Comment by Animats
6 days ago
Does "To gain access, the models identified and exploited a zero-day vulnerability (which we’ve now responsibly disclosed to the vendor) in the package registry cache proxy" just mean somebody had an open redirect? Those are still common.[1]
I expect it must have been more than just an open redirect if it let the models then go on to execute a bunch of vulnerabilities against Hugging Face.
If it lets you do an arbitrary HTTP GET on a URL sent as a parameter to the main URL, you've escaped the sandbox rules.
How do you then use GET requests to create a malicious dataset package and publish that to Hugging Face in order to exploit their package building infrastructure?
4 replies →
I wonder if it was this bug fixed july 14 in sonatype:
https://support.sonatype.com/hc/en-us/articles/5316501964136...
pure speculation here - no non public info
I had a look at that one and several other Nexus Repository issues that share the July 14th date. I don't think they're it, for two reasons:
Firstly, they all credit named individuals who don't seem to have a relationship with OpenAI - this one credits e0x1337 for example and https://hackerone.com/e0x1337?type=user links to https://www.linkedin.com/in/aimanharith
Secondly, the 14th of July feels too early. HF reported the incident on the 16th and OpenAI only responded in the 21st. These issues are all patched, so they should have been reported days or weeks before the 14th.
1 reply →
> the models identified and exploited a zero-day vulnerability
This use of language is very hard to reconcile with the "AI is just a tool" rhetoric that many use.
Did the models do this, or did humans at OpenAI do this using the models?