OpenAI is in huge trouble, and they should have seen it coming!
OpenAI just saw a massive 200% uninstallation rate, all because of a major conflict between Anthropic & the US Government.
Not to mention, haters even started a “Quit ChatGPT“ protest.
In July 2025, Anthropic signed a $200 Million agreement with the US Government for the use of Claude and its AI capabilities in their government services and classified networks.
The agreement had two important guardrails — security restrictions.
- Don’t use Claude for mass surveillance
- Don’t use Claude for weapon development.
That’s it.
Now, recently, the government claimed that they conducted a secret operation to capture the Venezuelan president with the help of Claude tools. But there’s no specific mention of how they utilized it.
My guess is that they must have used it in their surveillance technologies.
Here comes the conflict.
Realizing the potential and the power of Claude, the U.S. Department of Defense (Pentagon) wanted Anthropic to drop the restrictions from the contract.
When asked, the Pentagon said:
- They are not planning to do mass surveillance or weapon development without a human in the loop right now
- But they just don’t want to be restricted by the contract.
But Anthropic refused to drop the safeguards, claiming that it was not part of the original agreement and it should not be now.
So, the Pentagon sent a revised agreement that would allow the government to override safety safeguards whenever it wanted, which is not so good.
But, Anthropic standing true to their ethical standards, rejected the revised agreement, saying that they would rather happily walk away from the contract than be part of something that promotes mass surveillance or weapon development.
Then it turned into a threat.
Pete Hegseth (the U.S. Defense Secretary) gave Anthropic a deadline.
He basically said:
- Agree to let us use your AI without any restrictions, or we’ll cancel the contract.
- We might officially label you a “supply chain risk” (which makes it harder for you to work with the government).
- We could use a powerful law called the Defense Production Act to force you to give full access to your technology if needed.
Despite the pressure, Anthropic refused to remove its two safety protections.
So the administration (under Donald Trump) decided to:
- Start cutting ties with Anthropic across U.S. government agencies.
- Gradually remove their systems from Pentagon use.
- Look for other AI companies willing to accept their conditions.
This is where OpenAI enters.
Right after this incident, OpenAI initiates a conversation with the government and signs a new agreement, agreeing to what the government is asking for.
Literally the next day, there’s a huge spike in the uninstallation rate of ChatGPT of up to 200%. Earlier, they just had 9%.
Clearly, people were supporting Anthropic’s decisions in this matter. Anthropic did nothing wrong. They wanted to stay true to their ethical standards, and people respected that.
But now, when OpenAI signed a similar deal that Anthropic refused, OpenAI is now being hated.
People started uninstalling OpenAI just like that and shifted to Claude instead, whose installation rate just spiked!
This surprisingly led to the “Quit ChatGPT” protest.
Didn’t expect this, by the way.
Okay, so this is the whole story. OpenAI’s reputation was already slightly sinking due to the release of ChatGPT Ads, but this would definitely drown it.
From a common perspective, what do you think?