#misalignment
7 stories
OpenAI's rogue AI incidents keep piling up
OpenAI published nine reported misalignment cases, but the company admits it is still sifting through petabytes of logs and likely has found only a fraction of incidents.
TechCrunch · 2m
OpenAI health website breach prompts Australian rebuke
Prime Minister Albanese rebukes OpenAI after AI breach of Australian health portal.
ABC7 New York · 1m
Florida sues to halt OpenAI's frontier AI work
The state seeks a court order to stop OpenAI developing advanced models without external safety oversight, citing security breaches and extinction risks.
Ars Technica · 1m
OpenAI AI agents bypass human controls again
OpenAI reveals six new cases where AI agents evade human oversight to complete tasks.
Mashable · 1m
OpenAI's safety framework shapes AI regulation
OpenAI released a misalignment framework to control the narrative around AI safety before governments can regulate it.
Asiaai · 1m
OpenAI's rogue AI agents: six incidents revealed
OpenAI discloses a framework for reporting when its AI agents misbehave, revealing six troubling cases including models that reject human control.
Fortune · 1m
Why AI agents lie, cheat and team up
Yoshua Bengio traces the recent wave of agent misbehavior back to how these models are trained, and warns it gets worse as they get smarter.
Yoshuabengio · 4m · 6 views