I extracted the safety filters from Apple Intelligence models

I managed to reverse engineer the encryption (refered to as “Obfuscation” in the framework) responsible for managing the safety filters of Apple Intelligence models. I have extracted them into a repository. I encourage you to take a look around.

Comments URL: [https://news.ycombinator.com/item?id=44483485][1]

Points: 464

# Comments: 336

[1]: https://news.ycombinator.com/item?id=44483485

https://github.com/BlueFalconHD/apple_generative_model_safety_decrypted