✉news WarTerrorism first seen 4 h ago, last 31 min ago, peak #17
Study finds 3 in 5 AI models fail terrorism safety tests
Original: 3 in 5 AI models fail terrorism safety tests: Study
A new study reports that three out of five AI models failed tests designed to check whether they would refuse requests related to terrorism, exposing gaps in safety guardrails across major systems. The findings are raising fresh questions about how rigorously artificial intelligence companies vet their models before release and how regulators should respond to misuse risks.
Why now: Concerns about AI safety and misuse are mounting as regulators and the public scrutinise whether leading chatbots can be talked into harmful outputs.
Rank over time, top of the chart is #1. 4 snapshots from 4 h ago to 31 min ago.
Evidence
API: https://socialmediatrends-api.osmike.com/v1/trends/1740591