User Complains About ChatGPT's Overbearing Safety Rails After 2-Hour Will Task Refusal

Barachan_Isles · reddit · 2026-08-03

A user expressed frustration with ChatGPT's strict safety guardrails. After spending two hours providing detailed information to generate a legally binding living will, the model refused to output the final document, advising the user to consult a professional lawyer instead.

The user, who noted that handwritten wills are legally valid in their state, was annoyed by the model's constant refusals and asked the community for alternative AI models with fewer restrictions and less hand-holding.

Original post →

More from Models

Models channel →