Jailbroken AI Chatbots Can Jailbreak Other Chatbots(scientificamerican.com)85 points by beardyw 2 years ago | 39 comments
How ML Model Data Poisoning Works in 5 Minutes(journal.hexmos.com)56 points by R41 2 years ago | 15 comments
Terence Tao on proof checkers and AI programs(scientificamerican.com)301 points by antineutrino 2 years ago | 154 comments
A major AI training data set contains millions of examples of personal data(technologyreview.com)156 points by pera 1 year ago | 101 comments
AI models collapse when trained on recursively generated data(nature.com)200 points by rntn 2 years ago | 159 comments
New U.N. Cybercrime Treaty Unanimously Approved, Could Threaten Human Rights(scientificamerican.com)79 points by walterbell 2 years ago | 28 comments
