NSA Reportedly Using Anthropic's Restricted Mythos AI Amid DoD Feud
The National Security Agency (NSA) is reportedly utilizing Anthropic's highly restricted Mythos Preview AI model, a development that emerges despite the Department of Defense (DoD) having previously designated Anthropic
The National Security Agency (NSA) is reportedly utilizing Anthropic's highly restricted Mythos Preview AI model, a development that emerges despite the Department of Defense (DoD) having previously designated Anthropic as a "supply chain risk." This revelation, first reported by Axios, highlights a complex and seemingly contradictory relationship between the U.S. government and a leading artificial intelligence firm.
Anthropic officially unveiled Mythos earlier this month, characterizing it as a cutting-edge AI designed specifically for sophisticated cybersecurity applications. However, the company opted to withhold the model from public release, citing concerns that its advanced capabilities could be misused for offensive cyberattacks. Consequently, access to Mythos has been limited to approximately 40 select organizations, with Anthropic publicly identifying only about a dozen of these recipients.
The NSA, a key intelligence agency within the U.S. government, appears to be among the undisclosed entities granted access to Mythos. According to reports, the agency is leveraging the AI primarily to scan digital environments for exploitable vulnerabilities, a critical function in national security. The United Kingdom’s AI Security Institute has also confirmed its access to the specialized model.
This reported usage by the NSA creates a perplexing dynamic, occurring just weeks after the DoD, the parent agency of the NSA, officially labeled Anthropic a "supply chain risk." This designation stemmed from Anthropic's refusal to grant Pentagon officials unfettered access to the full functionalities of its AI models. The core of the dispute originated from Anthropic's principled stand against making its AI, specifically the Claude model, available for mass domestic surveillance or the development of autonomous weapons systems.
The disagreement escalated to the point where Anthropic initiated a lawsuit against the Defense Department, challenging the "supply chain risk" designation. The lawsuit underscores the tension between the military's demand for advanced technological capabilities and AI developers' ethical considerations and control over their powerful creations.
Yet, even as the legal battle and public dispute unfold, there are signs that Anthropic's relationship with the current administration may be undergoing a notable shift. Last Friday, Anthropic CEO Dario Amodei reportedly engaged in a productive meeting at the White House with Chief of Staff Susie Wiles and Secretary of the Treasury Scott Bessent. This high-level engagement suggests a potential thawing of relations, despite the ongoing friction with the Pentagon.
The contradictory situation — one arm of the U.S. military actively deploying a restricted AI from a company that another, overarching military branch has deemed a risk — underscores the challenging landscape of AI integration into national security. It highlights the intricate balance between leveraging powerful new technologies for defense and intelligence, and navigating the ethical and control frameworks that AI developers are attempting to establish.
Anthropic has declined to comment on the matter, and TechCrunch has reached out to the NSA for their statement.
FAQ
Q: What is Anthropic's Mythos model?
A: Mythos is a frontier AI model developed by Anthropic, designed specifically for advanced cybersecurity tasks. Anthropic chose not to release it publicly due to concerns over its potential misuse for offensive cyberattacks, instead limiting access to a select group of around 40 organizations.
Q: Why did the Pentagon label Anthropic a "supply chain risk"?
A: The Department of Defense (DoD) designated Anthropic as a "supply chain risk" because the AI firm refused to provide Pentagon officials with unrestricted access to the full capabilities of its models. This dispute specifically arose from Anthropic's unwillingness to allow its Claude model to be used for mass domestic surveillance or the development of autonomous weapons.
Q: How is the NSA reportedly using Mythos?
A: The National Security Agency (NSA) is reportedly utilizing Anthropic's Mythos Preview model primarily for scanning digital environments. Its main application is to identify and analyze exploitable vulnerabilities, enhancing the agency's cybersecurity capabilities.
Related articles
China delivers a one-two punch to America’s AI dominance: Moonshot
Chinese AI firms Moonshot and Alibaba have unveiled new frontier models, Kimi K3 and Qwen3.8, challenging US dominance by claiming they rival OpenAI and Anthropic's top offerings. Their open-source strategy marks a significant shift, intensifying the global AI race with implications for national security and economic power.
Samsung OLED Laptop: Brighter & More Durable
Samsung's latest OLED laptop panels are a significant leap forward, boasting VESA DisplayHDR True Black 1400 certification, up to 1,600 nits peak brightness, and enhanced longevity thanks to a new tandem design.
Nvidia's Huang Forges Deep AI Ties with Japan's Industrial Giants
Nvidia CEO Jensen Huang secured major deals in Japan for "physical AI," aiming to integrate advanced AI into factory floors and robotics. The partnerships involve Japan's government-backed Noetra initiative, a coalition of top robotics firms, and expanded collaboration with Toyota, positioning Japan to lead in AI-driven manufacturing.
Robotaxi Rules Spark Fierce Battle Between Uber and Waymo in D.C
Uber and Waymo are locked in a contentious battle over proposed robotaxi regulations in Washington D.C., with Uber pushing a "hybrid" model to protect human drivers and Waymo poised to gain a market advantage under the current bill. This regulatory clash highlights broader industry debates over autonomous vehicle deployment and market control. Simultaneously, Uber is expanding its global delivery presence with a $14.8 billion acquisition of Delivery Hero, while other mobility startups secure significant funding.
in-depth: The Search for Another Earth-Like Planet Just Took a Big
Astronomers have achieved a significant breakthrough, confirming the existence of LHS 1140 b, a rocky exoplanet boasting an atmosphere within its star's habitable zone. Located 48 light-years away, this discovery marks the first time all three crucial criteria for potential habitability—rocky composition, suitable orbital distance, and a confirmed atmosphere—have been met. The findings propel the search for life beyond Earth into an exciting new phase.
regional: Silicon Valley icon Vinod Khosla: What kind of Seahawks
Silicon Valley legend Vinod Khosla's family group has acquired the Seattle Seahawks for a record $9.6 billion. This article explores how Khosla's tech-driven, disruptive approach might reshape the NFL franchise, drawing insights from his past talks and identifying potential challenges in blending tech innovation with sports tradition.