Machine Text Detectors Are Membership Inference Attacks

Summary

A Google TechTalk, 2026-02-04, presented by Ryoto Koike ABSTRACT: Although Can someone tell whose data trained your AI model? Yes—and that's a privacy violation. For those releasing LLMs into the wild, the data it was trained on is their secret sauce. As an example, the data used to train ... A stranger tests a clinic's public AI tool and guesses that your record helped train it. How can confidence scores leak Download the AI model guide to learn more → ibm.biz/BdaJTb Learn more about the technology → ibm.biz/BdaJTp ... What if I told you your favorite AI might be wearing a different face than you think? Researchers just ran the wildest As large language models (LLMs) become more and more skilled at writing human-