Sam Altman is privately panicking. In the week that he was supposed to be celebrating ChatGPT’s third birthday, the OpenAI boss was facing a crisis on several fronts.
Just days after an investment report predicted that his company still won’t be profitable by 2030, OpenAI was caught up in a data breach that exposed the personal details of ChatGPT users. His lawyers were also forced to deny accusations that the artificial intelligence chatbot acted as a “suicide coach” for a teenager who killed himself earlier this year.
In an internal memo on Monday, Altman declared a “code red” – and not even for any of the reasons above. The tech boss told employees that the emergency situation was a result of the massive advances made by OpenAI’s rivals, which could threaten ChatGPT’s position as the world’s leading AI assistant.
Anthropic, DeepSeek and Meta have all made significant progress with their AI offerings, while Apple is also preparing to launch a revamped Siri early next year. Even Amazon’s AI assistant Rufus has seen success in recent days. But the biggest competitor comes in the form of Google’s Gemini.
When the search giant unveiled the latest version of Gemini in November, it was described as a “new era of intelligence”. Gemini 3 set record scores in benchmark tests, including the best ever result in Humanity’s Last Exam – designed by AI safety researchers to identify artificial superintelligence that matches or surpasses humans.
Publicly, Altman said it “looks like a great model”, and congratulated Google in a post to X. Privately, he acknowledged that ChatGPT was no longer the leading AI on the market. In a message to workers, he reportedly wrote: “We know we have some work to do, but we are catching up fast.”
ChatGPT users – there are now more than 800 million of them – are also realising that there are other options out there. Shortly after the release of Gemini 3, Salesforce CEO Marc Benioff said: “I’ve used ChatGPT every day for three years. Just spent two hours on Gemini 3. I’m not going back. The leap is insane.”
Gemini now counts around 650 million users, up nearly 50 per cent from 450 million in July. This is not even counting the roughly 2 billion people who use Gemini indirectly through the technology being integrated into Google services like Search. Figures from online analytics firm Similarweb show that users are also spending more time using Gemini than ChatGPT.
Losing users, or even just seeing growth stagnate, could prove catastrophic for OpenAI. To even approach the revenue figures projected in the recent investment report – the one claiming it would not be making any profit this decade – ChatGPT needs to see huge growth. The forecasts set out by HSBC Global Investment Research put the number of ChatGPT users in 2030 at 3 billion – nearly half of the world’s adult population. The report also predicted that subscription rates for premium ChatGPT models would rise to 10 per cent of users, though this seems increasingly unlikely if OpenAI falls behind its rivals.
Failing to hit this level of growth could see the collapse of OpenAI, and potentially most of the industry. The company that began blowing the AI bubble with the launch of ChatGPT in 2022 could be the one responsible for its deflation, leading to deserted data centres, trillions in lost wealth, and the risk of recession. If the bubble bursts, the blast radius could span the globe.
Google boss Sundar Pichai recently acknowledged in an interview a degree of “irrationality” in the current level of investment in artificial intelligence, warning that the fallout from any potential implosion would be unprecedented. “I think no company is going to be immune, including us,” he said.
In a financial stability report this week, the Bank of England warned that share prices in the UK are at their “most stretched” since the 2008 financial crisis. Governor Andrew Bailey likened the inflated valuations of tech companies in the US to the dotcom bubble.
“The governor’s parallel to the Dotcom bubble is spot on, but the reality is more insidious. We are funding a global gambling addiction with debt the real economy cannot afford, all concentrated in a handful of US giants,” Rohit Parmar-Mistry, an AI expert at business automation firm Pattrn Data, told The Independent.
“If this bubble bursts – and the stretched valuations suggest it will – it might finally flush out the hype merchants.”
There is still time for OpenAI to live up to the hype it has generated. And it may have a secret strategy to stay on top.
The company is currently working on a mystery device, and has recruited the man behind the iPhone, Jony Ive, to build it. Altman claims it will increase OpenAI’s value by $1 trillion and will ship “faster than any company has ever shipped 100 million of something before” when it launches next year.
Speculation surrounding what form OpenAI’s first hardware product will take has ranged from smart glasses to a clothespin. Should it be the latter, and fail as spectacularly as previous attempts, it could be the literal pin that pops the bubble.
Altman is yet to comment publicly about the internal memos or the unfolding turmoil (OpenAI did not respond to a request for comment from The Independent), but the head of ChatGPT, Nick Turley, has indirectly addressed the increased competition from Google and what OpenAI is planning to do about it.
“New products are launching every week, which is great – it pushes us to move faster and keep raising the bar for what an AI assistant can do,” he wrote in a post to X. “Our focus now is to keep making ChatGPT more capable, continue growing, and expand access around the world – while making it feel even more intuitive and personal.”
If OpenAI fails, and Google does become the most dominant in the AI space as it did with Search, then Altman’s company could go the way of AltaVista, Ask Jeeves and Lycos – an early pioneer now fondly remembered as one of the internet’s relics.