Revealing the intrinsic ethical vulnerability of aligned large language models

Journal:
Nature Communications
Published:
DOI:
10.1038/s41467-026-70917-y
Affiliations:
2
Authors:
8
Go to article
Institutions Authors Share
Northwestern Polytechnical University (NPU), China
4.500000
 0.56
The Hong Kong Polytechnic University (PolyU), China
3.500000
 0.44
Interested in Artificial Intelligence?