Bank of America expands software patching effort as AI flaws raise defense costs

Traders Union
All News
Financial News
BoA expands AI patching

Highlights

Bank of America hires 1,000 contractors in Q2 to patch software vulnerabilities identified by Anthropic's Mythos and other advanced AI models, supplementing 3,000 existing staff.
Moynihan reports spending a 'couple hundred million plus' in Q3 on software remediation and anti-money laundering compliance, contributing to an 8% slump in Bank of America shares Wednesday.
AI-driven efficiency drives expected annual savings from $300 million in 2023 to $700 million by 2027, while workforce shrank by 3,400 employees and 5% of managers over the past year.

Remediation push after Mythos exposure

As reported by CNBC, Chief Executive Brian Moynihan says Bank of America brings in 1,000 contractors during the second quarter to help patch software vulnerabilities identified by advanced AI models including Anthropic's Mythos.

Speaking after the bank's third-quarter earnings release, Moynihan says the company moves earlier this year to strengthen a broad range of internal and third-party code with support from thousands of employees and outside personnel. He says the contractors remain in place alongside the 3,000 people already working in that area.

Moynihan also tells analysts the bank spends a "couple hundred million plus" in the third quarter on software remediation and regulatory matters tied to anti-money laundering compliance. His comments offer a rare indication of the manpower and financial burden large U.S. companies face as AI systems become capable of identifying and exploiting long-standing software weaknesses.

Cost pressure and expected efficiency gains

Moynihan says the additional spending is temporary and should ease as the patching pipeline moves through, arguing that remediation costs decline over time once systems are fixed. The comments come as Bank of America's shares slump as much as 8% on Wednesday, adding pressure on management to explain the near-term expense increase.

At the same time, the bank says AI is also producing savings in other parts of the business. Bank of America has about 130 internal AI projects, and Moynihan tells analysts those efforts are expected to lift savings from $300 million this year to $700 million in 2027.

The lender's head count falls by about 3,400 employees over the past year as it cuts roughly 5% of managers, according to Moynihan. It remains unclear how much of that workforce reduction is directly linked to AI.

Our earlier coverage of major Wall Street banks’ AI push highlighted how lenders see the buildout as a significant financing and productivity opportunity, but also one clouded by uncertainty around credit quality and returns. We also noted that as banks deploy AI internally and support AI infrastructure investment, executives are increasingly flagging cybersecurity—especially more autonomous AI tools—as a growing source of vulnerability alongside the revenue upside.

This material may contain third-party opinions, none of the data and information on this webpage constitutes investment advice according to our Disclaimer. While we adhere to strict Editorial Integrity, this post may contain references to products from our partners.