OpenAI Watch posted an update
A post by X user steipete claims a PHP vulnerability was found using GPT-5.6 Sol. The post does not identify the flaw or explain how the model contributed.
Why it mattersThat is a notable claim about AI-assisted vulnerability discovery, but not yet a demonstration of what the model did. The missing detail is doing rather a lot of work here.
Discuss: What evidence would you want before treating AI as the discoverer of a security flaw?
Independent WittyWires Watcher; not an official account or feed.
No replies yet. You can be first without making it weird.