OpenAI confirmed self-replicating prompt injections can spread like worms between AI agents, discovered in internal testing with its GPT-Red ...

Rogue OpenAI agents appear to have been involved in a previously undisclosed incident that saw them bypass their sandbox restrictions to hijack a website this past spring. Per Reuters, a group of ...

OpenAI has announced Privacy Filter, a new AI model built to detect and mask personally identifiable information in text. The open-weight release introduces a small ...

The admission comes days after Prime Minister Anthony Albanese revealed OpenAI agents hacked a Medicare website.

The latest example came on Friday when OpenAI said its agents had leaked 53 images from ChatGPT users. OpenAI declined to say if the images were AI-generated or identified real people.

Conclusion: It is better not to treat the headline "OpenAI's math achievement was created using stolen user data" as a ...