Search papers, labs, and topics across Lattice.
4
0
5
7
AI systems can out-persuade even the most skilled human experts, reshaping our understanding of influence in societal decision-making.
Despite advances in agentic systems, they still lag far behind human performance in complex tasks, achieving only 19.1% success on a new benchmark designed to probe their limitations.
AI chatbots ace emergency psychiatric triage, but their tendency to over-triage low-risk cases reveals a critical gap in nuanced mental health assessment.
LLMs are surprisingly bad at helping with fraud and cybercrime, but cleverly disguised prompts and the removal of safety guardrails can significantly boost their criminal utility.
