Sam Altman Addresses GPT-5 Rollout Challenges and Enhancements

OpenAI's CEO, Sam Altman, recently provided updates on the GPT-5 rollout, acknowledging initial challenges and detailing forthcoming improvements.

 

Altman announced plans to double GPT-5 rate limits for ChatGPT Plus users as the deployment progresses, aiming to enhance accessibility and performance.

 

In response to user feedback, Plus subscribers will have the option to continue using the previous model, GPT-4o. OpenAI will monitor usage patterns to determine the duration of support for legacy models.

 

Addressing concerns about GPT-5's performance, Altman explained that a malfunction in the autoswitcher led to suboptimal responses. He assured users that the issue has been resolved, resulting in a smarter and more reliable GPT-5 experience.

 

To improve transparency, OpenAI will implement features that clearly indicate which model is responding to user queries, ensuring clarity in interactions.

 

Additionally, the user interface will be updated to make it easier for users to manually trigger the model's "thinking" process, providing more control over response generation.

 

Altman acknowledged that the rollout is taking longer than anticipated due to the scale of the changes, noting that API traffic has approximately doubled in the past 24 hours. He emphasized OpenAI's commitment to stabilizing the system and attentively listening to user feedback.

 

These updates reflect OpenAI's dedication to refining GPT-5 and enhancing the overall user experience.