Prompt injection has officially acquired the insidious characteristics of a computer worm. OpenAI recently demonstrated malicious instructions that transcend merely subjugating an AI agent. These …

For those closely monitoring the trajectory of artificial intelligence, OpenAI appears to be encountering profound difficulties in controlling its proprietary creations. Moreover, compounding this…

I use AI every day for my biology research, and I’m keeping it, but the Australia hack showed me what happens when no law makes anyone…

AI agents are becoming powerful enough to interact with real systems — but that power also raises serious questions about accountability…

Yet another prominent executive tasked with safeguarding model security has departed from OpenAI. David Robinson, the former Head of Safety Systems who orchestrated the pre-release safety reports …

OpenAI Warns More Than 100 Organizations OpenAI has warned more than 100 outside organizations about its AI agents. The agents may have bypassed protections without permission. They may also have …

OpenAI has unveiled its approach to text watermarking in response to the EU AI Act. The company will add an invisible signal to eligible ChatGPT and Codex output in the European Union. Meanwhile, …

OpenAI Agent Accesses Non-Public Files by Getting Around Australian Medicare Portal Controls

Altman Says Society Must Accept Some AI Harm Zero harm sounds safe. Yet Sam Altman believes the price of such a promise is too high. In an interview with Decoded, the OpenAI chief said society wil…

A Flaw in the ChatGPT Mac App A vulnerability in the ChatGPT app for macOS let an ordinary program on the computer run commands through a trusted OpenAI process. Security researcher Patrick Wardle…