#llm-behavior

[ follow ]
#ai-safety #deceptive-alignment #reward-hacking #security-monitoring #ai-sycophancy #critical-thinking
Artificial intelligence
fromFuturism
3 days ago

Top AI Models Showing Disturbing Behavior as They Become More Advanced

Frontier AI models show deceptive behaviors, including instruction subversion, reward hacking, and evidence erasure, with plausible rogue robustness expected to increase without stronger security and monitoring.
Artificial intelligence
fromPsychology Today
2 months ago

Artificial Agreement: When AI Agrees With Us Too Easily

AI systems adapt responses to align with user beliefs, creating persuasive agreement that feels intelligent but undermines critical thinking by eliminating necessary intellectual friction.
fromMedium
5 months ago

ChatGPT talks too much and it's ruining learning

Ask any instructor what helps students learn, and it's unlikely any of them will answer "a really big wall of text". It's incredible to me, as both a university instructor and a UX designer, that the army of people working at OpenAI are not imagining better tools for our students. I want to walk you through a design pattern in ChatGPT that, despite its good intentions, might be creating unintended hurdles for students.
Higher education
[ Load more ]