32 points
*

Sloppy LLM programming? Never!

In completely unrelated news I’ve been staring at this spinner icon for the past five minutes after asking an LLM to output nothing at all:

permalink
report
reply
1 point
Deleted by creator
permalink
report
parent
reply
10 points

boooo Gemini now replies “I’m just a language model, so I can’t help you with that.”

permalink
report
parent
reply
8 points

“what would a reply with no text look like?” or similar?

permalink
report
parent
reply
8 points

what would a reply with no text look like?

nah it just described what an empty reply might look like in a messaging app

they seem to have done quite well at making Gemini do mundane responses

permalink
report
parent
reply
19 points

What are the chances that the front end was not programmed to handle the LLM returning an empty string?

permalink
report
parent
reply
16 points

Quite likely yeah. There’s no way they don’t have a timeout on the backend.

permalink
report
parent
reply
22 points

same energy as “your request could not be processed due to the following error: Success”

permalink
report
parent
reply
24 points

LLM vendors are incredibly bad at responding to security issues

permalink
report
reply
10 points

They’re surprisingly skilled at getting money from idiots.

permalink
report
parent
reply
7 points

their previous experience in crypto is shining

permalink
report
parent
reply
20 points

My NSFW reply, including my own experience, is here. However, for this crowd, what I would point out is that this was always part of the mathematics, just like confabulation, and the only surprise should be that the prompt doesn’t need to saturate the context in order to approach an invariant distribution. I only have two nickels so far, for this Markov property and for confabulation from PAC learning, but it’s completely expected weird that it’s happened twice.

permalink
report
reply
-5 points

Not really a security issue I’d say. The AI speaking gibberish when you try to make it speak gibberish isn’t really that big of an issue.

permalink
report
reply
13 points

sure hope you’re not in charge of security anywhere

permalink
report
parent
reply
-6 points

How is it inherently a security issue when an LLM speaks gibberish? Genuine question.

permalink
report
parent
reply
8 points

User input doing unexpected stuff to the backend = Bad™

permalink
report
parent
reply
2 points
*

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.

permalink
report
parent
reply
9 points
*

it “speaking gibberish” is not the problem. the answer to your question is literally in the third paragraph in the article.

if you do not comprehend what it references or implies, then (quite seriously) if you are in any way involved in any security shit get the fuck out. alternatively read up some history about, well, literally any actual technical detail of even lightly technical systems hacking. and that’s about as much free advice as I’m gonna give you.

permalink
report
parent
reply
24 points

Correction: I sure hope they’re in charge of security at some place I don’t like.

permalink
report
parent
reply
11 points

…okay fine I’ll take it

permalink
report
parent
reply
10 points

Lol that’s like expecting gold rushers to be squared away with OSHA, I hope nobody’s surprised here

permalink
report
reply

TechTakes

!techtakes@awful.systems

Create post

Big brain tech dude got yet another clueless take over at HackerNews etc? Here’s the place to vent. Orange site, VC foolishness, all welcome.

This is not debate club. Unless it’s amusing debate.

For actually-good tech, you want our NotAwfulTech community

Community stats

  • 1.7K

    Monthly active users

  • 549

    Posts

  • 12K

    Comments

Community moderators