Microsoft founder Bill Gates has warned that tons of investments in artificial intelligence (AI) will be “dead ends”. During an appearance on CNBC’s “Squawk Box”, Gates said that his opinion is not a product of pure speculation. Stating that we are in the middle of an AI bubble, he pointed to the example of “tulip mania” during the 1630s in the Netherlands. During this period, the price of tulips soared for several years only to suddenly crash. “That’s not where we are,” Gates said. Instead, he compared the AI bubble to the dot-com era when several internet-based companies became overvalued, leading to a significant crash.“In the end, something very profound happened. The world was very different,” Gates said. “Some companies succeeded, but a lot of the companies were kind of me-too, fell behind, burning capital companies.”

“Absolutely, there are a ton of these investments that will be dead ends,” Gates added.

Gates calls AI ‘biggest technical thing ever’

During the conversation, Gates said that AI has been “the biggest technical thing ever in my lifetime.”“The value is extremely high, just like creating the internet ended up being, in net, very valuable,” Gates said. “But you have a frenzy. And some of these companies will be glad they spent all this money. Some of them, you know, they’ll commit to data centers whose electricity is too expensive.”

95% companies have seen zero return on their AI investments, says study

Gates’ views have a stark similarity with a Massachusetts Institute of Technology (MIT) study that claimed that 95% of organizations are getting zero return, despite $30–40 billion in enterprise investment into GenAI. MIT published a report in September this year titled ‘The GenAI Divide: State of AI in Business 2025’, which said that only 5% of integrated AI pilots are extracting millions in value, while the vast majority remain stuck with no measurable P&L (profit & loss) impact.The study says that more than 80 % of organizations have explored or piloted tools like OpenAI’s ChatGPT and Microsoft’s Copilot. Nearly 40% reported deployment of these tools. However, these AI tools, as per the study, enhance individual productivity primarily with no P&L performance. “Most fail due to brittle workflows, lack of contextual learning, and misalignment with day-to-day operations,” the study highlighted.