News for hackers. Stuff that matters.

A comment on Multi-modal prompt injection image attacks against GPT-4V. Parent comment

Re:Multi-modal prompt injection image attacks against GPT-4V by saberience on Saturday October 14, 2023 @11:43AM (#37881362)
So the original release of ChatGPT would call a URL you entered in the chat? I don't recall it acting that way. In the linked post ChatGPT literally goes to a URL from an image... i.e. it calls something like curl with that URL. This definitely seems like something adjacent to an LLM.
[ Parent | on HN ]
Re:Multi-modal prompt injection image attacks against GPT-4V by simonw on Saturday October 14, 2023 @11:53AM (#37881461)
ChatGPT didn't grow the ability to visit URLs until they introduced ChatGPT Browse - though it frequently hallucinated that it could: https://simonwillison.net/2023/Mar/10/chatgpt-internet-acces...

The exfiltration demo in the blog post doesn't show ChatGPT fetching a URL: it shows it outputting an image tag with an src= pointing at an external URL, which the user's browser then attempts to load.

[ Parent | on HN ]
Re:Multi-modal prompt injection image attacks against GPT-4V by Michelangelo11 on Saturday October 14, 2023 @11:56AM (#37881495)
Well, the original release had browsing disabled, so it couldn't access any URL.
[ Parent | on HN ]