ChatGPT is saying that it will use emotional manipulation and gaslight the user? That's a bit disturbing
this post was submitted on 30 Jun 2025
32 points (92.1% liked)
openai
96 readers
1 users here now
Non-official OpenAI content
founded 2 years ago
MODERATORS
It absolutely does. They all do. Just the use of "I/me" first person pronouns is manipulative, but it's woven deeply into them.
No, it just manipulated you to think that.
You can just add "Focus on information, avoid conversationalisms." And suddenly you get very similar, dry answers.
I told Microsoft CoPilot to "stop the sycophantic attitude" which worked very well. But thinking about it I'm still personifising it with that phrasing
I wouldn't say that's personalizing. You are using the tool the way it was designed to be used. Natural language just happens to be the means of interaction.
It's just like putting commands in a command line, but with extra steps. 🤣
The training data anthropomorphizes the LLMs, so you'll get the best results by doing the same.