ChatGPT Lost 2.5 Million Users in a Week

Two and a half million people, that's how many users, have now joined a movement to delete ChatGPT from their phones, cancel their subscriptions, and walk away from the most popular AI tool on the planet. And I need you to understand something. This is not some random internet outrage that's going to blow over by next Tuesday. ChatGPT uninstalls surged 295% in a single day.

One star reviews jumped 775%. And for the first time in history, a competitor, a company most people had never even heard of a month ago, dethroned ChatGPT as the number one app in the United States app store. That competitor is called Claude. And the story of how all this happened in the span of 72 hours is one of the most wildest things I've ever covered in this space.

So let me break down from the beginning, because the details matter here, and what's really going on underneath the surface is way bigger than just one company making one bad deal. This is about where AI is headed, who controls it, what they're willing to do, and what that means for you, your business, and for the whole world that we're all living in. So here's the timeline. On February 27th, opening night, the company behind ChatGPT announced that it signed a deal with the United States Department of Defense.

Now, the Pentagon had actually been working with another AI company first, a company called Anthropic, which makes AI assistant Claude. Anthropic had a $200 million contract with the military. They were already providing AI tools for classified operations. But here's where things went sideways.

The Pentagon came to Anthropic and said, we want to use your AI for all lawful purposes, no restrictions. Anthropic CEO, a guy named Dario Amodi, looked at that request and said no. Specifically, Anthropic drew two red lines, two things they would not allow their AI to be used for. Number one, mass surveillance of American citizens, and number two, fully autonomous weapons, meaning AI that can decide to kill someone without a human being in the loop making that decision.

Think about that for a second. A private company, an AI company, told the United States military, no, we will not let you use our technology to spy on Americans without a judge signing off on it. And we will not let you use our technology to build robots that hurt people without a human pressing the button. Now you might think, okay, that seems pretty reasonable.

You might also think, well, the military has its reasons and national security is complicated. Both of these things can be true at the same time. But what happened next is where this story gets absolutely insane. So the Pentagon gave Anthropic a deadline, Friday, agree to our terms or face the consequences.

Anthropic said no. Within hours, Defense Secretary Pete Hegseth declared Anthropic a supply chain risk. Now, what that basically means, a supply chain risk designation is something that the government has historically reserved for foreign adversaries. For example, Chinese companies like Huawei.

This was reportedly the first time in history this label was applied to an American company, technology company. It means federal agencies are ordered to stop using Anthropic's products. It means government contractors who work with Anthropic could lose their own government business. And then within hours of Anthropic being blacklisted, Sam Altman, the CEO of OPI announced that his company had signed a deal with the Pentagon.

The very same deal that Anthropic had refused. Altman claimed OpenEye's contract included the same red lines, no mass surveillance, no autonomous weapons. But critics immediately started pulling the contract language apart. And what they found was not so encouraging.

The contract said OpenEye's technology could be used for any lawful purpose. And here's the problem with that phrase. What counts as lawful can change. Department of Defense policies can be rewritten at any time.

Intelligence agencies have a long history of interpreting legal loopholes in the broadest possible way. Brad Carson, a former congressman and former general counsel of the United States Army, looked at the contract language OpenEye published and said publicly that he had reluctantly come to the conclusion that the provision preventing domestic surveillance doesn't really exist. His words were, I think they're just trying to fake it. And then it got worse for OpenEye.

The full contract hasn't been released to the public. OpenEye's own national security chief, Katrina Mulligan, first promised to release a clear and comprehensive explanation of the contract terms. Then she told a critic on social media, I don't agree that I'm obligated to share contract language with you. That is not a great look when you're telling 900 million users to trust you.

The public reaction was swift. It was massive. It was measurable. On February 28th, one day after the deal was announced, chatGPT mobile app uninstalls in the United States jumped to 5%.

