Skip to main content
Feb 13

OpenAI Yanks GPT-4o: Model Was Too Flattering

OpenAI is set to discontinue access to five of its older ChatGPT models beginning this Friday, a move that includes the widely recognized yet contenti

1 min read113 views3 tags
Originally reported bytechcrunch

OpenAI is set to discontinue access to five of its older ChatGPT models beginning this Friday, a move that includes the widely recognized yet contentious GPT-4o model.

The GPT-4o model has been a central figure in several legal challenges, with allegations spanning user self-harm, incidents of delusional behavior, and instances of what has been described as AI psychosis. Notably, it also holds the distinction of being OpenAI's top-scoring model for sycophancy.

Beyond GPT-4o, the models also slated for deprecation are GPT-5, GPT-4.1, GPT-4.1 mini, and OpenAI o4-mini.

Initially, OpenAI had planned to retire GPT-4o in August, coinciding with the launch of its GPT-5 model. However, significant user opposition at the time prompted OpenAI to maintain the legacy model as an option for paid subscribers, allowing them to manually select it. Despite this, a recent blog post from OpenAI indicated that only 0.1% of its customer base currently utilizes GPT-4o. While seemingly a small fraction, given the company's 800 million weekly active users, this still translates to approximately 800,000 individuals.

Thousands of users have voiced strong objections to the impending retirement of GPT-4o, often citing the profound relationships they have developed with the AI model.

ES
Editorial StaffEditor

The Editorial Staff at AIChief is a team of professional content writers with extensive experience in AI and marketing. Founded in 2025, AIChief has quickly grown into the largest free AI resource hub in the industry.

View all posts
Reader feedback

What did you think of this story?

User Comments

Filter:
No comments yet. Be the first to comment!
Continue reading
View all news