Bank of America brought in 1,000 contractors in the second quarter to patch software vulnerabilities that new AI models can dig up, CEO Brian Moynihan told CNBC on Wednesday. They're still there, working alongside the roughly 3,000 people the bank already has in that area.
"That risk is real risk," Moynihan said, explaining that when frontier models like Anthropic's Mythos arrived, "a lot of us sat down and said, 'How do we figure out how to get through this quickly?'" Mythos went out in April to a small group of banks, tech companies and government agencies, and the bank's response covered both its own code and code from its vendors.
It isn't cheap. Moynihan told analysts the bank spent a "couple hundred million plus" in the third quarter on that software cleanup plus regulatory work tied to anti-money-laundering compliance. He argued the bill will shrink: "Once you patch, you don't have to patch as much." He also had a pointed line about the vendors, saying "the people doing the software development are the people trying to sell us the services."
This is one of the first times a big U.S. company has put a headcount and a dollar figure on defending against Mythos-class bug hunting. The bank is also spending on AI itself. It has about 130 internal AI projects and expects savings from them to grow from $300 million this year to $700 million in 2027. The shares fell as much as 8% on Wednesday.
— via CNBC