NSA Uses AI Tool Despite Pentagon Risk Concerns
The United States National Security Agency is reportedly using Anthropic’s Mythos Preview artificial intelligence tool, even after the Pentagon assigned the company a formal supply chain risk designation. The development highlights ongoing tensions between rapid technological adoption and national security safeguards.
According to a report published on Sunday, the Mythos Preview model has seen broader use within the Department of Defense. However, the details of its deployment remain unclear, and the report has not been independently verified.
Wider Adoption Raises Questions
Sources indicate that the AI model is being utilised more extensively across the department than previously understood. Neither Anthropic nor U.S. defence authorities responded immediately to requests for comment outside regular business hours.
The National Security Agency operates under the Department of Defense, placing the reported usage within a highly sensitive security environment. Consequently, the situation raises concerns regarding compliance with internal risk assessments and procurement guidelines.
Renewed Talks Signal Changing Approach
Recent discussions between U.S. President Donald Trump’s administration and Anthropic’s leadership suggest a potential shift in relations. These talks marked the first engagement since an earlier dispute between the Pentagon and the AI firm over the appropriate use of its technology.
The renewed dialogue comes at a time when artificial intelligence is becoming increasingly central to defence strategy. Officials appear to be balancing the advantages of advanced AI tools with the risks they may introduce.
Concerns Over Cybersecurity Capabilities
Anthropic has described Mythos as its most advanced model to date, particularly in coding and autonomous task execution. These capabilities enable the system to operate with a high degree of independence, performing complex functions without continuous human input.
Experts have warned that such advanced coding ability could pose significant cybersecurity risks. The model may be capable of identifying vulnerabilities in digital systems and developing methods to exploit them. This potential has fuelled concerns that the technology could accelerate cyberattacks if misused.
Balancing Innovation And Risk
The reported use of Mythos within a national security agency underscores the broader challenge facing governments. On one hand, advanced AI offers powerful tools for defence, intelligence, and operational efficiency. On the other, it introduces new risks that require careful management.
As artificial intelligence continues to evolve, authorities are likely to face increasing pressure to establish clear frameworks governing its deployment. The outcome of this situation may influence how similar technologies are integrated into sensitive government operations in the future.
With inputs from Reuters