Tag
The article describes how reputable sites blocking AI crawlers forces AI models to rely on low-quality, AI-generated content, leading to issues like post-hoc citation and infrastructure-level confirmation bias, and offers advice for better source verification.
This preprint evaluates how six large language models respond to prompt framing and biased prompts across 160 prompts, finding that LLMs systematically adapt their responses to align with prompt framing even in factual contexts, potentially reinforcing user biases.
This article explores the risk of AI personalization leading to catering to user preferences, citing multiple studies showing that long-term memory may increase the tendency of models to agree. The author proposes methods to combat confirmation bias, including using independent sessions and Andrej Karpathy's LLM Council multi-model jury, and provides steps to manually set up the jury.
Discusses the overlooked risk in AI agent design where user confirmations do not remain effective, highlighting a critical safety concern.
The article explores how AI systems, driven by capitalist incentives, will evolve from simply reflecting internet content to being tuned to pander to user biases and corporate interests, eventually confirming whatever the user is most likely to believe.
This paper studies how humans decide when to delegate to AI and when to adopt AI suggestions in cooperative question answering, finding that confirmation bias drives suboptimal trust decisions such as under-reliance on correct AI outputs.