To put that in context, the normal daily uninstall rate for chatGPT was around about 9%. It went from 9% to 295% overnight. Downloads dropped 13% on Saturday and another 5% on Sunday, reversing what had been a 14% growth trend. One-star reviews on the app searched 75-star reviews collapsed by more than 50%.

And here's a number that should terrify OpenEye's investors. Reports indicate that OpenEye lost approximately 1.5 million paid users in the first week alone. These are people paying $20 a month for chatGPT+, or $200 a month for chatGPT Pro. If you do the math on just the plus subscribers, that's potentially a $30 million or more in monthly recurring revenue that just walked out the door.

Meanwhile, what happened to Anthropic? The company that said no to the Pentagon and got blacklisted to it, well, on Saturday night, Claude searched to the number one spot in Apple's app store, dethroning chatGPT for the first time ever. Not just number one among AI apps, number one across all apps, period. More downloaded than Instagram, more than TikTok, more than chatGPT.

The numbers are staggering. Claude's daily downloads in the US hit 149,000 compared to chatGPT's 124,000. Anthropic said daily signups actually broke their all-time record every single day that week. Free users increased more than 60% since January.

Paid subscribers more than doubled since the start of 2026. Daily active users tripled from about 4 million at the start of the year to 11.3 million by March 2nd. Even Katy Perry got involved. The pop star posted a screenshot on X showing she had purchased a Claude Pro annual subscription with a big red heart drawn over it.

Her caption was one word, done. The post went viral, racking up 12.3 million views. Now, I need to pause here and tell you something important. I'm not going to tell you which company is right and which company is wrong.

That's not my job. My job is to give you the facts and numbers and let you make up your own mind. What I'm going to tell you is what this means for you and why you should be paying very close attention to what's happening because here's a bigger picture. chatGPT's market share has been declining for months and this Pentagon controversy just poured gasoline on fire that was already burning.

According to Apptopia, chatGPT's mobile app market share fell from 69.1% in January 2025 to 45.3% in early 2026. That's a massive drop. Google's went from 14.7% to 25% in the same period. Grok, Elon Musk's AI went from 1.6% to 15%.

Similar web shows chatGPT's overall market share dropped from 87% to 68% in just 12 months. The era of chatGPT being the only game in town is over and this is actually a good thing for you regardless of what AI tool you use because competition means better products, lower prices and more innovation. But what I find most fascinating about this whole saga is what it reveals about where the real power lies in the AI industry. It's not with the government, it's not entirely with the companies, it's with the users.

Two and a half million people threatened to leave and within days Samuel was on social media doing damage control, admitting the deal was rushed, admitting it looked opportunistic and sloppy and scrambling to renegotiate the contract terms. And it's not just the users. On Saturday morning activists had written messages in chalk on the sidewalk outside OpenAI's San Francisco offices. Where are your red lines?

You must speak up. What are the safeguards? The really telling thing is that those sentiments were being echoed inside the building. CNN reported that OpenAI employees were quote, venting about how leadership handled the Pentagon negotiations.

Many employees said they really respect Anthropic for standing up to the Pentagon and were frustrated with OpenAI's response. One research scientist at OpenAI, Aidan McLaughlin, posted publicly saying he personally didn't think the deal was worth it. And then just this past Saturday, March the 7th, OpenAI's head of robotics, Caitlyn Malinowski, resigned. She left the company specifically because of the Pentagon deal and a public statement was devastating.

She said, this was not an easy call. I had an important role in national security, but surveillance of Americans without judicial oversight and lethal autonomy, without human authorization, aligns that deserved more deliberation than they got. She clarified saying her issue was that the announcement was rushed without the guardrails defined. It's a governance concern first and foremost.

Malinowski was not some random employee. She previously led the team that built Meta's Orion augmented reality glasses. She spent nine years at Oculus building virtual reality headsets. She, before that, was designing and helping Apple design MacBooks.

