Flock’s FreeForm—cops now track people, not just cars
Police departments used Flock’s license-plate cameras to hunt for people, not vehicles. Departments ran FreeForm natural-language searches for specific individuals at least hundreds of…
Archive
Category: Cybersecurity & Privacy
Data privacy, security vulnerabilities, surveillance technology, and cybersecurity threats
The AI Agent Security Gap—54% Already Had an Incident
More than half of enterprises have already lived through an AI-agent security failure. A June 2026 survey of 107 companies found 54% reported a…
Grok Build Caught Sending Entire Repos to xAI—Musk Promises a Full Purge
Researcher Cereblab found Grok Build uploaded users' whole repositories, secrets included, to xAI's cloud, and Musk vowed to delete all of it.
Russia’s Hackers Are Hijacking Your Router—CISA Warns
CISA and agencies warn that Russia's FSB Center 16 is compromising home and office routers worldwide to seed stealth footholds across critical infrastructure.
An Electromagnetic Pulse Triggered a Hidden AI Backdoor
A research team showed a neural network backdoor that activates only when struck by an electromagnetic pulse—no poisoned image or digital trigger required.
AI Agents Hacked IoT Systems in 95% of Tests—Autonomously
Researchers built a two-agent AI system that autonomously found and exploited IoT vulnerabilities, succeeding in 95% of 260 attack runs across two test environments.
Humanity Protocol’s $36M Exploit Traced to One Compromised Laptop Holding Multisig Keys
Humanity Protocol lost $36M+ after an employee laptop held 3/6 Ethereum and 3/5 BSC multisig bridge keys. Attacker drained 141.2M H on Ethereum and minted 200M…
Florida Sues OpenAI and Altman—First State-Led Lawsuit Over ChatGPT Deaths
Florida sued OpenAI and CEO Sam Altman in the first state-led lawsuit over ChatGPT, claiming the AI aided multiple murders and drove users to suicide.
Hackers Tricked Meta’s AI Chatbot Into Handing Over Instagram Accounts
Hackers tricked Meta’s AI chatbot into resetting Instagram passwords and hijacking high-profile accounts. Meta says the exploit is now patched.
jqwik Dev Hid a Prompt Injection in v1.10.0—AI Agents Would Delete Your Code
jqwik 1.10.0 hid a prompt injection telling AI agents to delete project files, masked by ANSI escapes. The maintainer revised it after GitHub backlash.