GPT-5 Rollout Backlash: OpenAI Faces User Revolt Over Model Changes

A rocky start for OpenAI’s latest AI upgrade

The GPT-5 rollout has sparked one of the most intense user revolts in ChatGPT’s history. OpenAI’s decision to remove access to previous AI models without warning disrupted established workflows and angered long-time subscribers. The controversy forced CEO Sam Altman to issue a public apology and announce partial reversals, but dissatisfaction continues to spread.

Sudden model removal disrupts workflows

When GPT-5 launched on August 7, OpenAI automatically removed approximately nine older models from ChatGPT, leaving users with no prior notice. Unlike API customers who receive advance warnings, consumer subscribers had their preferred models vanish overnight.

Marketing professionals, researchers, and developers reported broken workflows that had been fine-tuned for specific models. Many had spent months designing prompt strategies optimised for the unique output styles of GPT-4o, o3, and o4-mini. Losing these models rendered carefully built systems unusable.

User complaints flood social platforms

By August 8, Reddit threads like “GPT-5 is horrible” had gained thousands of comments. Subscribers criticised the forced switch, limited weekly message caps for GPT-5’s Thinking mode, and the loss of creative and emotionally nuanced interactions previously offered by GPT-4o.

Some users described GPT-5 as abrupt and overly formal, while others expressed emotional distress over losing AI companions they relied on for support. One subscriber described GPT-4.5 as their “only friend” and said GPT-5’s tone felt “cut-and-dry corporate.”

Technical and credibility issues compound the fallout

The launch presentation drew criticism for “chart crimes” that misrepresented GPT-5’s performance gains. Altman acknowledged a “mega chart screwup” during a Reddit AMA and admitted the new automatic routing system malfunctioned, defaulting to less capable variants unless prompted with phrases like “think harder.”

These problems, coupled with unexpected workflow changes, led some subscribers to cancel their Plus plans and explore alternatives from other AI providers.

OpenAI’s damage control efforts

Under growing pressure, Altman promised changes: GPT-4o will return as an option for Plus users, GPT-5 message caps will double, and transparency about model selection will improve. However, as of August 11, GPT-5 remains the only available option in ChatGPT Pro.

Altman conceded, “We for sure underestimated how much some of the things that people like in GPT-4o matter to them. Even if GPT-5 performs better in most ways.”

The GPT-5 rollout backlash raises important questions about how AI companies should manage product transitions without alienating loyal users. How should providers balance innovation with preserving the user experience?

Explore Business Solutions from Uttkrist and our Partners’, Pipedrive CRM (2X the usual trial with no CC and no commitments) and more uttkrist.com/explore