Sam Altman regrets rushed defense deal as ChatGPT uninstalls surge by 295%

In a dramatic turn of events that has sent shockwaves through the tech world, OpenAI finds itself at the center of a firestorm. A recently signed defense deal with the US Department of War (DoW) has sparked a massive user backlash, leading to a 295% surge in ChatGPT uninstalls, according to data from Sensor Tower.

This isn’t just another tech controversy. It’s a critical moment where the rapid advancement of artificial intelligence collides with fundamental questions about privacy, ethics, and national security. If you’re a ChatGPT user concerned about where your data might go, a tech professional watching the AI landscape, or simply someone trying to make sense of the news, this article is for you. We’ll break down exactly what happened, why users are abandoning the app in droves, and what Sam Altman’s “sloppy” admission means for the future of AI.

The Catalyst: A “Rushed” Deal with the Department of War

The root of the problem lies in a partnership between OpenAI and the US Department of War. While the specifics were initially vague, the mere association with a military body—especially one with a name evoking historical conflicts—triggered immediate alarm among privacy advocates and ethically-conscious users.

Why the Deal Caused an Uproar

For many, the fear was not just about autonomous weapons, which Altman has previously spoken against. The core anxieties centered on two key areas:

1. Domestic Surveillance: The terrifying prospect of ChatGPT’s powerful analytical capabilities being used to monitor US citizens.
2. Unchecked Military Power: The fear that advanced AI could be used in ways that violate international laws or ethical norms without proper oversight.

This anxiety was amplified when Anthropic, the developer of Claude and a company also known for its focus on AI safety, walked away from a similar deal last week. Anthropic reportedly could not secure the safety and privacy assurances it wanted, making OpenAI’s decision to proceed seem even more concerning by comparison.

“Opportunistic and Sloppy”: Altman’s Mea Culpa

Facing a PR crisis and a user exodus, Sam Altman took to X (formerly Twitter) to do damage control. In an internal memo he shared publicly, he admitted the announcement was mishandled.

“We shouldn’t have rushed to get this out on Friday,” Altman wrote. “It just looked opportunistic and sloppy.”

This rare admission of fault from a CEO was a key moment. It acknowledged that the rollout was botched, but did it go far enough to address the deeper concerns?

The Fallout: The Great ChatGPT Exodus

The numbers paint a stark picture of user sentiment. According to mobile data analytics firm Sensor Tower (via TechCrunch), the impact has been swift and severe:

• ChatGPT uninstall rates in the US surged by 295% in the days following the announcement.
• Nearly three times as many users are removing the app from their phones compared to an average day.

The Winner? Claude’s Astonishing Rise

Where are these users going? The data suggests they are flocking directly to the competition.

• Claude app installs soared by 37% last Friday and a massive 51% last Saturday.
• Claude has rocketed to the top of the Apple App Store charts, positioning itself as the ethical alternative for AI-powered assistance.

This mass migration highlights a growing trend: users are increasingly willing to vote with their wallets (and their app storage) based on corporate ethics. The fact that Claude recently made its chat memory feature available to all users only sweetens the deal for newcomers.

Breaking Down the Damage Control: What OpenAI Changed

In his efforts to stem the tide, Sam Altman announced specific amendments to the DoW agreement. These changes were clearly designed to tackle the most significant privacy fears head-on.

Key Amendment: No Domestic Surveillance

The most crucial addition to the contract is a specific clause stating that ChatGPT-powered systems at the DoW “shall not be intentionally used for domestic surveillance of US persons and nationals.”
This is a direct response to the primary fear of critics. It attempts to draw a hard line between using AI for foreign threats and using it to spy on American citizens.

The Jail Vow: A Promise Against Unconstitutional Orders

Perhaps trying to underscore his personal commitment, Altman made a dramatic pledge: “If I received what I believed was an unconstitutional order, of course I would rather go to jail than follow it.”
While powerful rhetoric, critics might argue that this is a promise for a hypothetical future, not a solution for the inherent risks of the current partnership. It raises the question: would he even know if such an order was being carried out deep within a classified military program?

