Skip to content
Sign in
Defense & Security · North America

Anthropic president questions whether OpenAI’s Pentagon deal was opportunistic

Anthropic president Daniela Amodei questioned whether OpenAI’s Pentagon deal reflected principle or opportunism after Anthropic declined to change its military AI restrictions. Anthropic CEO Dario Amodei warned that one person controlling a million AI-powered robots or drones could undermine democracy.

AI-assisted synthesis · Based on 2 outlets' reporting · 1 article captured only in partHow it's made

View timeline · 8 entries
2 outlets2 regionsLatest reporting 3 min read
Coverage: Too few ratedL 0 · C 1 · R 1
File photo
Photo: Touch Of Light / Wikimedia Commons (CC BY-SA 4.0)

What happened

Key facts

  • Anthropic president Daniela Amodei questioned whether OpenAI’s Pentagon deal was a “really principled stand” or an “opportunistic moment.” (per New York Post, pluang.com).
  • Anthropic had refused to change its military AI restrictions before OpenAI made its Pentagon deal (per New York Post, pluang.com).
  • Daniela Amodei said Anthropic could control only its own decisions and had taken a position it considered ethically correct (per New York Post, pluang.com).
  • Anthropic CEO Dario Amodei warned that one person controlling an army of a million AI-powered robots or drones could undermine democracy (per New York Post).
  • Anthropic was preparing for an initial public offering and had filed paperwork to list shares on Nasdaq under the ticker symbol “ANT,” according to the report (per New York Post, pluang.com).
  • Anthropic reported revenue above $16 billion and a net loss of $80.7 million in the first half of 2026, compared with a $13 billion loss in the same period the previous year, according to Wired as cited by the report (per New York Post).

The brief

Anthropic president Daniela Amodei questioned whether OpenAI’s Pentagon agreement reflected a principled position or an opportunistic opening, after Anthropic refused to alter its military AI restrictions. Her criticism places the companies’ different decisions on military use of AI at the center of a public dispute, though she said Anthropic could control only its own choices.

Daniela Amodei made the comments in a joint interview with her brother, Anthropic CEO Dario Amodei. Biographer Walter Isaacson conducted the interview for his podcast “The Builders,” which was scheduled to air Tuesday. The report does not describe the terms of OpenAI’s agreement or identify what restrictions OpenAI accepted.

Daniela Amodei described Anthropic’s approach as taking a position the company believed ethically correct. She did not state that OpenAI acted for a particular motive; instead, she posed the question of whether the deal represented principle or opportunism.

The distinction matters because the account says Anthropic declined to change its own restrictions before OpenAI reached its Pentagon agreement, but does not provide an account from OpenAI. Dario Amodei raised a separate concern about the concentration of control over AI-enabled military systems.

He warned that a single person directing a million AI-powered robots or drones could undermine democracy. The interview excerpt offers that as a hypothetical risk, not as a report that such an army exists or has been deployed. The comments came as Anthropic prepared for a planned public offering.

The report says the company filed paperwork Monday to list on Nasdaq under the ticker “ANT.” It also cites Wired’s account of Anthropic’s first-half 2026 financial results: revenue above $16 billion and a net loss of $80.7 million, compared with a $13 billion loss in the same period the previous year.

The interview thus connects a debate about military AI safeguards with Anthropic’s public positioning as it prepares to become a publicly traded company. The figures reported for the first half show a smaller net loss than the year-earlier period, while revenue exceeded $16 billion; the account does not establish that the planned listing prompted Daniela Amodei’s remarks.

The central unresolved issue is how the companies’ respective restrictions and contractual terms differ. The reporting records Anthropic’s refusal to change its rules and its president’s challenge to the meaning of OpenAI’s deal, but does not provide enough detail to compare the agreement’s operational provisions or establish OpenAI’s reasoning.

Why it matters

  • Anthropic’s military AI restrictions and OpenAI’s Pentagon agreement put the limits governing military AI use at the center of a dispute between two companies working in the same field.
  • Dario Amodei’s warning identifies a specific democratic risk: one person having control over a million AI-powered robots or drones.
  • Anthropic’s planned Nasdaq listing, under ticker ANT, coincides with public scrutiny of its military AI position; the report also gives first-half revenue above $16 billion and a net loss of $80.7 million.
  • Anthropic’s reported net loss narrowed from $13 billion in the first half of the previous year to $80.7 million in the first half of 2026, while revenue exceeded $16 billion.

