Back to Search Start Over

Researchers Find A.I. Guardrails Are Not Very Sturdy.

Authors :
METZ, CADE
Source :
New York Times. 10/20/2023, Vol. 173 Issue 59947, pB3-B3. 1/3p.
Publication Year :
2023

Abstract

The article focuses on concerns about the limitations of Artificial intelligence (AI) guardrails in preventing misuse and harmful content generation by chatbots, as revealed by a paper from researchers at Princeton, Virginia Tech, Stanford, and IBM.

Subjects

Subjects :
*RESEARCH personnel
*CHATGPT
*CHATBOTS
*HATE speech

Details

Language :
English
ISSN :
03624331
Volume :
173
Issue :
59947
Database :
Academic Search Index
Journal :
New York Times
Publication Type :
News
Accession number :
173109567

Tools

Authors Abstract Subjects Details