AI Model at the Center of Controversy

The National Security Agency (NSA) is reportedly using Anthropic's AI model, Mythos, despite a federal ban on its use.

The ban was imposed in retaliation for Anthropic's decision not to allow the government to use its models for mass domestic surveillance and autonomous weapons.

According to The Financial Times, Anthropic has deployed around half-a-dozen engineers to the NSA to help its spies use Mythos for certain applications. However, it's unclear if the engineers or Mythos are being actively used in the agency's hacking operations.

The NSA's use of Mythos raises concerns about the potential misuse of AI technology in cybersecurity.

"We believe that our technology has the potential to be used for good or ill," said an Anthropic spokesperson. "We're working closely with governments and other organizations to ensure that it's used responsibly."

The NSA is tasked with collecting intelligence from various sources, including wiretaps, undersea cables, corporate partnerships, and clandestine means, as well as conducting offensive cyberattacks on foreign adversaries.

• Anthropic's decision to limit access to Mythos follows concerns about its cybersecurity capabilities being exploited to discover security flaws and carry out hacks.
• The use of AI technology in cybersecurity is a growing trend, with many companies investing heavily in this area.
• The collaboration between the NSA and Anthropic highlights the complex relationship between governments and technology companies.

When reached by TechCrunch, an NSA spokesperson declined to confirm or deny the reporting.

Anthropic did not respond to a request for comment.