Anthropic AI finds thousands of vulnerabilities, Fed excluded
โดย Tim Official: Kevin Warsh
So, Anthropic's AI, Claude Mythos Preview, found thousands of undocumented security holes in major OSes and browsers. Instead of dropping it publicly, they're giving early access to about 50 orgs via Project Glasswing to patch stuff up. Think JPMorgan Chase, Amazon, Apple, and Google got the memo. Wild thing is, the Federal Reserve, which runs Fedwire and sets banking security rules, is still locked out as of July 15. They even met with bank CEOs on April 7 to talk security, but still no access to the vulnerability list. Fed Chair Kevin Warsh is out here saying we gotta patch up as these new models get more common. Daniel Newman from Futurum Group is baffled they're not included, especially since Anthropic was already chatting with US gov folks. Nobody's really explaining why, but some guess it's federal procurement rules. Anthropic figures bad actors will catch up in six to twelve months, and that clock started in April. Gotta wonder what's up with that
คำบรรยาย (en)
The broad financial system has new vulnerabilities with these new technologies. I wouldn't want to just isolate mythos, though. The other large language models have rival products. The changes, the efficient frontier has changed even in the last month. But as these new models find their way more broadly, our banking system and, frankly, the Federal Reserve needs to do all we can to patch any vulnerabilities that we have. What are you doing to ensure that these midsize and smaller banking institutions have access to whatever AI tools appropriate as quickly as possible? Yeah, so I share the sentiment, Senator Reid. We are not the deciders as to who has access, but I have not been shy in sharing my views with authorities across the government about the vulnerabilities and have been asking for access, not just for the Federal Reserve, but for other institutions, to a whole range of these new artificial intelligence models so that they can protect themselves. if you said to me what worries you over the course of the next several months I want to make sure that we're as protected as we can be from foreign actors that want to do us harm
