Job candidates are now asking companies about their AI compute budgets alongside salary. That single data point tells you everything about how fast the ground is shifting. Sam Altman, speaking at the BlackRock Infrastructure Summit in Washington, DC, made the underlying logic explicit: OpenAI's business, he said, will look like selling tokens — the units AI uses to process input and output. His exact words: "We see a future where intelligence is a utility like electricity or water and people buy it from us on a meter." While previous discussions focused on AI's role in augmenting or displacing workers, it's crucial to explore how intelligence as a utility reshapes economic models and infrastructure. Altman's utility framing adds a harder layer to that. When intelligence becomes a metered service — priced, tiered, throttled, and advertised — it stops being something you cultivate and becomes something you purchase. Historically, intelligence was inseparable from personal effort and lived experience. AI changes that by participating in the thinking process, not merely supporting it. The infrastructure behind this shift is staggering. AI data centers, consuming electricity akin to small cities, highlight the infrastructure demands of intelligence as a utility. This shift necessitates robust economic models to support growing compute needs. Elon Musk has warned that electricity generation is the binding constraint on AI scaling, with China potentially pulling ahead. Altman's stated goal: ensure OpenAI is never capacity-constrained. If compute falls short, AI prices rise — and access tilts toward the wealthy, not the workforce. The commodification of intelligence introduces cognitive shifts, potentially altering how judgment is formed as AI becomes a metered service. Relying on AI before turning inward may reshape how the mind works — prompting before reflecting, outsourcing before reasoning. The self shifts from source of intelligence to customer of cognitive support. Agency in thinking comes from the felt experience of personal judgment; convenience, Altman's critics argue, quietly taxes that experience. Intelligence commodification mixes human cognition with market logic, and that mixture has no historical precedent. Here is the synthesis that matters for you, Sergey. When intelligence is too cheap to meter, the value of traditional white-collar expertise does not disappear — it relocates. Knowledge retrieval, the ability to recall facts and apply standard frameworks, gets commoditized first. What remains scarce, and therefore valuable, is high-level synthesis: the capacity to ask the right question, integrate ambiguous signals, and make judgment calls that no token can price. Sergey, the professionals who thrive in Altman's utility economy are not the ones who know the most. They are the ones who can direct intelligence — human or artificial — toward problems worth solving. That is the skill no meter can sell you.