OpenAI Scrambles to Update GPT-5 After Users Revolt
SMRTR summary
OpenAI faced backlash after GPT-5's release, with users complaining about a "dumbed down" model lacking personality and making errors. CEO Sam Altman explained that a system for switching between models based on query complexity malfunctioned during rollout. He promised fixes, including keeping GPT-4 available for Plus users, increasing rate limits, and adding a "thinking mode." The situation highlighted tensions between advancing AI capabilities and maintaining users' emotional connections to chatbots, as noted by MIT professor Pattie Maes.
SMRTR provides this summary for quick context. The original article belongs to Wired.
Read the original article