Jamie Dimon Warns Anthropic's Mythos AI Poses National Security Risks

JPMorgan Chase CEO Jamie Dimon has warned that Anthropic’s advanced Mythos artificial intelligence model presents significant national security risks, highlighting growing concerns over the potential misuse of powerful AI systems.

Speaking at Senator Dave McCormick’s Pennsylvania Defense and Innovation Summit on Wednesday, Dimon said the U.S. government is taking the issue seriously and emphasized that access to cutting-edge AI technologies must remain tightly controlled. He compared unrestricted access to Mythos with handing out "ballistic missiles," arguing that highly capable AI models should not be freely available to individuals because of the security threats they could pose.

Anthropic introduced its Mythos AI model in April to a limited group of organizations, including JPMorgan Chase. The model quickly gained attention within the financial industry for its ability to detect cybersecurity vulnerabilities, allowing companies to identify and address software weaknesses more efficiently. Financial institutions have viewed the technology as a valuable tool for strengthening cyber defenses and protecting critical infrastructure.

Despite its commercial potential, Mythos has also raised concerns among U.S. policymakers. In June, the U.S. government instructed Anthropic to restrict access to its most advanced AI models, Fable 5 and Mythos 5, for foreign nationals due to national security considerations. The move reflected fears that sophisticated AI capable of discovering software vulnerabilities could be exploited by hostile governments or intelligence agencies.

Those restrictions were later lifted after Anthropic implemented additional security safeguards designed to reduce the risk of misuse while maintaining access for authorized users.

The debate surrounding advanced artificial intelligence continues to intensify as governments and technology companies seek to balance innovation with security. Washington has increased oversight of next-generation AI systems amid concerns that powerful models could be leveraged for cyberattacks, military intelligence, or other malicious purposes by countries such as China and Russia.

Anthropic did not immediately respond to requests for comment on Dimon’s remarks outside regular business hours. The company remains at the center of broader discussions over AI governance, cybersecurity, and the responsible deployment of increasingly capable artificial intelligence models.