Sam Altman said AI budgeting had recently become a “huge issue” for some companies \u2014\u00a0and it sent AI bubble watchers and doomers into a frenzy.<\/p>\n
During a Tuesday enterprise event<\/a>, Altman referenced memes like, “My company spent my entire 2026 budget in Q1.” <\/p>\n “That went from, at the beginning of this year, an issue that never came up \u2014 people were totally happy with the amount they were spending \u2014 to all of a sudden, a huge issue,” Altman said.<\/p>\n The reader response was loud. Some said that it was a warning of dark times or a failure in AI business model. The word “bubble”<\/a> came up often. Others said it was par for the course, a normal stage as people learn what to actually spend their tokens on after a period of experimentation.<\/p>\n Commentators from Gary Marcus to Michael Burry<\/a> got involved. Here are some of the most interesting reactions.<\/p>\n Some say it’s a dark warning<\/p>\n Ed Zitron, one of the internet’s foremost AI bubble warners, wrote on X<\/a> that OpenAI was “absolutely cooked.”<\/p>\n “This is loser language,” Zitron wrote. “You can’t be four years into the bubble saying ‘yeah our customers have a huge issue with how expensive our business is.’ You just raised $122 billion!”<\/p>\n Programmer Eric S. Raymond (often referred to as ESR) agreed that<\/a> the “bubble is popping.”<\/p>\n “Make no mistake, it’s a hugely useful technology and uptake will continue, even accelerate,” Raymond wrote. “But the overinvestment in datacenters that we’ve been seeing is not sustainable; the business model of the big providers doesn’t work, and is floating on VC money.”<\/p>\n Academic and author Vivek Wadhwa wrote that<\/a> it seemed like AI researcher Gary Marcus<\/a> was right: “the AI revenue models are imploding.”<\/p>\n Marcus himself commented that the death of tokenmaxxing<\/a> was “potentially a very serious issue for all three big IPOs.”<\/p>\n Michael Burry, the “Big Short” investor who’s taken an AI-skeptical turn<\/a>, also referenced the story on X<\/a>.<\/p>\n Others say it’s token misuse<\/p>\n As the tokenmaxxing frenzy dies down<\/a>, some engineers are wondering: are we spending in the right ways? Altman’s comments may be less of a dark warning and more of a rational cost-benefit analysis.<\/p>\n Google software engineer Patrick Toulme commented that<\/a> “getting value out of agents is still too difficult for most engineers, so they end up just burning tokens.”<\/p>\n “80% of the economic value of LLMs come from 20% of the tokens,” BCA Research’s chief strategist Peter Berezin<\/a> wrote. “There is a long tail of dubious token usage that can be greatly curtailed without much negative impact on productivity.”<\/p>\n Kun Chen worked at Meta, Microsoft, and Atlassian<\/a>. He wrote that AI spending was “driven by FOMO,” so some cutback was “inevitable.” <\/p>\n