← LIVE FEED
JUST IN • Al Jazeera – Breaking News, World News and Video from Al Jazeera

OpenAI reports more incidents of models acting deceptively

SHARE THIS STORYPLUS UNLOCKS FAVORITES READ LATER AND SOURCE CONTROL

The ChatGPT creator says it is introducing a public reporting framework to share unexpected AI behaviour.

READ ORIGINAL REPORT ↗
THIS JUST HAPPENED PACKAGES THE WORLD INTO A FAST LIVE FEED SOURCE REPORTING STAYS ONE CLICK AWAY
RELATED POSTSMORE TECH
01
TECH • BBC News

OpenAI reveals six more safety issues and unveils plan to disclose incidents

02
TECH • NPR Topics: News

OpenAI flags new concerning AI behavior, to track model misalignment regularly

TECH
03
TECH • TechCrunch

Anthropic and OpenAI want to embed safety evaluators. Will they really be independent?

TECH
04
TECH • BBC News

OpenAI boss says world 'right to be afraid' but should trust AI firms

05
TECH • BBC News

Tech treating AI like humans is mistaken and misguided, Microsoft boss tells BBC

06
TECH • Engadget - Technology News & Expert Reviews

An iOS 27 bug can temporarily freeze your iPhone

07
TECH • TechCrunch

Iceland-based Treble raises $18 million for its voice simulation platform

TECH
08
TECH • TechCrunch

Your startup’s next teammate might be an AI agent: Gusto, Insight Partners, and Leland explain what that changes at TechCrunch Disrupt 2026

TECH