Bank of America has mobilized a workforce of 4,000 people, including 1,000 outside contractors, to address software vulnerabilities uncovered by artificial intelligence models such as Anthropic’s Mythos. According to reporting by The Decoder, the massive remediation effort highlights the dual-edged nature of advanced code analysis tools in corporate environments.
The Scale of AI-Driven Code Review
The involvement of thousands of personnel demonstrates the sheer volume of output generated when AI models scan legacy and production codebase systems. While models like Anthropic’s Mythos are capable of accelerating vulnerability discovery at unprecedented speeds, they simultaneously create substantial human workloads as internal teams and hired contractors must manually review, verify, and patch the flagged security holes.
What Remains Unclear
The public reports do not specify the timeline for this remediation project, nor do they detail the exact financial cost of hiring the 1,000 outside contractors. Additionally, the announcement did not specify which other AI models besides Anthropic’s Mythos were utilized in uncovering the vulnerabilities, or how many total flaws the systems identified.
Observers will be watching to see how other major financial institutions manage the human resource bottlenecks that accompany automated security scanning.
Lead image: Calvin Seng / Pexels
