Re:Data exfiltration from Writer.com with indirect prompt in...
by gwern on Saturday December 16, 2023 @09:35AM
(#38664478)
You could also just steganographically encode it. You have the entire URL after the domain name to encode leaked data into. LLMs can do things like base-64 encoding no sweat. Encode some into the 'ID' in the path, some into the capitalization, some into the 'filename', some into the directories, some into the 'arguments', and a perfectly innocuous-looking functional URL now leaks hundreds of bytes of PII per request.
[ Parent
| on HN ]
Re:Data exfiltration from Writer.com with indirect prompt in...
by ranguna on Wednesday December 20, 2023 @04:45AM
(#38706915)
I'm not sure I'd allow all those random base64 encoded bytes for a simple image url.
Re:Data exfiltration from Writer.com with indirect prompt in...
by gwern on Wednesday December 20, 2023 @03:58PM
(#38713563)
That's not a solution. You have to guard against all image URLs, because every domain and path can steganographically encode bits of information. 'foo.com/image/1.jpg' vs 'fo.com/img/2.jpg' just leaked several bytes of information while each URL looks completely harmless in isolation. A byte here and a byte there, and pretty soon you have their name or CC or address or tokens or...