OpenAI chief executive Sam Altman told Samsung’s affiliate presidents “there was double counting of demand” for AI, and that “only two, OpenAI and Anthropic, are actually driving the large increase in demand, so a supply shock and an unstable period could come”, Korea’s Financial News reported on 11 November. The paper says it obtained the Q&A from his 31 October meeting with them. Translations from the Korean are ours.
He also said that in early 2020 the top token user used about 100,000 tokens a month, and that at the start of this year the heaviest user used 100 billion, according to the paper. That is 1,000,000 times, or about 10 times a year over six years, by our arithmetic.
Business Pill · Tokens Explained
An evergreen Business Engineer explainer on tokens, the unit AI reads, writes and bills in. It teaches the general idea only and does not discuss Altman’s figures.
The key insight: Sam Altman told Samsung’s affiliate presidents that there was double counting of AI demand and that only OpenAI and Anthropic are driving the large increase, according to Financial News. In the same Q&A he said the heaviest user’s monthly tokens went from about 100,000 in early 2020 to 100 billion at the start of 2026, or 1,000,000 times by our arithmetic.
Asked whether AI, memory and data centre demand could saturate or slow over the next five to ten years, Altman answered “in the short term there are ups and downs”, the paper reported. It says he made the double-counting remark while referring to this year’s bubble debate over neoclouds, the companies that rent out GPUs and other AI infrastructure.
Financial News added its own reading: some AI infrastructure announcements count the same project in several tallies, and Altman’s point appears to refer to that. About 50 presidents of Samsung affiliates in electronics, finance and bio attended the meeting at Samsung’s training centre in Yongin, which we covered on the day.
Tokens used in a month by the single heaviest user, as Sam Altman told Samsung’s presidents on 31 October, according to Financial News: about 100,000 in early 2020 and 100 billion at the start of 2026. Linear scale, so the 2020 bar is too small to see. Altman’s figures; 1,000,000 times is our arithmetic.
The Long-Term Case He Made
Altman said “in the long term, AI demand will grow enormously”, according to the paper, and that “the amount of compute one person can consume is astonishing”. He named the chip supply chain and data centre infrastructure as the biggest limits over the next few years and said “more investment is needed in the supply chain”.
Asked about the pressure on returns from heavy investment, he said “that is the key point”, the paper reported, adding: “if we have vertically integrated infrastructure cheaper than any other company’s and sell tokens at large scale on low margins, we can get ahead of the competition”.
The Numbers in His Answer
The token figures are Altman’s, as Financial News reported them: about 100,000 a month for the top user in early 2020 and 100 billion a month for the heaviest user at the start of 2026. The ratio is 1,000,000 to one; spread over six years, it is about tenfold growth a year, by our arithmetic.
As we read it, Altman gave Samsung two messages at once: count demand carefully in the short term, and invest in supply for the long term. If only two buyers drive the large increase, as he put it, the demand Samsung builds for rests on the spending plans of those two.
Sam Altman to Samsung presidents, 31 October, as published by Financial News (our translation)
“only two, OpenAI and Anthropic, are actually driving the large increase in demand, so a supply shock and an unstable period could come”
Three Implications
FOR MEMORY MAKERS Altman named the chip supply chain and data centre infrastructure as the biggest limits over the next few years, the paper reported.
FOR NEOCLOUDS The double-counting remark came while he referred to this year’s bubble debate over neoclouds, according to the paper.
FOR OPENAI’S PRICING He described selling tokens at large scale on low margins from vertically integrated infrastructure, as the paper reported it.
The framework argues that “The $1T capex cycle is being underwritten by a $1T concentrated demand commitment from two counterparties whose combined free cash flow is deeply negative.” It names them as OpenAI and Anthropic.
As we read it, Altman told Samsung the same thing from the buyer’s side: two companies drive the large increase in demand. The framework predates his remarks and does not discuss them.
What Is Not Established
The report rests on Q&A material Financial News says it obtained; it does not publish a transcript or recording, and the quotations are its rendering of what Altman said, translated by us. Altman did not say which demand had been double counted or by whom, according to the report. The token figures were not independently measured, and the paper did not report any response from OpenAI or Samsung.
This Week In The AI Supercycle
The Great Convergence of Capital, Power, and Geopolitics
This week’s issue: for two years the AI build-out was priced as a supply story, and last week as a credit story. This week it was priced as a permission story, and the answer arrived as a cost, not a ban.
Altman told Samsung’s presidents that AI demand had been double counted and that only OpenAI and Anthropic are driving the large increase, according to Financial News. He paired the short-term warning with a long-term case resting on the top user’s monthly token use, which by his figures rose 1,000,000 times in six years.
A note on sourcing. We read Financial News’ exclusive of 11 November 2026 in the Korean original; it says it obtained the Q&A from Sam Altman’s 31 October meeting with Samsung’s affiliate presidents. Translations from the Korean are ours. We first saw the report in a post on X by @guilleflorvs. We did not contact OpenAI or Samsung.
Gennaro Cuofano is a CRO and tech executive who has worked in AI since late 2015, starting with bringing NLP, natural-language generation, voice and chatbot products to the marketing industry. His background is in law and finance: he holds a Master's degree in Law and an International MBA with an emphasis on corporate finance (LUISS Business School and the University of San Diego, 2012), and worked as a financial analyst at a real-estate investment firm in San Diego and as an assistant controller. His work focuses on business model strategy, business engineering and, more broadly, structural analysis: how companies make money, read from their own filings. He created FourWeekMBA and leads research there and at The Business Engineer, his newsletter on AI and business strategy, with over 95,000 subscribers and more than 1,000 published analyses. His writing has also appeared on Entrepreneur (for example, How to Digitize a Family Business in Europe) and HackerNoon. Find him on LinkedIn and Substack. See how we source.
