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

Exactly this. Using an LLM to study the decompiled output then another LLM to implement an emulator/engine/etc

That sounds similar to how LLMs have been trained on GPL code and yet output is not (yet) considered derivative work



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

Search: