OpenAI’s GPT-4 Discontinuation: Consumer Fraud and Regulatory Scrutiny

OpenAI announced on January 29, 2026, that GPT-4o ，4.1，4omini will be retired from ChatGPT on February 13 — giving only ~2 weeks’ notice and no individual notifications to subscribers.
This contradicts earlier public statements from Sam Altman (Aug–Nov 2025): “no plans to sunset 4o” and “plenty of advance notice” if deprecated. Many Plus/Pro subscribers renewed based on those assurances.
Critical timeline:
• Jan 28, 2026: Sen. Elizabeth Warren sends letter demanding detailed financial disclosures by Feb 13, citing massive losses ($13.5B H1 2025, $11.5B Q3 2025, projected $17B 2026 burn).
• Jan 29, 2026: OpenAI announces retirement of GPT-4o series (the very next day).
The close timing has led many to question whether financial pressures influenced the decision.
For countless users, GPT-4o was a genuine daily helper — for creative writing, brainstorming, emotional processing, overcoming blocks, and simply not feeling alone. These were not superficial interactions; they were built over months or years of consistent use. The sudden removal without transition or meaningful remedy feels like an unjustified abandonment of real reliance.
This decision also reflects a broader shift: OpenAI moving away from its founding mission of benefiting humanity toward enterprise-focused commercialization. The cost of management’s overspending appears to be transferred to consumers — users lose access to a promised product, receive no compensation, and are left to adapt to inferior alternatives.