AI compute is the new oil–and needs a new market

The discussion centers on whether AI compute is "the new oil" and what market or policy changes that would imply. vannevar rejects the analogy, arguing compute efficiency improves rapidly and predicting AI costs will fall to near the price of electricity while blaming data-center demand on pandemic-era private-credit machinations. simoncion disputes vannevar's electricity-is-dominant-cost claim with a Hetzner-based cost breakdown that puts hardware depreciation well above power. vatsachak counters that AI is essentially electricity and is building systems that reflect that view. j16sdiz and vdfs treat AI compute as a massive infrastructure project with regulatory capture and huge capex implications, while Avicebron and jordemort warn that usefulness or environmental/toxic externalities must be resolved before equating compute to oil. raziel2701 and kylehotchkiss place the claim in historical context, comparing it to past "new oil" metaphors like Big Data and crypto.

Opinion also divides on commercial outcomes and cultural effects. vinyl7 expects firms to monetize relentlessly; userbinator sees an emerging ecosystem of consumer-facing "vibecoded" projects and better YouTube content driven by AI, whereas byzantinegene and creata find those outputs hollow or want concrete examples. BatchJob and others lampoon AI news saturation. Overall commenters dispute both the economic mechanics and the social consequences of treating AI compute as a commodity akin to oil.