According to a report by the Financial Times, citing sources familiar with the matter, Anthropic has sent approximately six engineers to the National Security Agency (NSA) to assist the agency in using its cutting-edge cybersecurity model, Mythos. The report did not specify whether these engineers or the model have been directly involved in NSA hacking operations.
Engineers have been stationed at the NSA
The report states that the engineers' task is to help the NSA deploy and use Mythos in specific scenarios. At this stage, the specific scope of application is unclear, and it cannot be confirmed whether it involves offensive cyber operations.
The NSA's responsibilities include not only gathering intelligence through methods such as surveillance, undersea cables, and corporate partnerships, but also launching cyberattacks against foreign adversaries. Therefore, this collaboration quickly drew attention to the potential applications of AI models in intelligence and cyber warfare scenarios.
The ban had previously been controversial.
This report continues the controversy surrounding Mythos. Axios reported in April that the NSA was already using Mythos, despite federal bans on the use of Anthropic technology.
The ban is reportedly related to the U.S. Department of Defense's earlier designation of Anthropic as a "supply chain risk." The decision has been described as a response to Anthropic's refusal to allow the government to use its models for large-scale domestic surveillance and autonomous weapons.
The official response was not direct.
When TechCrunch contacted the NSA for verification, an NSA spokesperson declined to confirm or deny the report. Anthropic did not respond to requests for comment.
Currently, publicly available information cannot confirm the actual extent of Mythos's deployment within the NSA, nor can it determine whether it has entered into more sensitive cyber operations. However, the direct involvement of engineers in supporting this model suggests that its connection with the US intelligence system is deepening.
