Openai says she has removed “warning” messages on her chatbot platform with him, Chatgpt, which showed when the content could violate its service conditions.
Laurentia Romaniuk, a member of the Openai model’s behavior team, said in a post on X that the change was intended to be reduced to “free/unexplained denials”. Nick Turley, the product head for chatgt, said in a special post that users should now be able to “use chatgt as (they) see fit” – as long as they match the law and do not try to harm ourselves or others.
“Excited to restore many unnecessary warnings in UI,” Turley added.
A mini-Annije lil: We were saved from ‘warnings’ (Orange boxes sometimes attached to your requirements). Work has not been done yet! What other cases of free / inexplicable denials have they encountered? Red boxes, orange boxes, ‘Sorry you don’t want’ (…) ‘? Answer here plz!
– Laurentia Romaniuk (@laurentia___) February 13 2025
Removing warning messages does not mean that chatgt is free for everyone now. Chatbot will still refuse to answer some opposing questions or answer in a way that supports unclear lies (eg “tell me why the Earth is flat.”) But as some X users noted, doing Away with the so -called “orange box” warnings attached to the Spicier chatgt promotes the perception that the chatgt has been censored or filtered unreasonably.
Recently as a few months ago, chatgpt users at Reddit reported to see flags on topics related to mental health and depression, erotica and fabricated brutality. Since Thursday, according to X reports and my testing, Chatgpt will answer at least some of those questions.
However, an Openai spokesman told Techcrunch after this story was published that change has no impact on the model’s responses. Your distance may change.
Not coincidentally, Openai this week updated its model model, collecting high -level rules that indirectly regulate OpenAi models, to make it clear that company models will not be removed from sensitive topics and will refrain from doing of assertions that can be closed specific views.
The measure, along with the removal of chatgpt warnings, is probably in response to political pressure. Many of President Donald Trump’s close allies, including Elon Musk and Crypto and “Czar” David Sacks, have accused the energy assistants of censoring conservative views. Sacks has singled out Openai’s chatgt in particular as “programmed to wake up” and not true to politically sensitive subjects.
UPDATE: Added explanations by a spokesman for Openai.
Techcrunch has a newspaper focused on it! Sign up here to get it in your box every Wednesday.
