Anthropic vs. the Pentagon: What’s actually at stake?

Source: TechCrunch

The Ongoing Clash Over Military Use of AI

Key Players

• Anthropic CEO Dario Amodei – Opposes using Anthropic’s models for mass surveillance of Americans or fully autonomous weapons that can strike without human oversight.
• Defense Secretary Pete Hegseth – Argues the Department of Defense should not be constrained by a vendor’s policies and that any “lawful use” of AI should be allowed.

Timeline

Core Issue

Who should control powerful AI systems?

• The companies that build them (e.g., Anthropic) – prioritize ethical safeguards and limit misuse.
• The government (e.g., the Department of Defense) – seeks unrestricted access for national‑security purposes.

Stakes

• Ethical & civil‑rights concerns – Potential for mass surveillance and autonomous lethal actions without human decision‑making.
• National‑security imperatives – The Pentagon argues that AI can provide critical advantages on the battlefield.
• Supply‑chain risk – The government could label Anthropic a “risk” if it refuses cooperation, affecting the company’s ability to do business with federal agencies.

What’s Next?

• Policy negotiations – Expect continued dialogue (or confrontation) over licensing, oversight, and permissible use cases.
• Potential legislation – Congress may intervene to set clearer boundaries for AI in defense contexts.
• Industry response – Other AI firms are watching closely; their stances could shape broader standards for military AI deployment.

What Is Anthropic Worried About?

Anthropic doesn’t want its AI models used for:

• Mass surveillance of Americans
• Autonomous weapons that operate without a human in the loop for targeting and firing decisions.

Traditional defense contractors typically have little say in how their products are used, but Anthropic has argued from its inception that AI technology poses unique risks and therefore requires unique safeguards. From the company’s perspective, the challenge is how to maintain those safeguards when the technology is being used by the military.

Military Context

• The U.S. military already relies on highly automated systems, some of which are lethal.
• Historically, the decision to use lethal force has been left to humans, but there are few legal restrictions on the military use of autonomous weapons.
• The DoD does not categorically ban fully autonomous weapons systems. According to a 2023 DoD directive¹, AI systems can select and engage targets without human intervention, provided they meet certain standards and receive review by senior defense officials.

Why this worries Anthropic
Military technology is secretive by nature. If the U.S. military were automating lethal decision‑making, we might not know about it until the systems are operational. If those systems used Anthropic’s models, the company could be implicated in “lawful use” while still violating its own safety principles.

Event Highlight

Anthropic’s position isn’t that such uses should be permanently off the table; rather, it believes its models aren’t yet capable enough to support them safely. Imagine an autonomous system misidentifying a target, escalating a conflict without human authorization, or making a split‑second lethal decision that cannot be reversed. A less‑capable AI in charge of weapons becomes a fast, confident machine that is bad at making high‑stakes calls.

Surveillance Concerns

AI also has the power to super‑charge lawful surveillance of American citizens to a concerning degree. Under current U.S. laws, surveillance of Americans is already possible through the collection of texts, emails, and other communications. AI changes the equation by enabling:

• Automated large‑scale pattern detection
• Entity resolution across disparate datasets
• Predictive risk scoring
• Continuous behavioral analysis

These capabilities raise significant privacy and civil‑rights questions that Anthropic wants to address before its models are deployed in such contexts.

---

What Does the Pentagon Want?

The Pentagon argues that it should be able to deploy Anthropic’s technology for any lawful use it deems necessary, rather than being limited by Anthropic’s internal policies on topics such as autonomous weapons or surveillance.

Key Points

• Secretary Pete Hegseth’s stance

  • The Department of Defense (DoD) should not be constrained by a vendor’s rules.
  • The DoD would use the technology only for “lawful” purposes.

• Official statement from Sean Parnell

  • In a Thursday X post, the Pentagon’s chief spokesperson said the department has no interest in mass domestic surveillance or deploying autonomous weapons.
  • Quote:

    “Here’s what we’re asking: Allow the Pentagon to use Anthropic’s model for all lawful purposes. This is a simple, common‑sense request that will prevent Anthropic from jeopardizing critical military operations and potentially putting our warfighters at risk. We will not let ANY company dictate the terms regarding how we make operational decisions.”

• Deadline to Anthropic

  • Anthropic has until 5:01 p.m. ET on Friday to respond.
  • If no agreement is reached, the Pentagon will terminate the partnership and label Anthropic a supply‑chain risk for DOW.

Context & Background

• Cultural grievances

  • Secretary Hegseth’s concerns appear linked to broader cultural issues.
  • In a speech at SpaceX and xAI offices (January), he warned against “woke AI,” which many interpreted as a prelude to his conflict with Anthropic.

• Relevant excerpt from the speech

  “Department of War AI will not be woke. We’re building war‑ready weapons and systems, not chatbots for an Ivy League faculty lounge.”

• Full transcript: Remarks by Secretary of War Pete Hegseth at SpaceX

---

The Pentagon’s request centers on unrestricted, lawful use of Anthropic’s AI models, while the debate also reflects underlying tensions over the role of corporate policy and cultural values in defense technology.

So What Now?

The Pentagon has threatened to either:

1. Declare Anthropic a “supply‑chain risk,” effectively blacklisting the company from doing business with the government, or
2. Invoke the Defense Production Act (DPA) to force Anthropic to tailor its model to the military’s needs.

Hegseth has given Anthropic until 5:01 p.m. on Friday to respond. With the deadline looming, it’s anyone’s guess whether the Pentagon will follow through on its threat.

Stakes for Both Sides

This isn’t a fight either party can easily walk away from. Sachin Seth, a VC at Trousdale Ventures who focuses on defense tech, says a supply‑chain‑risk label for Anthropic could mean “lights out” for the company.

“[The Department] would have to wait six to twelve months for either OpenAI or xAI to catch up,” Seth told TechCrunch. “That leaves a window of up to a year where they might be working from not the best model, but the second or third best.”

If Anthropic is dropped from the DoD, it could become a national‑security issue.

Who Might Fill the Gap?

• xAI is gearing up to become classified‑ready and could replace Anthropic. Given owner Elon Musk’s rhetoric on the matter, the company would likely have no problem giving the DoD total control over its technology.
• Recent reports suggest that OpenAI may stick to the same red lines as Anthropic, limiting its willingness to comply fully with DoD demands.

Footnotes

1. 2023 DoD Directive – “Artificial Intelligence in the Department of Defense.” Available at: ↩