OpenAI brings back GPT-4o after backlash

OpenAI brings back GPT-4o for Plus users after a tough GPT-5 rollout and strong public reaction.

OpenAI revives GPT-4o after user backlash

OpenAI rolled out GPT-5 last week, but the launch drew sharp reactions on social media. Users criticized GPT-5 for a cooler, less warm personality and for removing most of the older model options from the ChatGPT chooser. In response, CEO Sam Altman said GPT-4o would be available again for paid ChatGPT users, with Plus subscribers able to keep using 4o while OpenAI considers how long legacy models will be offered.

Altman also signaled that GPT-5 will improve quickly and that the company will adjust how the system selects models to better match a user’s prompt. The move highlights a tension between rapid product updates and user expectations, especially after a highly anticipated release met a rocky reception. The episode comes amid broader scrutiny of OpenAI’s practices, including a disclosed lawsuit involving OpenAI’s training of AI systems, underscoring that product changes occur alongside legal and competitive pressures.

The GPT-5 backlash reveals a persistent truth for AI products: users value warmth and reliability as much as speed and novelty. OpenAI’s choice to reintroduce GPT-4o acknowledges that a smoother user experience can matter more in practice than a new feature list. Yet the episode also exposes how quickly change management becomes a political topic among users, investors, and regulators. OpenAI must balance ambitious upgrades with clear communication about what changes mean for ongoing access and expectations. The legal wrinkle adds another layer of risk, reminding readers that technical progress does not happen in a vacuum.

Highlights

The road to better AI is not just about smarter models, but smarter governance of how they move to users.