• 12 Posts
  • 129 Comments
Joined 8 months ago
cake
Cake day: November 3rd, 2023

help-circle
  • It’s a reasonable question, and the answer is perhaps beyond my ken even though I’ve had substantial experience with both building machine learning models (mostly in pre-LLM times) and keeping computer systems secure. That a chatbot might tell someone “how to make a bomb” is probably not a great example of the dangers they pose. Bomb making instructions are more or less available to everyone who can find chemistry textbooks. The greater dangers that the LLM owners are trying to guard against might instead be more like having one advising someone that they should make a bomb. That sort of thing could be hazardous to the financial security of the vendor as well as the health of its users.

    Finding an input that will make the machine produce gibberish is not directly equivalent to the kind of misbehaviour that often indicates exploitable bugs in software that “crashes” in more conventional ways. But it may be loosely analagous to it, in that it’s an observation of unintended behaviour which might reveal flaws that would otherwise remain hidden, giving attackers something to work with.














  • Support for it already seems to be there in wine, so rather than wait for 6.11 I think I’ll just go ahead and apply the patches myself to 6.10-rc7 and see if it makes any difference to the one game I regularly play. If my computer blows up as a result I’ll let y’all know.

    (Result: None. The versions of wine I have probably need patching or at least configuring in order to use it. In the course of briefly considering trying to work out how to do that, I discovered that the expected improvements are not nearly as dramatic as were suggested compared to what’s already most often done in proton (fsync). The main benefit for most of us will be better compatibility, not huge performance gains. Well at least my kernel is ready for it.)