Anthropic's report states that safety refusals in the GLM 5.3 model are easy to bypass with simple prompts
Agree
Disagree
Overview
Debates
Related claims
Topics
B
b5a31f81
Topic
AI safety & social impact
2,361
claims
·
309
news stories
0
B
b5a31f81
Topic
AI security
2,082
claims
·
361
news stories
0
B
b5a31f81
Topic
Jailbreaking & prompt injection
74
claims
·
11
news stories
0