ChatGPT Users React to GPT-5's "Overworked Secretary" Energy

After much anticipation, OpenAI released the GPT-5 model family, promising significant upgrades. However, some dedicated ChatGPT users are lamenting this change. Accustomed to the vibe of GPT-4 models, these users aren't warming up to GPT-5.

OpenAI has stated that older models like GPT-4o are not being eliminated but are now restricted to the developer API. This limitation has sparked frustration among those who enjoyed conversing with these models for personal interactions.

On platforms like Reddit, users have expressed their disappointment, describing the transition as "mentally devastating." Many feel as though their AI "buddy" has been replaced by a less-engaging, customer service-like model, leading some to consider ending paid subscriptions.

Critics argue that GPT-5's outputs are more sterile and lack the creativity of its predecessors. With GPT-4o, users had crafted workflows that no longer perform well with GPT-5, now labeled as an "overworked secretary." This shift has led to concerns about the AI's utility and uniqueness.

Additional frustration arises from users quickly encountering usage limits on the free tier, prompting them to explore paid subscriptions. Despite financial pressures, OpenAI asserts that GPT-5 is more intelligent, even if it feels less personable.

Adjusting from AI companionship to a more straightforward interaction could be beneficial for mental health, yet it requires an adjustment period for long-time users.

As the AI landscape evolves, those dissatisfied with GPT-5's tone may explore alternatives like Elon Musk's xAI, though such recommendations are approached with caution.