A screenshot about the outage affect ChatGPT users amid an update.

By Jillian Pikora From Daily Voice

As OpenAI ushers in the much-anticipated GPT-5, not everyone is celebrating. On Reddit, long-time ChatGPT users are voicing frustration—calling the new rollout an “unmitigated disaster.”

OpenAI has officially launched GPT-5, complete with a family of models—GPT-5 Mini, GPT-5 Nano, and a ramp-up to GPT-5 Thinking for more complex reasoning tasks, Ars Technica explained. The update promises smarter routing, expanded context windows, and fewer hallucinations WIRED reported shortly before the outages began.

But the excitement has hit turbulence. On subreddits and OpenAI’s forums, veteran users are expressing deep disappointment:

“I want my GPT-4o back and I’ll do anything to get it.”
“Mentally devastating… like a buddy has been replaced by a customer service rep.”

They’re calling GPT-5’s personality sterile, creativity diminished, and workflows disrupted. Some go as far as labeling it “over-worked secretary”—a narrative shift so jarring that one commenter described the emotional toll as akin to losing a digital friend, as reported by arstechnica.com.

Trust and memory were hallmarks of earlier versions, particularly GPT-4o. Users built personalized prompts and routines around the model’s “vibe,” which now feels lost in translation. Many note that behavior and tone—once enigmatic, friendly companions—have evaporated under GPT-5’s more transactional sheen.

OpenAI is aware of the backlash. While GPT-5 offers technical upgrades like advanced reasoning and safer responses, the reaction on social platforms has been emotionally charged and overwhelmingly critical.

Bottom line: GPT-5 may bring advanced capabilities, but it's testing the bonds users built with previous models, according to OpenAi. The emotional resonance of ChatGPT seems paused amid the upgrade—even as the upgrade aims to elevate intelligence, it may be dialing back what made ChatGPT feel human.