AI Apps Are Undressing Women Without Consent — And It’s a Problem AI-powered “nudification” apps are making it horrifyingly easy to generate non-consensual intimate images of women and teens—an urgent and harmful misuse of generative AI. Read more
SECURITY
-
Shadow AI in Workplace: Undeclared Tool Usage and Data Risk
By
–
1 salarié sur 3 utilise l’IA… sans prévenir sa direction.
C’est le Shadow IA.
+250 % d’outils non déclarés en 1 an.
8 % des prompts révèlent des données sensibles. Tolérance zéro ou innovation cachée ? Bienvenue dans la leçon 6 de mon #FlashCahierDeVacances – IA -
Cyber Deterrence Strategy Against China Requires Stronger Action
By
–
In my opinion, that would not leave the PRC trembling with fear. I’m all for smart moves, but if we want to make the imaginary cyber deterrence she references a reality we have to do things that scare the shit out of them, and a digital twin would not do that, would it?
-
Major Prompt Injection Vulnerabilities Found in Leading AI Products
By
–
Took some time today to catch up with @wunderwuzzi23
's Month of AI Bugs and wow… 15 examples so far of major prompt injection vulnerabilities in products including ChatGPT, Codex, Cursor, Amp, Devin, Claude Code, GitHub Copilot and Google Jules -
Sophisticated Cyber Threats Targeting Manufacturing Environments Rise
By
–
Cyber threats targeting manufacturing environments are growing more sophisticated—and relentless. For many industrial orgs, the biggest challenge isn’t ransomware itself, but knowing where to start. Learn more: https://
buff.ly/wPJfRv1 #sponsored #fortinet_ics #cybersecurity -
Recovering Model Weights Exactly Post-LoRA: New Discovery
By
–
just learned that under certain conditions you actually CAN recover model weights *exactly*, post-LORA pretty crazy
-
New Vulnerability Enables Pre-Fine-Tuning Weight Recovery Attack
By
–
— dr. jack morris (@jxmnop) 15 août 2025
We uncover a new vulnerability- Pre-Fine-Tuning Weight Recovery
With a few LoRA fine-tuned models we recover the pre-fine-tuning weightsof SoTA models, undoing Stable Diffusion personalization training and Mistral alignment
Project: http://
vision.huji.ac.il/spectral_detun
ing
… -
Q-Day Threat: Quantum Computing And Digital Security Collapse
By
–
Why You Should All Be Worried About Q-Day And The Collapse Of Digital Security #QuantumComputers are advancing fast, and with them comes the looming threat of Q-Day, a moment when #encryption could be shattered and #digital #security #collapse. This article explores how close
-
Trust Over Technology: Human Behavior in Cybersecurity Breaches
By
–
When digital breaches strike with such magnitude, it becomes clear that trust, rather than technology, is often the weakest layer in cybersecurity, especially when human behavior turns into the attack surface. Infographic by @VisualCap via @antgrasso #cybersecurity
-
Comet for Enterprise: AI-Powered Browser Agent Launches
By
–
Comet for Enterprise is here.
— Perplexity (@perplexity_ai) 14 août 2025
Comet is an AI-powered browser agent that thinks with you, linking tools for streamlined workflows and trusted answers.
Enterprise Pro users maintain the security, privacy, and compliance standards that come with an Enterprise subscription. pic.twitter.com/HdD7t0uswpComet for Enterprise is here. Comet is an AI-powered browser agent that thinks with you, linking tools for streamlined workflows and trusted answers. Enterprise Pro users maintain the security, privacy, and compliance standards that come with an Enterprise subscription.