3 ms·
Once you have a reliable model to detect AI-content, you also have a new tool to train AI-models against. So ironically, any tool to detect AI content will just
by juujian 3y ago
Once you have a reliable model to detect AI-content, you also have a new tool to train AI-models against. So ironically, any tool to detect AI content will just serve to make AI content more difficult to detect.
- PartiallyTyped 3y agoWhich you can use to train a better model to detect AI content. I wonder whether some models fit this - well, you could certainly call it generative, and obviously adversarial - framework of networks?
- passion__desire 3y agoThe new AI models will seek novelty[0], pieces of facts not recorded before. It will directly hone in on newness not how it was delivered. Writing styles could be another vector in which novelty could be detected but that's a different matter. [0] https://people.idsia.ch/~juergen/sice2009.pdf https://people.idsia.ch/~juergen/sice2009.pdf
- bee_rider 3y agoBut the ability to detect, say, content generated by pre-2033 chatbots will open up a whole decade of previously contaminated content.
- collegeburner 3y agoright, you're describing a gan-like system/positional arms race. which is how adversarial tech always develops. doesn't mean we won't get better AI out of it.