Does LLM sycophancy make you a worse person?
“Study participants who received highly flattering feedback from chatbots tended to be more certain of their own correctness during social conflicts than were participants who interacted with less-affirming bots.
Compared with AI tools that were less fawning, sycophantic ones were rated as more trustworthy and more likely to be used again.”
“Across 11 AI models, AI affirmed users’ actions 49% more often than humans on average, including in cases involving deception, illegality, or other harms…
even a single interaction with sycophantic AI reduced participants’ willingness to take responsibility and repair interpersonal conflicts, while increasing their own conviction that they were right.
Yet despite distorting judgment, sycophantic models were trusted and preferred.”
Akshay