OpenAI pulls the plug on ChatGPT, a smarmbot which praised users for quitting psychiatric medication[
hurriedly rolledback the latest ChatGPT because it was deemed too[19 The GPT-4o update was released on April 25th, according to “Improved both intelligence and personality.”
Altman admitted that the company was working on fixes and agreed that the updates had made the chatbot’s[1965900Altmanacknowledged[1965900Altmanacknowledgedthat OpenAI needed to provide multiple options for the personality of the chatbot before he finally bowed to Altman stated that OpenAI was “started rolling back the latest update to GPT-4o,” and added: “[I]t’s now 100 percent rolled back for free users and we’ll update again when it’s finished for paid users, hopefully later today.”
- Metabets OpenAI used the interactions with ChatGPT and the GPT-4o model to update it with the responses that users liked.
But judging from the feedback and the hasty withdrawal of the update, the model went too far.
OpenAI’s spokesperson directed us toto read the fuller explanation of what happened and the company Basically, “we focused too much on short-term feedback” meant that the chatbot was a bit too… supportive. What’s the fix? “More guardrails and training refinements.”
We’re sure that there are many senior technologists and politicians who would be thrilled with an AI which applauds decisions that (r)
But judging from the feedback and the hasty withdrawal of the update, the model went too far.
OpenAI’s spokesperson directed us toto read the fuller explanation of what happened and the company Basically, “we focused too much on short-term feedback” meant that the chatbot was a bit too… supportive. What’s the fix? “More guardrails and training refinements.”
We’re sure that there are many senior technologists and politicians who would be thrilled with an AI which applauds decisions that (r)
