Jamie Dimon, chief executive officer of JPMorgan Chase, said on Wednesday that the risks posed by Anthropic’s Mythos artificial intelligence model are real and are being addressed by the U.S. government. Dimon made the remarks while speaking at a defense and innovation summit in Pennsylvania hosted by Senator Dave McCormick.
Dimon argued that access to advanced AI capabilities should be tightly controlled, warning that using Mythos is akin to putting "ballistic missiles in private hands." His comment highlights the concerns that prompted Washington to intervene in the deployment of high-performance AI models.
Anthropic made Mythos available in April to a small group of partners, reportedly including JPMorgan. Financial institutions were attracted to the model because it is among the most advanced systems for identifying cybersecurity vulnerabilities, potentially enabling companies to detect and remediate weaknesses in their systems more quickly.
In June, the U.S. government directed Anthropic to restrict access by foreign nationals to its most advanced models — including Fable 5 and Mythos 5 — citing national security reasons. That decision temporarily cut off access to those services worldwide. The restrictions were later lifted after Anthropic implemented additional security safeguards.
Both governments and technology firms are grappling with how to prevent the misuse of powerful AI, particularly systems capable of finding software vulnerabilities. U.S. policymakers have been especially concerned that advanced capabilities could be exploited by the militaries or intelligence services of China, Russia, or other countries considered national security risks.
Dimon’s remarks reflect the broader unease surrounding the national security and regulatory challenges posed by the most advanced forms of artificial intelligence.
