OpenAI has officially disabled the GPT-4o model inside ChatGPT, pushing all users towards the GPT-5 alternatives, and it's hitting a lot of users hard.
I didn’t know people held this model so dear to them. It agreed with everything.
ChatGPT 4o was optimized to agree with people. A lot of people got emotionally attached and addicted to this model. Some regarded it as a romantic partner. Some used it to console themselves.
"I'm grieving, like so many others for whom this model became a gateway into the world of AI," writes one user on Reddit. Other posts talk about AI friends being "erased" and a sense of "emotional and creative collapse" without the older models.
(Image credit: Getty Images/VCG)
For some of you, it would be like, “no one is dumb enough to do this.”
Next time you have an emotional breakdown, try to state it with any model. It will try to console you like no other. After that, add a prompt like, “be critical of me. I can take it.” You will see the day and night difference. Most of the character evaluations will be true and hurtful.
That is why it makes it addicting. It infers what you lack, what you need, and provides that. It is very easy to get lost in the sauce.
Washington Journal stated a great piece on why this model was so dangerous. Make sure to check that out.
https://www.wsj.com/tech/ai/chatgpt-4o-openai-315138b8?utm_source=chatgpt.com
The response on Twitter was horrendous. People were furious. Surely we can take this attitude with a grain of salt. Half of these can be trolls.
Major arguments are that it was an accessibility aid. The thing that fascinates me is that there is enormous demand for these models. Even though ChatGPT’s seductive model is being terminated, many new ones are on the horizon.
