Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

>>> If you are wondering what the catch is: you still need the network weights.

That's why I'm very worried about neural nets stuff : you can share any code you want, the secret sauce (the weights) will be easy to hide. GPL don't quite apply here.

It's probably true of hundreds of other applications, but somehow, I have the feeling it's different with AI 'cos the data "embody" the behavior much more.



This is why lots of large companies are happy to open-source internal code, no? Because ultimately it's data which is valuable, not the algorithms.


If you compile the data inside the program it's probably GPLed too and must be made public as soon as the program is distributed. Workarounds: the data go inside an external library or are downloaded from a service. Then you're right to worry: the real algorithm would be secret.


> GPL don't quite apply here.

You can license your trained network data GPL if you choose. Applies fine.

You are right that the valuable IP is in the data, but this is no different to Doom without the WADs, or MAME without ROMS.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: