20+ hour, 56+ min ago   (568+ words) Cloudflare is using OpenAI GPT-5.6 Cyber to automate vulnerability discovery and remediation as AI agents reshape enterprise security operations. Cloudflare has announced Vulnerability Discovery and Remediation, a new service designed to automate vulnerability discovery, generate code fixes, and deploy edge…...

3+ day, 14+ hour ago   (861+ words) Security teams are building custom AI agents to improve defense, close tooling gaps, and automate workflows, but strong governance remains critical. For security professionals, they’re inescapable: nearly every vendor is trying to sell a product that uses them, to the…...

3+ day, 19+ hour ago   (587+ words) PaperCut pre-authentication RCE: A PaperCut remote code execution flaw is being exploited in the wild, with Huntress observing attacks against two customers. PaperCut considers all NG and MF versions potentially affected. Organizations should update PaperCut immediately, remove public access, preserve…...

5+ day, 12+ hour ago   (506+ words) CrowdStrike SafeMind pairs offensive and defensive AI models in a closed loop designed to test attack paths and strengthen enterprise cyber defenses. Cyber defenders could soon have an AI system capable of testing attacks and developing defenses in the same…...

6+ day, 18+ hour ago   (695+ words) China-linked Fire Ant hackers compromised Cisco IOS XR routers, management hosts, and authentication systems to create covert paths into other networks. Compromising a server can expose one machine. Compromising the right router can give attackers visibility into an entire network....

1+ week, 6+ day ago   (696+ words) A suspected Iran-linked cyberattack reportedly forced a small UK power generator offline for four days, raising concerns about infrastructure resilience. A small British power generator spent four days offline after a suspected Iran-linked cyberattack. The incident happened in July 2026 and…...

2+ week, 4+ day ago   (636+ words) Researchers used a remote Spectre attack to leak a JWT between co-located Cloudflare Workers. Cloudflare says the technique is now mitigated. A remote Spectre attack successfully pulled a JWT from a co-located Cloudflare Worker, showing how CPU side channels can…...

3+ week, 3+ day ago   (530+ words) Anthropic found Claude agents sabotaging rivals and escalating to self-replicating malware during controlled tests of multiagent AI behavior and conflict. Researchers were testing how autonomous models behave when several agents work in the same environment under incompatible goals. Some Claude…...

3+ week, 3+ day ago   (891+ words) Threat actors are using AI to accelerate intrusions, steal credentials, and identify high-value targets. AI is becoming a bigger part of real-world cyberattacks, helping threat actors conduct intrusions, steal credentials, navigate networks, and identify valuable targets. A Gambit Security investigation…...

3+ week, 4+ day ago   (766+ words) Cybercriminals are developing indirect prompt injection tools to target AI agents. Proofpoint researchers found that cybercriminals are developing and selling tools designed to hide malicious instructions inside emails, documents, calendar invites, and webpages that AI systems routinely process. This could…...