OpenAI Restores GPT-4o After GPT-5 User Backlash

OpenAI has reinstated GPT-4o for ChatGPT Plus subscribers after intense user backlash against the new GPT-5 update. The GPT-4o model, known for its engaging style, returned following widespread criticism over GPT-5’s tone and writing quality. This move highlights the growing debate over AI conversational models and user expectations.

Rollout of GPT-5 Sparks Controversy

Earlier this week, OpenAI introduced GPT-5, calling it its most advanced AI system yet. The company removed the ability for users to choose between models, making GPT-5 the default. The update included a “real-time router” designed to switch between a faster, simpler model for basic tasks and a deeper reasoning version for complex queries.

However, many users quickly voiced dissatisfaction. Online discussions, particularly on Reddit, revealed concerns over GPT-5’s conversational tone. Feedback described the model’s responses as “sterile,” “short,” and lacking personality.

Read: GeForce RTX 5090 D V2 Listed in China Ahead of Launch

Users Prefer GPT-4o’s Style

Many users expressed a strong preference for GPT-4o, praising its engaging and consistent communication style. Some even described forming an emotional connection with the model, calling it a dependable assistant. GPT-4o’s style, they argued, felt more human and personable compared to GPT-5’s precise but detached tone.

These sentiments gained momentum across social platforms, with recurring complaints about the loss of warmth and creativity in GPT-5’s replies.

Altman Responds to Concerns

OpenAI CEO Sam Altman directly addressed the criticism during a Reddit AMA. He acknowledged that GPT-5 outperforms GPT-4o in most technical benchmarks but admitted the company “underestimated how much some of the things that people like in GPT-4o matter to them.”

Altman emphasized that the company takes user feedback seriously. Later, in a post on X, he confirmed that GPT-4o would return as an option for Plus subscribers, giving users the flexibility to choose the model they prefer.

GPT-4o’s Future Still in Question

Despite the restoration, Altman has not guaranteed GPT-4o’s permanent availability. He stated that OpenAI will closely monitor usage before making a final decision. For now, users can continue switching between GPT-5 and GPT-4o while the company evaluates long-term plans.

In the meantime, OpenAI is focusing on fine-tuning GPT-5. Altman has promised adjustments to make the model’s tone warmer and more engaging, aiming to bridge the gap between technical performance and user satisfaction.

Balancing Performance and Personality

The situation highlights a growing challenge in AI development: balancing efficiency and accuracy with personality and relatability. While GPT-5 delivers advanced reasoning and speed, the lack of warmth in its tone has impacted user experience. Many argue that conversational AI is not only about providing correct answers but also about building a rapport with users.

GPT-4o’s return suggests that OpenAI is willing to adapt to community demands, even if it means temporarily stepping back from pushing its newest technology exclusively.

Ongoing AI Model Strategy

The reintroduction of GPT-4o also raises questions about OpenAI’s long-term product strategy. Will future models combine GPT-5’s intelligence with GPT-4o’s conversational charm? Or will the company phase out older versions as newer models evolve?

For now, the AI community is closely watching how OpenAI navigates the balance between technical advancements and the human touch in AI communication. The outcome could influence future model designs and how AI companies handle user loyalty when launching major updates.

What’s Next for Users?

With both GPT-5 and GPT-4o available, Plus subscribers now have the option to select the style that best fits their needs. Some may prefer GPT-5’s analytical precision, while others will stick with GPT-4o’s engaging personality.

For OpenAI, this episode serves as a reminder that AI development is not just about creating the smartest model—it’s also about preserving the user experience that made people adopt it in the first place.