Is AI a bubble waiting to burst...
My idea for this article stemmed from the disappointing release of GPT-5, which underwhelmed many in the tech world, in addition to reading MIT’s article, which found that 95% of enterprise AI pilots are failing to produce financial returns.
Sam Altman’s Statement
Recently, Sam Altman, CEO of OpenAI, stated that he believes we are in an AI bubble, specifically comparing the current hype surrounding AI to the dot-com bubble of the late 1990s (1995-2000), which resulted in losses of up to 80% in high-value tech companies. Ultimately, people became overexcited by everything related to it, and web startups were launching left, right and centre, with infeasible and unsustainable business models.
“Are we in a phase where investors as a whole are overexcited about AI? My opinion is yes. Is AI the most important thing to happen in a very long time? My opinion is also yes,” said Sam Altman.
Arguments for the AI Bubble Theory
Overinvestment – lots of investment is being poured into AI startups with little to no revenue, with the valuations being based on hype, potential and in an attempt to launch quicker than competitors. OpenAI is valued at $300 billion (March 2025), yet other smaller AI companies, such as the French AI firm Mistral, are valued at around £13 billion, yet have similar capabilities. Try out Mistral here.
Overpromising – with the release of GPT-5 being disappointing after lots of anticipation built up over social media. OpenAI acknowledged problems, including glitches and a problem with the model’s routing system, causing GPT-5 to perform poorly on certain tasks. In some cases, previous models performed far better than the new GPT-5.
History repeating itself – many companies are racing into AI without viable business models which are sustainable long-term. Read about the dot-com bubble here.
Arguments against the AI Bubble Theory
Widespread adoption – 52% of businesses in the UK now use AI, with an impressive 92% of companies reporting increased revenue due to AI (according to the AWS report). For example, think about the online services you use daily – can you think of any which don’t have some AI feature? No, I thought not!
Infrastructure Investment – Not just big tech companies, but various different companies are betting trillions on the development of their own AI-ready infrastructure, although many companies decide to rent infrastructure from Amazon or Google.
Real gains – unlike the dot-com bubble, where companies didn’t have working products, only promises, and the ecosystem wasn’t ready (think dial-up internet, very few people online and expensive device & internet costs); however, now we have AI tools, which are resulting in real gains in industries such as software engineering, healthcare, finance, and logistics, and now in 2025, the technology is already adopted, and anybody can use these tools from their phone.
Conclusion
In conclusion, while there are certainly signs of hype and overvaluation, reminiscent of the past dot-com tech bubble, these developments in AI are resulting in real, measurable gains for companies and individuals. The already widespread adoption of the internet and investment in more AI infrastructure provide market readiness for AI growth, and demonstrate a shift in the way businesses are operating, indicative of AI being a lasting and revolutionary change.
🔗 Visit my website: www.samuelforrest.me