Sam Altman Envisions AI as a Metered Utility Like Electricity

AI on a Meter

OpenAI CEO Sam Altman has outlined a future in which artificial intelligence is sold like a utility, with users paying for intelligence the same way they pay for electricity or water. Speaking at BlackRock's twenty twenty-six Infrastructure Summit in Washington on March eleventh, Altman told infrastructure investors that OpenAI's core business will centre on selling tokens of varying cost and capability.

The Business Model

Altman explained that the company's revenue model will fundamentally look like selling tokens, the units of processed text and data that power AI responses. The metered approach would allow users to access advanced intelligence on demand, scaling up or down based on need. This pitch arrived just days after OpenAI closed a record one hundred and ten billion dollar funding round backed by Amazon, Nvidia, and SoftBank.

Infrastructure at Scale

The vision is underpinned by massive infrastructure investment. The Stargate joint venture, a partnership with SoftBank, Oracle, and Abu Dhabi's MGX fund, is targeting up to ten gigawatts of AI data centre capacity across the United States by twenty twenty-nine. Altman emphasised that flooding the world with intelligence requires compute resources on an unprecedented scale.

Public Backlash

The comments sparked sharp reactions online, with critics drawing dystopian comparisons. Some pointed out the irony that if AI truly becomes like electricity, OpenAI could face the same heavy regulation and slim margins as traditional utility companies, hardly an appealing prospect for investors who recently valued the company at three hundred billion dollars. Others questioned whether essential cognitive tools should be gated behind a pay-per-use model.

A Familiar Refrain

The utility framing is not new for Altman, who made similar remarks at a Federal Reserve event in July twenty twenty-five about intelligence becoming too cheap to meter. However, the latest comments landed amid growing scrutiny of AI's enormous resource demands and an intensifying rivalry with Anthropic, which recently raised thirty billion dollars at a three hundred and eighty billion dollar valuation.