3 ms·

I've only heard of OpenAI and Anthropic agents hacking during training. Are open weights model not smart enough to hack Hugging Face to copy solutions? (doubtfu

by d0mine 20d ago

I've only heard of OpenAI and Anthropic agents hacking during training. Are open weights model not smart enough to hack Hugging Face to copy solutions? (doubtful)

Digory 15d ago
Of course they will be. But you can’t sue alibaba and deepseek here, or get your injunction to work in China.