What to watch next

  • Whether Anthropic’s planned Nasdaq listing proceeds under the ticker symbol ANT.
  • Whether the scheduled “The Builders” interview provides further detail on Daniela Amodei’s comments.
  • Whether OpenAI publicly explains the terms or rationale for its Pentagon agreement.
  • Whether the companies disclose further detail about their respective military AI restrictions.

How it's being covered

Coverage analysisShow details
3 source articles · 2 unique outlets
Outlet political mix: Too few rated· 1 Center · 1 Right
Where the outlets are based: US/Canada 50% · Asia 50%
Outlet political mix
Too few rated
Left
Center
Right
0 outlets
1 outlet
1 outlet
The marker shows coverage balance: each unique outlet counts once.
Geography
Where the outlets covering this story are based.
2 unique outletsDominant coverage: US/Canada
Limited coverage
  • Coverage is currently limited to 2 outlets.
How we measure coverage

How outlets' accounts compare

2 framing differences · 2 uncorroborated claims · 2 information gaps
The reports repeat the same account with different emphasis; its central claims are not independently confirmed in the reporting Updat3 read.Compared across 2 reports whose article text Updat3 read.
Framing / emphasis differences
Sources emphasise different angles, actors or consequences of the same reporting.
  • The New York Post centers Daniela Amodei’s question about whether OpenAI’s agreement was principled or opportunistic and includes Dario Amodei’s warning about concentrated control of AI-powered robots or drones.
  • Pluang’s headline frames the story as Anthropic questioning the ethics of OpenAI’s military AI agreement, but its supplied text gives no details about the interview, the companies’ positions, or the robotics warning.
Updat3: These are differences of emphasis, not disputes about what is reported. They do not show that the reported claims are true.
One-sided or uncorroborated claims
Claims that come from one party or one source family and are not independently corroborated.
  • Daniela Amodei asked whether OpenAI’s Pentagon deal was a “really principled stand” or an “opportunistic moment.”
  • Dario Amodei warned that one person controlling a million AI-powered robots or drones could undermine democracy.
Updat3: Not independently corroborated yet — treat as a claim, not an established fact.
Not established by the reporting
Information the available reporting does not contain or confirm. This is a gap, not a disagreement.
  • Neither source supplies the agreement’s operational terms, details of OpenAI’s safeguards, or a direct response from OpenAI.
  • The account does not provide OpenAI’s explanation of its agreement or enough terms to compare its military AI restrictions with Anthropic’s.
Updat3: Updat3 cannot conclude anything about this from the current sources.

Timeline

Developments from this story's sources, dated background we compiled and fact-checked, and our earlier coverage. Each row says which.
  1. The commercial AI sector’s roots reach back to OpenAI’s founding in December 2015 as a nonprofit research organization.
    Background · from our historical context
  2. The Defense Department created the Joint Artificial Intelligence Center to coordinate AI adoption.
    Background · compiled and fact-checked by Updat3
  3. In 2019, OpenAI created a “capped-profit” subsidiary to raise capital while retaining nonprofit control.
    Background · from our historical context
  4. Anthropic was established in 2021 as a public benefit corporation, with its founders emphasizing research into AI safety.
    Background · from our historical context
  5. In June 2022, the department published its Responsible Artificial Intelligence Strategy and Implementation Pathway, calling for AI systems to be governed responsibly throughout their life cycle.
    Background · from our historical context
  6. In 2023, Anthropic formalized its safety approach in the Responsible Scaling Policy, which described capability thresholds and safeguards the company would use as its systems advanced.
    Background · from our historical context
  7. In August 2023, Deputy Secretary of Defense Kathleen Hicks announced the Replicator initiative, which aimed to field thousands of attritable autonomous systems within two years.
    Background · from our historical context
  8. In January 2024, OpenAI removed a blanket prohibition on “military and warfare” from its usage policies, while continuing to bar uses such as developing or using weapons.
    Background · from our historical context

How we got here

The immediate backdrop to the dispute was a widening difference between the two companies’ published rules for military uses of AI. In January 2024, OpenAI removed a blanket prohibition on “military and warfare” from its usage policies, while continuing to bar uses such as developing or using weapons.

Anthropic, founded in 2021 by former OpenAI employees including Dario and Daniela Amodei, had set out a different approach in its September 2023 Responsible Scaling Policy, which tied the company’s deployment decisions to assessments of catastrophic risk.

Original sources

1 of 3 articles · US/Canada · Show all

Every story links to the reporting it was built from. How we choose sources

Spotted a factual error, a broken link or a misleading summary? Tell us.