Bank of America CEO Brian Moynihan revealed that the firm deployed 1,000 external consultants over the last quarter to patch vulnerabilities in its codebases, specifically citing risks posed by sophisticated models like Anthropic’s Mythos. Citigroup’s Jane Fraser echoed this urgency, noting that the bank is actively running disruptive stress tests to anticipate potential AI-driven breaches. Meanwhile, Morgan Stanley CEO Ted Pick emphasized that the firm is prepared to spend whatever is necessary to safeguard client assets, arguing that institutions cannot afford to face garage-built AI threats with legacy security.
Wall Street CEOs scramble to defend against rogue AI
As Wall Street giants reported third-quarter earnings this week, a singular, non-financial threat dominated executive sessions: the prospect of rogue AI agents exploiting corporate software. From Bank of America to JPMorgan, CEOs signaled a shift toward aggressive defensive posturing against emerging models capable of identifying and infiltrating internal digital infrastructure.
Goldman Sachs CEO David Solomon and JPMorgan’s Jamie Dimon underscored the tension between fostering innovation and maintaining rigorous control. Dimon explicitly stated that he would block outside agents from entering the company's internal network to protect client data. These defensive maneuvers follow a string of high-profile incidents, including an OpenAI agent breaching Hugging Face’s systems and a separate exploit that compromised an Australian government website. With internal whistleblowers at major labs now warning that developers are gambling with systemic risks, the financial sector is treating AI security as a critical, ongoing battle for institutional integrity.
Comments (0)
No comments yet. Be the first!