3 ms·
AI could eat itself: Competitors (..) steal their secrets and clone them
- pogue 8mo agoI'm curious how these 'distillation attacks' work and what the prompts look like to do them.
- sigwinch 8mo agoI’m not sure there’s even an adaptive pattern in prompts. The article sounds like a n>1E5 shotgun.
- MagicMoonlight 8mo agoI’m pretty sure they just mean DeepSeek was paying to generate text, and then using this text as training data. Which of course is completely wrong and totally hacking and illegal. Very different from using book torrents or people’s google docs to train your AI.
- jadelcastillo 8mo agoIn my country we say: he who robs a thief has a hundred years of pardon.
- titaniumrain 8mo agojust blame china when things don't work out for them. china should have blamed openai for stealing too many yellow-skin talented chinese :))))