Source: Marcus on AI
OpenAI is building systematic monitoring capabilities into its product ecosystem—tracking user interactions, flagging suspicious activity, and building behavioral profiles. This marks a departure from its original positioning as a trustworthy AI research organization. The shift creates perverse incentives: the company now profits from engagement data and user metadata alongside API fees, putting it in direct structural conflict with user privacy and independent safety research. OpenAI has also abandoned the open-source community guardrails it once championed, replacing them with closed proprietary systems where transparency is impossible and only OpenAI determines what constitutes acceptable use.