level of coding capability where they can surpass all but the most skilled humans at finding and exploiting software vulnerabilities,\u201d Anthropic said in a blogpost earlier this month. \u201cThe fallout \u2013 for economies, public safety, and national security \u2013 could be severe.\u201d<\/p>\n Finance ministers, executives and regulators have discussed the potential threats as they gathered in Washington this week for the IMF and World Bank spring meetings<\/a>, while also handling concerns over the global ramifications spilling over from the US-Israeli war with Iran.<\/p>\n The Canadian finance minister, Fran\u00e7ois-Philippe Champagne, told the BBC: \u201cCertainly it is serious enough to warrant the attention of all the finance ministers \u2026 The difference with the strait of Hormuz<\/a> is that we know where it is and we know how large it is.<\/p>\n \u201cThe issue that we\u2019re facing with Anthropic is that it\u2019s an unknown unknown. It requires a lot of attention so that we have safeguards, and we have processes in place to make sure that we ensure the resiliency of our financial system.\u201d<\/p>\n Andrew Bailey, the governor of the Bank of England who also chairs the Financial Stability Board of global regulators, said: \u201cIt is a very serious challenge for all of us. It reminds us how fast the AI world moves.\u201d<\/p>\n However, he said regulators were having to consider whether, and how hard, to clamp down on the technology, as governments seek to reap AI\u2019s economic rewards. \u201cWhat is the optimum moment to frame the rules of the road?\u201d Bailey asked. \u201cIf you go too early you a) risk missing the target and b) you risk distorting the evolution, and if you go too late things can get out of control.\u201d<\/p>\n The European Central Bank\u2019s president, Christine Lagarde, said: \u201cThe development we\u2019ve seen with Anthropic and Mythos is a good example of a responsible company that is suddenly thinking: \u2018Ah, that could be really good\u2019 \u2013 but if it falls in the wrong hands, it could be really bad.<\/p>\n \u201cEverybody is keen to have a framework within which to operate,\u201d Lagarde told Bloomberg TV. But she added: \u201cI don\u2019t think there is a governance framework that is there to actually mind those things. We need to work on that.\u201d<\/p>\n