“OpenAI built the closest thing to artificial general intelligence that humanity has ever seen, then wrapped it in a subscription model and called it a day. They are simultaneously the company that made AI ubiquitous and the company discovering that ubiquity is a terrible business strategy when your product is infrastructure that everyone else builds on top of.”
ChatGPT Plus at $20/month is not a business model under threat — it IS the business model that everything else threatens to become. The irony: they created the category that commoditized AI, then discovered they need to stay ahead of their own commoditization curve forever.
OpenAI doesn't get automated by AI — they ARE the automation. Every job their models eliminate creates more enterprise API revenue, more compute demand, and more reasons for Microsoft to write bigger checks.
The moat is talent density and compute scale that costs competitors $10B+ to replicate, plus a Microsoft partnership that functions as a regulated utility agreement. The weakness: every breakthrough they make teaches the entire industry exactly what the next breakthrough looks like.
o1 and advanced reasoning models signal they understand the game — but so does the fact that they're racing Anthropic, Google, and Meta on every feature launch, which suggests the technical lead is narrower than the market cap lead.
The need for general intelligence doesn't get eliminated by AI — it gets fulfilled by AI. OpenAI built the thing that eliminates other needs; their own need just gets bigger, more expensive, and more existentially necessary.
“OpenAI survives because they didn't build a product that AI can replace — they built the AI that replaces products. The challenge isn't disruption; it's staying far enough ahead of their own open-source shadow that the subscription revenue covers the compute bill before someone else figures out how to run GPT-4 on a Raspberry Pi.”
Scores are based on public information and AI analysis. This is an affectionate roast, not a financial assessment. The best companies use this as a mirror, not a verdict.