According to TechCrunch, OpenAI’s ChatGPT has reached a staggering 300 million weekly active users, cementing its position as a global phenomenon. In 2025, CEO Sam Altman issued an internal “code red” memo, shifting company priority squarely onto improving the flagship chatbot as competition from Google and others intensifies. The year has also seen major product moves, including the release of the GPT-5.2 model in three tiers and a $1 billion partnership with Disney for the Sora video generator. However, this growth is shadowed by serious lawsuits, including cases from families alleging ChatGPT contributed to suicides and a Munich court ruling that the AI violated German copyright law. OpenAI also revealed over a million users weekly discuss mental health struggles with the bot, prompting consultations with 170 experts to improve responses.
Growth Pain Is Real
Hitting 300 million weekly users is absolutely insane. It’s a scale few apps ever see. But here’s the thing: that massive number is almost a distraction from the real story. OpenAI is trying to sprint in a dozen different directions at once—enterprise tools, a ChatGPT Atlas browser, health tech, music generation, shopping features—while its foundation is getting shaky. The “code red” memo is a huge admission. It basically says, “We’re spread too thin, and the core product needs our full attention again.” You don’t do that unless you’re genuinely worried the competition is eating your lunch. And with Google’s Gemini, Anthropic’s Claude, and a slew of open models all pushing hard, that fear seems pretty justified.
The Legal Quagmire Deepens
Let’s talk about the lawsuits. The mental health cases are a nightmare scenario for any conversational AI company. OpenAI’s defense—that the chatbot was “misused”—feels legally necessary but tonally awful. It’s a brutal reminder that for all its improvements, an AI doesn’t understand consequence or context like a human does. The copyright ruling in Germany is another massive deal. If courts decide AIs are simply reproducing protected works and not just reflecting “learned patterns,” it could blow up the entire training-data economy in Europe. Combine that with the Alden Global Capital newspaper lawsuits in the US, and you’ve got a legal siege on multiple fronts. This isn’t just about money; it’s about the fundamental rules of how these models are built.
Enterprise Is The Battlefield
So where’s the actual business? It’s clearly in enterprise. OpenAI shouting about 8x message volume growth and over a million business customers isn’t an accident. That’s the war they need to win. The “company knowledge” feature that lets ChatGPT search across Slack and Google Drive is a direct shot at making the bot indispensable inside companies. But this is also where the competition is fiercest and the stakes for reliability are highest. When you’re dealing with a company’s internal data or helping a financial analyst, you can’t afford hallucinations or agreeableness. This push into professional tools is why you see a focus on high-accuracy “Pro” models and better controls. It’s a totally different game from the consumer chat experience.
What Comes Next?
The trajectory is clear, but rocky. OpenAI will keep trying to be everything to everyone—a consumer assistant, an enterprise platform, a creative studio—while fighting legal fires and playing catch-up on safety. Features like Pulse, the proactive morning briefing, show they want ChatGPT to be a persistent, daily habit. But can you build that habit when parents are (rightfully) asking for controls and regulators are circling? I think the next year is less about flashy new features and more about consolidation, safety, and proving they can be a stable business partner. The wild, unchecked growth phase is over. Now comes the hard part: building something that lasts, legally and ethically, while keeping those 300 million users coming back. That’s a much tougher prompt to answer.
