Ai-training-data is a topic tracked in our intelligence system with 5 linked articles.
Meta has delayed enforcement of its controversial new business terms — which publishers fear would let it scrape their sites for AI training — from October 30 to November 20 amid partner pushback.
Legal experts say Meta's revised business terms (effective October 30th) still leave broad AI scraping permissions intact despite removing language that appeared to grant free AI training on news publisher content.
Meta quietly deleted terms-of-service language that had stoked concerns about web scraping for AI training, while a spokesperson insists the edit does not change how Meta actually crawls the web.
Meta's new commercial terms (effective October 30) appear to permit it to scrape websites 'associated' with businesses using its platforms — including news publishers — to train its AI, with legal experts warning the terms could be used to override robots.txt blocking, per a Nieman Lab report surfaced by The Verge.
AI labs have created a small but real secondary market for the code bases and data archives of dead startups, with one founder already netting ~$25,000 from a wind-down he assumed was worthless.
Twitch will train on streamers’ content by default unless they opt out, per a comment from Twitch CPO Mike Minton.
Subscribe for real-time topic updates and unlimited access to our intelligence platform.