Beyond Politics: Is ChatGPT’s Quality Declining?

Interestingly, the Reddit discussions surrounding this event reveal a secondary, but important, layer of user dissatisfaction. Many users are not just angry about the military deal; they feel the product itself is getting worse.

Common complaints across various threads include:

• A perceived decline in the quality and reasoning of responses.
• Frustration over the recent retirement of the GPT-4o model.
• A general feeling that the platform is becoming less reliable as it scales and pivots to new markets.

This combination of ethical unease and product frustration creates a powerful incentive for users to explore alternatives like Claude.

Common Mistakes Users Make in a Crisis Like This

If you are considering leaving ChatGPT or changing how you use AI, avoid these pitfalls:

1. Mistake: Deleting your account immediately out of anger.
   Solution: First, download your data. Go to your account settings and request a data export. This ensures you don’t lose your valuable chat history and custom instructions.
2. Mistake: Assuming a new app is automatically more private.
   Solution: Read the privacy policy of any alternative, like Claude. Check if they have similar clauses or partnerships that might conflict with your values.
3. Mistake: Only looking at the headlines.
   Solution: Read the actual updates from the companies. As we saw, OpenAI added specific language to its contract. Understand the current facts before making a final decision.

Pro Tips for Navigating the AI Ethics Landscape

Here’s how to be a smarter, more secure AI user in light of these events:

1. Treat All Free Tiers with Caution: If you aren’t paying for the product, you are the product. Be mindful of what you share on any free platform, as data usage policies can change.
2. Follow the Developers: Anthropic showed that walking away from a bad deal is an option. Pay attention to the actions of AI companies, not just their mission statements. Which companies are willing to lose a contract to uphold their principles?
3. Diversify Your AI Toolkit: Don’t rely on a single AI model. Use ChatGPT for some tasks, Claude for others, and experiment with open-source models. This reduces your dependency and gives you a broader perspective.

Frequently Asked Questions (FAQ)

1. Is my personal data on ChatGPT now being shared with the US military?

Based on OpenAI’s recent statement, the agreement now specifically prohibits the use of ChatGPT for domestic surveillance of US persons. However, the full scope of data usage in other contexts is not publicly detailed. If you are concerned, you should assume any data you input into a cloud-based AI is not fully private.

2. What is the Department of War? I thought it was called the Department of Defense.

The Department of War was the original name for what is now the Department of Defense (DoD). The use of the older, more militaristic name in the TechRadar article and by some commentators highlights the historical weight and aggressive connotations that critics associate with this partnership.

3. How do I permanently delete my ChatGPT account?

Go to chat.openai.com, log in, and navigate to your account settings. Look for the “Delete account” option in the data controls section. Remember to export your data first if you want to keep your history.

4. Is Claude really a more ethical choice?

Claude’s developer, Anthropic, has built its brand around “Constitutional AI” and safety. Their decision to walk away from a DoW deal due to a lack of safety assurances supports this reputation. For many users, this action speaks louder than words, making Claude the current favorite for ethically-concerned users.

Conclusion: A Defining Moment for Consumer AI

The 295% surge in ChatGPT uninstalls is more than just a bad week for Sam Altman. It is a clear signal that a significant portion of the user base cares deeply about how AI is developed and deployed. The “sloppy” handling of the Department of War deal has crystallized a growing divide between rapid commercialization and responsible innovation.

For now, the ball is in OpenAI’s court. Will these contract amendments be enough to win back trust, or is the damage done? And for users, the path forward is clearer than ever: your choices matter. By moving to platforms whose actions align with your values, you are helping to shape the future of ethical AI.

Actionable Next Step:
Take a moment to review your own AI usage. Are you comfortable with the policies of the platforms you use most? If you have concerns, take control by exporting your data, researching alternatives like Claude, and making an informed choice. The future of technology shouldn’t just be built for us, but with our consent.