Or will it? Mustard’s Andrew Wiseman looks at recent disruptions in AI and similarities to the dot-com boom.
I’m vaguely in the West Midlands today, so what better way to start with a tip of the hat to Stourbridge’s finest PWEI?
I read with interest about the impact of DeepSeek at the weekend, and the resulting fallout on the financial markets in the US. You may have read about Nvidia, who lost $600bn of its market value in a day, Yes, you read that right. $600 BILLION. Tech billionaires were reported to have individually lost a total of $100bn in the same day. This got me thinking of similarities between the current AI surge and the dot-com bubble in the late 1990s. The whole era was defined by speculative investments and unrealistic market valuations. Webvan or Boo.com anyone?
To set that context, Webvan filed for bankruptcy in 2001 with losses of over $800m. Boo.com spent over $135m to establish itself as online fashion retailer, but entered liquidation in 2000. These stories demonstrate the risks associated with jumping on the latest tech bandwagon without a credible business plan.
It’s exactly what we’re seeing with AI right now. At a best guess, there were around SEVENTY THOUSAND AI companies in 2024, with 17,500 startups in the US alone. Industry estimates suggest that up to EIGHTY FIVE PERCENT of these startups will fail within three years. Indeed, this has already started, with the failure of Metigy, an AI marketing startup valued at $1bn. This was a slightly unique case, with extensive financial mismanagement and fraudulent activities by their CEO. We can add AllHere Education to the list too. An investor defraud of $10bn forced the company into liquidation.
So, clearly, AI is doomed, right? WRONG. When we look at the dot-com boom of the 1990s and the current surge in AI, there’s one important difference. The dot-com boom was characterised by numerous startups with unproven track records. This time, the agenda is being set by the behemoth tech giants across the world, leaving them far less susceptible to the market volatility and potential downturns.
The emergence of DeepSeek certainly gives us all questions on our own AI journeys. It’s an industry still in rapid flux. Technology and use cases continue to evolve, and providers and users need to keep up. At Mustard, we’re continually trialing the latest ‘restech’. Sometimes we think it’s the best thing since sliced bread. Then a couple of months later, something better comes along.
These are exciting times, and there’ll be winners and losers along the way. One thing’s for sure, AI is here to stay, and it’s up to us as adopters to cautiously evaluate developments to increase our efficiency and create better insights.
