arXivGianluca Barmina, Federico Torrielli, Sven Harms, Jacob Nielsen, Felix Mächtle, Stine Lyngsø Beltoft, Peter Schneider-Kamp, Thomas Eisenbarth, Lukas Galke Poech, Anne LauscherMon, Jun 8, 2026, 9:19 AM PDT
score 17.1
AI models learn to refuse requests with psychological support
Original: PsychoSafe: Eliciting Psychologically-Informed Refusals in Large Language Models
Source: arxiv.org ↗
Writing ELI5 summary…