Anthropic Briefs House Homeland Security as Pentagon Court Filing Flags Foreign-Worker Security Risks
Anthropic briefed the House Homeland Security Committee behind closed doors as a Pentagon court filing on March 17 warned that the company’s large number of foreign nationals — reportedly including many from the People’s Republic of China — create "adversarial" supply‑chain risks because they could be compelled under China’s National Intelligence Law. The filing contrasts Anthropic with other labs even as the Defense Department continues to use its tools and may extend off‑boarding deadlines; Axios also notes industry recognition of Anthropic’s operational‑security measures (including disrupting an alleged Chinese cyber‑espionage campaign and banning PRC users) and that a hearing on its request for temporary relief is set for March 24.
📌 Key Facts
- In a March 17 court declaration, Pentagon undersecretary Emil Michael says Anthropic employs a large number of foreign nationals, including many from the People’s Republic of China (PRC).
- The Pentagon argues PRC‑national employees increase “adversarial risk” because they could be compelled to comply with China’s National Intelligence Law, and says its concerns go beyond policy disputes over surveillance and autonomous weapons.
- The court filing contrasts Anthropic with “other labs,” saying those firms’ leadership and behavior provide more acceptable technical and security assurances to the Department of Defense.
- Despite those concerns, the DOD is currently relying on Anthropic tools and may extend off‑boarding deadlines for their use.
- Axios reports Anthropic has been an early adopter of operational‑security practices: it helped disrupt an AI‑orchestrated Chinese cyber‑espionage campaign on its platform, has banned PRC users, and is widely viewed in the industry as among the most serious about insider‑threat controls (per Samuel Hammond).
- A hearing on Anthropic’s request for temporary relief from its “supply chain risk” designation is scheduled for March 24.
📊 Relevant Data
In 2020, China produced 3.57 million STEM graduates, far exceeding the United States' 820,000, contributing to the large pool of Chinese-origin AI talent in the U.S.
CNBC's The China Connection newsletter: U.S-China AI talent race — CNBC
China's National Intelligence Law of 2017, under Article 7, obligates all citizens and organizations to support, assist, and cooperate with national intelligence work, regardless of location.
What China's National Intelligence Law Says, And Why it Doesn't Matter — China Law Translate
From 2000 to 2023, there have been 224 reported instances of Chinese espionage directed at the United States, with a focus on technology and military sectors.
Survey of Chinese Espionage in the United States Since 2000 — Center for Strategic and International Studies (CSIS)
Foreign-born individuals constitute 30-60% of technical staff at major U.S. AI and technology companies as of 2025.
Impact on Tech and AI Companies — Threads
Restricting high-skilled immigration could reduce U.S. innovation and productivity, as foreign-born workers have contributed substantially to advancements in AI and technology.
Restricting high-skilled immigration is a blow to American innovation — YouTube (Brookings Institution)
📰 Source Timeline (2)
Follow how coverage of this story developed over time
- In a March 17 court declaration, Pentagon undersecretary Emil Michael asserts that Anthropic employs 'a large number of foreign nationals,' including 'many from the People’s Republic of China (PRC),' to build and support its large language model products.
- The Pentagon argues those PRC‑national employees increase 'adversarial risk' because they could be compelled to comply with China’s National Intelligence Law, and says concerns about Anthropic go beyond policy disputes over surveillance and autonomous weapons.
- The filing contrasts Anthropic with 'other labs,' claiming those other firms’ leadership and behavior provide more acceptable technical and security assurances to the Pentagon, while still noting that DOD is relying on Anthropic tools for now and may extend off‑boarding deadlines.
- Axios notes that Anthropic was an early adopter of operational‑security practices, helped disrupt an AI‑orchestrated Chinese cyber‑espionage campaign on its platform, and has banned PRC users, and quotes Samuel Hammond saying the firm is widely viewed in the industry as among the most serious about insider‑threat controls.
- The article confirms that a hearing on Anthropic’s request for temporary relief from its 'supply chain risk' designation is set for March 24.