4 ms·

will happily run this on several hundred highly starred GitHub projects if someone ports the approach to code... (of course if Microsoft let me opt out I'd hav

will happily run this on several hundred highly starred GitHub projects if someone ports the approach to code... (of course if Microsoft let me opt out I'd have no reason to)

artninja1988 3y ago
It's not about code Generation amigo. This is for images
- lawlessone 3y ago
  I think what he meant by port was to create an analogous software for code.
  - Tokumei-no-hito 3y ago
    wasn't there an attack a while ago that used hidden characters to break compilers? i'll see if i can dig it up. maybe something like that could be used for github. you'd have to have a pre commit and pull hook that would encode / decode the malicious characters
  - artninja1988 3y ago
    Wouldn't you just need to write a bunch of bad code?
    - blibble 3y ago
      applying it to good (highly rated) code would do more damage to the model the behaviour would be identical for the end user, but very different for those stealing, sorry, "training AI models" from it
      - andybak 3y ago
        > Stealing Are you old enough to remember "You wouldn't download a car"?
      - pixl97 3y ago
        Did you read "the right to read" and take it as a training manual and not a warning?
  - Tokumei-no-hito 3y ago
    found it. here's the paper https://trojansource.codes/trojan-source.pdf https://trojansource.codes/trojan-source.pdf
  - astrange 3y ago
    Code generators are autoregressive (code->code) not labeled (text->image) so the attack wouldn't work. Also, you can actually tell if a code model works or not while you're training it, by running tests on the output. You could put up a lot of misleading code where the comments are wrong or there's bugs in it… seems bad for obvious reasons though.
schleck8 3y ago
The solution is not publicly sharing your code if you do not want it being used. Yes, you can spin up a whole debate about monetization etc pp but generative ai is the future of development too and will make it more accessible in a way that we can hardly imagine right now. So just stop with these arbitrary red lines of voluntarily published sources.

will happily run this on several hundred highly starred GitHub projects if someone ports the approach to code... (of course if Microsoft let me opt out I'd have no reason to)

artninja1988 3y ago

It's not about code Generation amigo. This is for images

lawlessone 3y ago
I think what he meant by port was to create an analogous software for code.
- Tokumei-no-hito 3y ago
  wasn't there an attack a while ago that used hidden characters to break compilers? i'll see if i can dig it up. maybe something like that could be used for github. you'd have to have a pre commit and pull hook that would encode / decode the malicious characters
- artninja1988 3y ago
  Wouldn't you just need to write a bunch of bad code?
  - blibble 3y ago
    applying it to good (highly rated) code would do more damage to the model the behaviour would be identical for the end user, but very different for those stealing, sorry, "training AI models" from it
    - andybak 3y ago
      > Stealing Are you old enough to remember "You wouldn't download a car"?
    - pixl97 3y ago
      Did you read "the right to read" and take it as a training manual and not a warning?
- Tokumei-no-hito 3y ago
  found it. here's the paper https://trojansource.codes/trojan-source.pdf https://trojansource.codes/trojan-source.pdf
- astrange 3y ago
  Code generators are autoregressive (code->code) not labeled (text->image) so the attack wouldn't work. Also, you can actually tell if a code model works or not while you're training it, by running tests on the output. You could put up a lot of misleading code where the comments are wrong or there's bugs in it… seems bad for obvious reasons though.

schleck8 3y ago

The solution is not publicly sharing your code if you do not want it being used. Yes, you can spin up a whole debate about monetization etc pp but generative ai is the future of development too and will make it more accessible in a way that we can hardly imagine right now. So just stop with these arbitrary red lines of voluntarily published sources.