NSA using Anthropic's Mythos model: Report

Key Concepts

• Mythos Model: A specialized AI model developed by Anthropic, currently restricted to select corporate and government entities for cybersecurity purposes.
• Supply Chain Risk: The classification used by the Department of Defense (DoD) to justify banning Anthropic’s software.
• Contagion Risk: The concern among investors that government-level regulatory or legal conflicts could negatively impact the company’s broader business operations and market standing.
• Market Share: The percentage of total paid AI subscriptions held by a company; Anthropic currently holds ~30% compared to OpenAI’s ~35%.

NSA Utilization of Mythos

Recent reports indicate that the National Security Agency (NSA) is utilizing Anthropic’s "Mythos" AI model. This development is significant because the NSA operates under the Department of Defense (DoD), which recently placed Anthropic on a blacklist, labeling its software a "supply chain risk."

• Purpose: The Mythos model is currently being deployed specifically for cybersecurity applications.
• Availability: Anthropic has limited the release of Mythos to approximately 40 companies and select government agencies, with no current plans for a broader public release.
• Official Stance: Both the NSA and Anthropic have declined to provide specific comments regarding the deployment of the model.

The DoD Blacklist and Regulatory Conflict

The inclusion of Anthropic on the DoD’s blacklist was described as an "unprecedented move," as such designations are typically reserved for U.S. adversaries.

• Root Cause: The ban stemmed from a fundamental disagreement between the DoD and Anthropic regarding safety protocols and usage restrictions.
• Diplomatic Efforts: To address these tensions, Anthropic CEO Dario Amodei held meetings in Washington D.C. with Treasury Secretary Janet Yellen and White House Chief of Staff Susie Wiles. Both parties characterized these discussions as "productive," suggesting a potential "thawing" of the relationship despite the ongoing legal and regulatory friction.

Market Performance and Investor Sentiment

Despite the government-level scrutiny, Anthropic’s commercial growth remains robust, providing relief to investors concerned about "contagion risk."

• Growth Metrics: According to data from Ramp, Anthropic has captured approximately 30% of all paid AI subscriptions.
• Competitive Landscape: Anthropic is rapidly closing the gap with the current market leader, OpenAI, which holds a 35% market share.
• Investment Outlook: The company continues to attract interest, reportedly fielding new offers for equity investments, indicating that the government drama has not yet stifled investor confidence or the company's valuation trajectory.

Synthesis and Conclusion

The situation highlights a complex paradox in U.S. technology policy: while the Department of Defense has officially flagged Anthropic as a supply chain risk, the intelligence community (NSA) continues to rely on the company’s specialized AI tools for critical cybersecurity infrastructure. The ongoing high-level meetings between Anthropic leadership and the White House suggest that the government is seeking a middle ground between strict security compliance and the strategic necessity of utilizing advanced AI capabilities. For now, the company’s strong market growth and competitive positioning against OpenAI suggest that the regulatory conflict has not yet hindered its commercial momentum.