A major clash is brewing between the tech industry and the U.S. Department of Defense (DoD), sparked by a recent government action against a leading AI company.
At the heart of the issue is a letter from the Information Technology Industry Council (ITI), a powerful lobby group whose members include Nvidia, Amazon, and Apple. They've raised alarms that the Pentagon's decision to label AI firm Anthropic a 'supply-chain risk' could cause chaos far beyond just one company. The group fears this move will disrupt the entire defense technology ecosystem, affecting cloud computing services and access to the best available AI tools.
So, what led to this? The story unfolds in a few key steps. First, President Trump issued a directive on February 27, 2026, ordering all federal agencies to stop using Anthropic's technology. Following this, Defense Secretary Pete Hegseth publicly declared Anthropic a national security risk. The problem, as legal experts point out, is that the law Hegseth seems to be using likely doesn't grant him the power for such a sweeping ban that affects all private defense contractors. This gap between the public declaration and its legal foundation is what has the tech industry so concerned about uncertainty.
Second, this isn't happening in a vacuum. Anthropic is a major partner for Amazon Web Services (AWS), with its Claude AI models deeply integrated into the Amazon Bedrock platform used by many companies. A broad ban, even if legally questionable, could scare defense-related businesses away from using Claude, complicating major cloud contracts like the Pentagon's $9 billion JWCC program, which was designed to use multiple cloud and AI providers.
Third, the competitive landscape adds another layer of complexity. Just as the news broke, rival AI company OpenAI announced it had secured a new agreement with the Pentagon that includes specific safety rules. This move suggests the DoD's issue might be less about a total opposition to AI safety 'guardrails' and more about gaining negotiating leverage over a specific company, Anthropic. This changes the narrative and puts pressure on companies reliant on Anthropic's technology.
While the stock market showed some jitters, with Nvidia and Apple dipping, the direct financial risk to these giants is relatively small. For Amazon, the at-risk revenue from DoD cloud contracts is less than 1% of AWS's total. The real threat is strategic: it introduces political risk into government AI projects and could fragment the carefully built tech supply chain the Pentagon relies on.
- Supply-Chain Risk: A designation that a company or its products pose a potential threat to national security, often leading to restrictions on its use by government agencies and contractors.
- JWCC (Joint Warfighting Cloud Capability): A major multi-billion dollar Pentagon program to procure cloud services from multiple vendors (like Amazon, Google, Microsoft) to modernize its IT infrastructure.
- 10 U.S.C. §3252: A section of U.S. law that allows the government to exclude certain contractors from procurements related to 'National Security Systems', but legal experts argue it does not authorize a blanket ban on all commercial activity with a company.