Bank of America moved quickly earlier this year to shore up a wide range of software after cutting edge AI models started revealing weaknesses, CEO Brian Moynihan told CNBC on Wednesday after Q3 earnings. He cited Anthropic's Mythos as one of the tools involved. Mythos was made available in April to a small set of banks, technology companies and government agencies.
To handle the workload, Moynihan said the bank "put on 1,000 contractors to help us do the patching work in the second quarter, and those people remain here today, helping us, on top of the 3,000 people we have in that area." He told analysts the bank spent a "couple hundred million plus" in the third quarter on software fixes and regulatory matters related to anti-money laundering compliance.
Moynihan described the risk from frontier models as urgent, saying, "A lot of us sat down and said, 'How do we figure out how to get through this quickly?'" His remarks are among the first from a large U.S. company to detail both headcount and dollars devoted to protecting systems from Mythos. He also pointed to AI's mixed impact: it can lift productivity, but the same tools can dig up decades-old flaws that force costly cleanup. "We've got to stay ahead of them, and we've got to be mindful that the people doing the software development are the people trying to sell us the services," he said. With shares down as much as 8% on Wednesday, he emphasized the spike in spending should be short lived: "Once you patch, you don't have to patch as much. … The pipeline moves through."
AI is not only a defensive line item at BofA. Inside the bank, roughly 130 homegrown AI initiatives are underway. Moynihan told analysts that anticipated savings are set to rise to $700 million in 2027, up from $300 million this year.
When technology and finance intersect, it helps to stay informed, so subscribe to the free Market Briefs for daily context
Head count fell by roughly 3,400 over the past year and about 5% of managers were cut, though how much of that ties to AI is unclear. Translation for your wallet: AI can lower costs and raise efficiency, but new models can also trigger sudden spend to fix old code, and that tug of war can show up quickly in a stock price.
Because small developments can have broader implications, get a clear daily summary from Market Briefs, our free newsletter in five minutes
