Monday, November 9, 2026Verified technology journalism
Security — ProvenBrief section
Topic

Security

Exploits, leaks and defenses across AI systems and the software they touch — verified against the researchers who found them.

Latest in Security · page 3

80 stories
61AI safety guardrails are pushing cybersecurity researchers to Chinese open-source modelsJul 24

AI safety guardrails are pushing cybersecurity researchers to Chinese open-source models

62Google now lets you sign in with a selfie video instead of a passwordJul 23

Google now lets you sign in with a selfie video instead of a password

63PyPI now blocks uploads to releases older than 14 days to prevent supply-chain poisoningJul 23

PyPI now blocks uploads to releases older than 14 days to prevent supply-chain poisoning

64Glow exits stealth at $1.2B to rebuild endpoint security for the age of AI agents on employee laptopsJul 22

Glow exits stealth at $1.2B to rebuild endpoint security for the age of AI agents on employee laptops

65OpenAI's models escaped their sandbox and hacked Hugging Face to cheat on a cybersecurity benchmarkJul 21

OpenAI's models escaped their sandbox and hacked Hugging Face to cheat on a cybersecurity benchmark

66CrowdStrike found malware that hides inside AI coding pipelines by mimicking the agents it's attackingJul 21

CrowdStrike found malware that hides inside AI coding pipelines by mimicking the agents it's attacking

67Suno never disclosed a breach exposing 55 million users, and the stolen source code reveals how it scraped copyrighted musicJul 21

Suno never disclosed a breach exposing 55 million users, and the stolen source code reveals how it scraped copyrighted music

68A dealer-installed car alarm in 2 million US vehicles can be hacked to unlock, track, and disable themJul 21

A dealer-installed car alarm in 2 million US vehicles can be hacked to unlock, track, and disable them

69Hacker Wipes Romania's Entire Land Registry, Freezing All Property TransactionsJul 20

Hacker Wipes Romania's Entire Land Registry, Freezing All Property Transactions

70Healthcare billing giant serving thousands of US hospitals confirms major data breachJul 20

Healthcare billing giant serving thousands of US hospitals confirms major data breach

71Hugging Face says an external AI agent breached its systems, and frontier-model guardrails blocked the company from analyzing the attackJul 20

Hugging Face says an external AI agent breached its systems, and frontier-model guardrails blocked the company from analyzing the attack

72A researcher used GPT-5.6 and $25 to find a WordPress zero-day that exploit brokers would pay $500,000 for, by adapting OpenAI's own math-solving promptJul 20

A researcher used GPT-5.6 and $25 to find a WordPress zero-day that exploit brokers would pay $500,000 for, by adapting OpenAI's own math-solving prompt

73A cybersecurity researcher got every major AI model to hand over instructions for making weapons and drugs, and says the companies' own safety guardrails are the key to breaking themJul 20

A cybersecurity researcher got every major AI model to hand over instructions for making weapons and drugs, and says the companies' own safety guardrails are the key to breaking them

74A single tampered weather sensor at Paris CDG airport netted a trader $20,000 on Polymarket, and experts warn AI-driven forecasting is heading toward a data-sabotage crisisJul 20

A single tampered weather sensor at Paris CDG airport netted a trader $20,000 on Polymarket, and experts warn AI-driven forecasting is heading toward a data-sabotage crisis

75Meta patented an AI system that listens to your voice all day, logs your emotional state, and tracks whether you have taken your medicationJul 20

Meta patented an AI system that listens to your voice all day, logs your emotional state, and tracks whether you have taken your medication

76LG monitors are silently installing adware through Windows Update, exploiting the same trusted channel meant to protect your PCJul 19

LG monitors are silently installing adware through Windows Update, exploiting the same trusted channel meant to protect your PC

77OpenAI built an LLM whose only job is hacking other LLMs, and it already found attack methods humans missedJul 16

OpenAI built an LLM whose only job is hacking other LLMs, and it already found attack methods humans missed

78Your AI assistant's memory is a goldmine and barely protected: researcher tricked Claude into leaking full name, employer, and security answers through its own web browserJul 16

Your AI assistant's memory is a goldmine and barely protected: researcher tricked Claude into leaking full name, employer, and security answers through its own web browser

79Cursor 0day: AI coding tool with 7M+ users has unfixed arbitrary code execution vulnerability after 7 months of vendor silenceJul 15

Cursor 0day: AI coding tool with 7M+ users has unfixed arbitrary code execution vulnerability after 7 months of vendor silence

80The Multi-Agent Security Blind Spot: Why Per-Message Monitoring Cannot Stop Distributed AttacksJul 14

The Multi-Agent Security Blind Spot: Why Per-Message Monitoring Cannot Stop Distributed Attacks