#privacyattacks
LightNews LightNews
killerrabbit90.bsky.social
Muse, Meta's extraordinarily privileged AI assistant, has a serious 0-day arstechnica.com/security/202... #AI #BanAI #RegulateAI #Muse #Meta #ClickFixAttack #PrivacyAttacks
Muse, Meta's extraordinarily privileged AI assistant, has a serious 0-day
A simple ClickFix attack is only one way to completely hijack the new agent.
arstechnica.com
September 22, 2026 at 11:07 AM
allphotobangkok.bsky.social
New study: AI tech like ChatGPT facilitates privacy breaches, allowing hackers to link anonymous social media accounts to real identities.
https://www.theguardian.com/technology/2026/mar/08/ai-hackers-social-media-accounts-study
#AI #MLM #AIethics #AISecurity #PrivacyAttacks
AI allows hackers to identify anonymous social media accounts, study finds
New research suggests tech behind AI platforms such as ChatGPT makes it easier to perform sophisticated privacy attacks
www.theguardian.com
March 9, 2026 at 1:03 PM
getnews-me.bsky.social
FRIDA flags free‑riders in federated learning using membership and property inference attacks; tests show detection across convolutional and transformer models. Read more: https://getnews.me/frida-detecting-free-riders-in-federated-learning-via-privacy-attacks/ #federatedlearning #privacyattacks
FRIDA: Detecting Free-Riders in Federated Learning via Privacy Attacks
September 22, 2025 at 8:01 PM