Sam Altman is addressing one of investors' biggest concerns about AI.
During a CNBC interview on Monday, the OpenAI CEO said questions about whether massive AI spending will ultimately pay off are legitimate, as companies continue pouring billions into infrastructure, chips, and software.
"So I think this is the most fair criticism right now of AI," Altman said. "You hear companies saying, I am spending a ton of money on AI. And I know some great stuff is happening, but I know there's a ton of waste."
He went on to summarize concerns he hears from businesses.
"How long do I have to wait for it to really show up in revenue, and how long do I have to wait to really get the costs under control?" he said. "I assume that the industry will figure that out pretty quickly, but I think that is a fair, a fair issue."
The comments come as investors increasingly scrutinize whether the AI boom can generate enough revenue to justify its staggering costs. According to an April report from The Wall Street Journal, OpenAI itself had missed some key targets for revenue and user growth last year.
Recent data from cloud optimization platform Cast AI suggests that many companies are paying for far more AI computing power than they actually use. In an analysis of 23,000 clusters across thousands of companies, the company found average GPU utilization was just 5%, meaning that roughly 95% of provisioned graphics-processing capacity sat idle.
Cast AI cofounder and president Laurent Gil said companies are often hoarding scarce AI chips simply because they fear missing out, not because they have an immediate need for them, resulting in a growing stockpile of underused computing resources.
Gary Marcus, a longtime AI researcher, author, and professor emeritus at New York University, has also said that the hyperscalers are committing unprecedented sums to AI without demonstrating returns that match the scale of those investments.
"Greatest capital misallocation in history," Marcus wrote in a post on X, describing some companies' AI capex plans and adding that Amazon, Google, Microsoft, and Meta collectively are spending more money than the Manhattan Project each month.
