1.5 million users cancel ChatGPT plus over OpenAI's multi-billion Pentagon deal

In a stunning display of unity against the militarization of artificial intelligence, OpenAI has seen a record-shattering 1.5 million individual users cancel their "ChatGPT Plus" subscriptions in just 48 hours. The mass cancellation event began late on February 27, 2026, shortly after the company announced a multi-billion-dollar contract to provide its frontier models to the Department of War (Pentagon), effectively filling the void left by Anthropic’s high-profile refusal to cooperate with military demands.

The exodus marks the first major "ethical revolt" by consumers in the generative AI era, wiping out an estimated $360 million in projected annual recurring revenue (ARR) almost overnight.

The viral #DeleteChatGPT movement

The "Deal with the Pentagon" was intended to be a victory lap for OpenAI’s enterprise growth, but it instead ignited a firestorm across social media and developer communities.

Users began posting screenshots of their cancellation confirmations on X and Reddit with the hashtag #DeleteChatGPT, citing OpenAI’s abandonment of its original "benefit to humanity" charter.

Analytics firm Sensor Tower reports a simultaneous 400% surge in new subscriptions for Anthropic’s Claude, as users migrate to the only major AI lab currently banned by the Trump administration for its anti-weapons stance.

Sources within OpenAI’s San Francisco headquarters report that the decision has caused "significant internal friction," with at least twelve senior researchers resigning in protest over the weekend.

The dual-use controversy

The Pentagon deal specifically involves the use of OpenAI’s newest multimodal models for strategic logistics, "cognitive electronic warfare," and real-time battlefield analysis.

OpenAI CEO Sam Altman defended the move in a leaked internal memo, arguing that "supporting the defense of democracy" is a moral obligation and that OpenAI’s safety guardrails remain the most robust in the industry.

The critic's view

Privacy advocates and AI ethicists have slammed the deal as a "betrayal of trust," arguing that OpenAI’s models were trained on the public’s data under the guise of open research, only to be sold to the military for kinetic operations.

The lesser of two evils

Some industry analysts suggest OpenAI took the deal to prevent a total monopoly by Elon Musk’s xAI, which has reportedly been seeking exclusive "sovereign" status within the administration’s tech circle.

Financial fallout vs government gain

While the loss of 1.5 million subscribers is a PR nightmare, the financial math remains complex for the company recently valued at $840 billion.

he Pentagon contract is rumored to be worth between $5 billion and $10 billion over three years, which would technically more than offset the loss of consumer subscription revenue.

Still, experts warn that the real danger lies in "brand dilution." If OpenAI loses its status as the world’s "default" AI for creative and personal use, it may struggle to justify its trillion-dollar IPO aspirations later this year.

Lead investors like Microsoft and SoftBank have remained silent, though Microsoft’s stock saw a 1.2% dip on news of the subscriber churn.