My paper on AI influence & propaganda, and the real risks, will (hopefully) be public this week. The aim is to study offensive capabilities to better inform defense. Some results are quite surprising. This is my contribution to the AI safety field. We're in for a wild run.
Follow
Feel free to share here, as I doubt most main stream media outlets will either report or report what the paper actually says or importantly concludes, we can read ourselves, happy to put a link on my website too.
