Bank of America Fixes Security Flaws Identified by AI

Large-Scale Efforts to Strengthen Cybersecurity

Bank of America, one of the world's largest financial institutions, has launched a major operation to secure its digital infrastructure. The bank's leadership announced that approximately 4,000 personnel have been mobilized to resolve software vulnerabilities identified by modern artificial intelligence models, including Anthropic's Mythos. To achieve this, the bank hired 1,000 external contractors who have been working alongside 3,000 of the bank's internal staff since the second quarter.

CEO Brian Moynihan explained that the bank spent several hundred million dollars in the third quarter on software updates and ensuring compliance with anti-money laundering regulations. While this specific funding was not allocated exclusively for fixing these flaws, the investment highlights the seriousness with which the financial sector is addressing threats identified by AI.

Artificial Intelligence as a New Security Tool

In April of this year, Anthropic introduced Mythos, an AI model specifically designed for security audits, which was provided to select partners under the Project Glasswing initiative. The model identified thousands of serious security vulnerabilities. This was followed in May by a similar project from OpenAI called Daybreak. Initially, many viewed these initiatives with skepticism, dismissing them as marketing ploys; however, real-world cyber incidents, such as the hack of Hugging Face, have shifted public opinion.

Independent evaluations have confirmed that new-generation AI models are significantly more effective at finding weaknesses in code than traditional scanning systems. Bank of America has become one of the first companies to publicly disclose the resources required to address problems identified by these new technologies.

What Awaits the Financial Sector?

According to Brian Moynihan, the bank expects the costs associated with fixing software vulnerabilities to decrease over time. This is logical, as the initial phase requires paying down a large amount of accumulated technical debt, which will lead to more stable system performance in the future.

For banks and technology companies in Georgia, this example serves as an important lesson. As the financial sector digitizes, cybersecurity standards are evolving. Using artificial intelligence not only for attacks but for proactive system audits will soon become standard practice. Companies that fail to adopt these technologies will struggle to cope with modern cyber threats.