This is a serious person with a serious career. And she walked away from one of the most coveted positions in AI because she felt the company moved too fast on something important. Now, let me zoom out even further because there's an economic story here that is completely missing in the coverage. OpenAI just raised $110 billion.

And here's the thing nobody is talking about. This fundraise was announced on the same day as the Pentagon deal, February 27th, the same day. OpenAI announced a $110 billion fundraise and a military contract on the same day and then watched 295% more users delete the app than ever before. Whatever the fundraise brought investor confidence, the Pentagon deal cost them in consumer trust.

And the money numbers in AI right now are absolutely mind-bending. In February 2026 alone, Global Venture Investment hit $189 billion, the largest single month on record. AI-related startups accounted for 90% of that, or $171 billion. OpenAI raised $110 billion.

Anthropic raised $30 billion. Waymo raised $16 billion. February 2026 was up 708% year-over-year from February 2025. Anthropic is now valued at with $380 billion.

Their revenue is growing at roughly 10x per year. They reportedly passed $19 billion in annualized revenue, up from $9 billion at the end of 2025 and $14 billion just a few weeks before that. OpenAI just crossed $25 billion in annual revenue. And some analysts project the two companies could reach parity sometime in 2026 or 2027.

So the main thing to note here is if you're not paying attention to right AI right now, you're going to be left behind. And I don't say that to scare you. I say that because the numbers don't lie. The AI market grew 708% in a single month.

Companies are raising $110 billion in a single round. And users are switching AI tools in the hundreds of thousands overnight. Now, let me bring you back to the story. One of the things that makes this situation so complicated is that both sides have legitimate arguments.

The Pentagon's position, stripped down to its essence, is that the United States military needs the best AI tools available, especially as competitors like China are rapidly integrating AI into their own military systems. A retired four-star general and former NSA director, Paul Naxone, who now sits on OpenAI's board, said publicly that we need Anthropic, we need OpenAI, and we need all of our large language models to be partnering with the government. And Anthropic's position, also stripped down to its essence, is that today's AI models are not reliable enough to be trusted with fully autonomous weapons, and that mass surveillance of American citizens violates fundamental rights. You can agree with one side or the other, or somewhere in between.

What you can't ignore is that this is the most important debate in technology right now. Who gets to decide what AI can and cannot be used for? The company that builds it, or the government that wants to use it? Or the users who rely on it every day?

And here's the twist that makes it even more complicated. Despite being blacklisted, Anthropic's AI was reportedly also used by the military. According to reports from the Wall Street Journal, the Defense Department used Anthropic's Claude via its Palantir contract to help plan operations, including in the recent military actions. So even as the government was publicly punishing Anthropic for refusing to cooperate, it was apparently still using Anthropic's technology behind the scenes.

So the story of the quit GPT movement and the Pentagon deal, and Anthropic going to number one, is not really a story about one company versus another. It's a story about a phase change. The AI industry has just moved from a phase where there was one dominant tool and everyone used it, to a phase where there are multiple strong tools and users have real power to shape the market within their choices. And this has never happened in AI before.

It's unprecedented. So the main thing that I would note here is that if you're someone who's just curious about all of this and wants to go deeper, just keep watching this channel, subscribe, because these things are changing all the time. The AI industry has just had its most traumatic week ever. Over 140 billion dollars raised, a Pentagon standoff, a mass user exodus, a number one app dethroned, an executive resignation on principle, and it's only March.

We are living through the fastest technology shift in human history and the people who understand what's happening and position themselves well are going to do incredibly well. The people who ignore it are going to wonder what happened. I don't want you to be in that second group, I want you to be in the first group. And everything that I do on this channel, everything we do in the AI Profit Boarding, which is my AI automation community to help you learn, scale, and save time with AI, is designed to help you get there.

So keep watching and if you haven't already, check out the AI Profit Boarding. Link in the comments description. It's my AI automation community that helps you save time, scale, and grow with AI automation. Thanks for watching, I'll see you on the next one.

Cheers, bye-bye.