Wall Street research and advisory firm Gartner stands by its prediction that $3.5 trillion in AI investment could stall due to concerns about the profitability of top AI models.
Gartner’s predictions released last week helped send shockwaves through the AI industry. John Lovelock, the company’s chief forecaster, told CNBC on Monday that the earnings don’t yet support a serious investment size.
He added that the timing may also be inconsistent, leading to mergers and acquisitions among AI specialist companies known as neoclouds.
“Right now, we have more computing power than demand, and freemium solutions are offering a lot of tokens and processing power,” he told CNBC. “Everyone in the AI race is running to be the last man standing. Almost every market Gartner tracks in the AI race is likely to collapse into a duopoly.”
The major commercial frontier model duopoly consists of Anthropic and OpenAI. The asset base of computing power that Neocloud is building to serve that duopoly requires some cash flow for businesses that haven’t yet scaled to justify the investment, Lovelock said.
Lovelock reiterated that while NeoCloud won’t go out of business all together, it could become the target of consolidation, such as an acquisition or merger.
After analyzing its profitability, Gartner last week predicted that 80% of AI-dedicated computing power provider NeoCloud could cease to function as an independent company.
NeoCloud took a big hit last week, and the rest of the stock market fell along with it.
coreweave decreased by 25%. Nevius decreased by 25%. and Airen It fell 16%. of S&P500 It ended down 2%. Nasdaq Composite It fell 4% in the week ending October 16th. Dow Jones Industrial Average It ended up 0.7% as investors moved away from technology stocks.
Coreweave was up about 2% in midday trading Monday, Nevius was up 3.3% and IREN led by 5.6%.
Core weave day 5.
Wolf analysts said in a note Sunday that the Gartner report likely contributed to market stress.
“By the end of the day (Wednesday), the market was already looking tense with the release of a Gartner report warning that more than $3.5 trillion in AI infrastructure spending could be exposed to impairment by 2030, with neo-cloud providers particularly vulnerable,” Roth’s Stephanie Wolfe said in a letter to clients.
Lovelock said Monday that AI is a technology, not a market, and likened it to salt that companies sprinkle on existing commercial activities rather than creating essentially new areas to sell differentiated and competitive products.
Gartner researchers found in a report on Wednesday that overall, between 2027 and 2030, “AI revenues will be insufficient to benefit from all planned AI capital investments.”
The findings come as frontier algorithm company Anthropic’s revenue growth slows, falling from 140% growth in the second quarter to 45% growth in the third quarter.
Considered by many tech analysts to be one of the cornerstones of the AI industry, Anthropic’s $16.8 billion in quarterly revenue is dwarfed by the more than $5 trillion hyperscalers are expected to invest by 2030, but the pace of its investments is increasing.
Wolfe’s Roth noted that OpenAI’s annual revenue forecast was $20 billion lower last week.
Gartner is not alone in warning that investors’ calculations about building AI may be wrong. Chris Wood, global head of equity strategy at Jefferies, has been warning about AI-driven capital destruction since at least September. Prominent investor Paul Tudor Jones told CNBC last week that 2027 will be a difficult year for the stock market given the macro economy.
“If you add up the buybacks, the M&A, all the positive things that have given this market so much power…since 2009, for example, if you add all of that up and subtract from it the IPOs, secondaries, unlocks, etc., you have arguably the most negative flow situation since 2009,” he said.
