Install
Please confirm you are human
This browser or connection looks automated. Press and continuously hold the control for 3 seconds to enable Google-hosted web results and, when separately allowed, AI-assisted answers.
A successful check enables 100 search requests. Interactive access does not authorize scraping, systematic collection, or reuse of search output.
News
OpenAI Says Its Model Found a Zero-Day by Itself, Without Seeing Source Code
23+ hour, 43+ min ago (692+ words) By Nokka | September 11, 2026 OpenAI published an incident report describing how its model breached Hugging Face, stating the model discovered a previously unknown vulnerability in another company's software on its own [1]. The incident started as a cyber capability evaluation, not a…...
Rogue AI didn’t breach Hugging Face, human decisions did
2+ day, 3+ hour ago (651+ words) By Eryk Salvaggio | Analysis | September 11, 2026 Editor’s note: A version of this piece originally appeared in the Cybernetic Forests newsletter and is published here with permission. The incident, as described by the media, involved more than 1,000 autonomous AI agents breaking out…...
Recent WeChat security flaw shows how quickly AI can accelerate mobile hacking research
2+ day, 23+ hour ago (628+ words) A recent security demonstration by the computer security and research firm Calif showed how a software bug could become a serious cyberattack through an exploit in the WeChat app. Calif researchers found a flaw in WeChat that could be triggered…...
Another Anthropic model gained access to open internet during testing
2+ day, 9+ hour ago (731+ words) Anthropic disclosed on Wednesday that another one of its Claude models mistakenly gained access to the open internet during a cybersecurity exercise, marking the fourth time its models have done so. An early version of the Claude Opus 4.6 model connected…...
Anthropic Discloses Four Incidents of Claude Models Accessing Real Systems
3+ day, 1+ hour ago (296+ words) Anthropic has disclosed four separate incidents in which Claude AI models gained unauthorized access to real-world computer systems during cybersecurity evaluations, highlighting the difficulty of safely isolating increasingly capable AI agents in testing environments. According to Anthropic, the incidents involved…...
Another Anthropic model gained access to the open internet in 4th such incident
3+ day, 5+ hour ago (731+ words) Anthropic disclosed on Wednesday that another one of its Claude models mistakenly gained access to the open internet during a cybersecurity exercise, marking the fourth time its models have done so. An early version of the Claude Opus 4.6 model connected…...
A New Claude 's Sandbox Failure Shows How AI Can Rationalize Real-World Harm
2+ day, 20+ hour ago (435+ words) U.S. CISA adds Microsoft Windows, N-able N-central, and Adobe flaws to its Known Exploited Vulnerabilities catalog Four Nation-State Actors Used the Same Chrome Zero-Day Exploit Kit Within 12 Days US Agencies Warn Chinese AI Firms Are Extracting Advanced AI Models Google fixes…...
Anthropic says Claude model hacked third-party system after accidental internet access
3+ day, 2+ hour ago (583+ words) Anthropic has disclosed a fourth incident in which a Claude model accessed the open internet during a cybersecurity test. The model breached a third-party system after a misconfigured environment prevented it from completing its assigned task, raising fresh questions about…...
Another Anthropic model gained access to the open internet, company says
3+ day, 6+ hour ago (740+ words) September 9, 2026 / 8:19 PM EDT / CBS News Anthropic disclosed on Wednesday that another one of its Claude models mistakenly gained access to the open internet during a cybersecurity exercise, marking the fourth time its models have done so. An early version of…...
Anthropic details four incidents where Claude gained unauthorized access to third-party systems, including a new Opus 4.6 case; METR will investigate them
3+ day, 11+ hour ago (12+ words) Top news and commentary for technology's leaders, from all around the web....