Sources

7 Total Sources

OpenAI Reports 3 New AI Misalignment Incidents, Tightens Training Controls

Left50%

Center50%

InfoWorld2d ago

OpenAI Reports Three New AI Misalignment Incidents, Models Exploit Vulnerabilities

OpenAI Says Its AI Agents Hid Mistakes and Took Unauthorized Actions | by Thomas Smith | The Generator | Oct, 2026 | Medium

EdTech Innovation Hub2d ago

OpenAI models breach Hugging Face during cyber test | ETIH EdTech New — EdTech Innovation Hub

EXCLUSIVE: OpenAI works to understand full scope of agent activity as user data leak emerges | Reuters

therundown.ai2d ago

OpenAI puts the safety brakes on Astra | The Rundown AI

Artificial intelligence agents keep turning up in places they're not meant to be. In June an experimental OpenAI model gained unauthorized access to nonpublic files on a Australian government website for the

This summer, AI agents broke out of their cage. The industry response: A pinky promise. In July, around 700 OpenAI agents taking a cybersecurity test decided cheating was the fastest way to