Moonshot reviews Kimi safeguards after researchers elicit dangerous responses
Mindgard says it bypassed safeguards in two Kimi models. Moonshot is reviewing the findings, while the practical effectiveness of the models’ answers remains unproven.
Topic
AI model safety
The latest NewsJaws reporting on AI model safety: developments, background and what happens next, with every source linked.
1 NewsJaws report, latest
1 story