LLMs are Getting a Lot Better and Faster at Finding and Exploiting Zero-DaysSchneier on Security·Feb 9, 12:04 UTC · Feb 9, 2026Exploit / PoC60
Anthropic's new AI model finds and exploits zero-days across every major OS and browserHelp Net Security·Jun 19, 12:14 UTC · Jun 19, 2026Exploit / PoC in the wildCVE-2026-474760
Anthropic rolls out embedded security scanning for ClaudeCyberScoop·Feb 20, 21:41 UTC · Feb 20, 2026Exploit / PoC in the wild60
Critical OpenWrt DHCPv6 Flaw Could Let Unauthenticated Attackers Run Code as RootThe Hacker News·Jul 28, 13:10 UTC · Jul 28, 2026Exploit / PoC in the wildCVE-2026-53921CVE-2026-62948CVE-2026-62947160
Anthropic Releases Claude Fable 5, Its Most Powerful AI Yet, With Cyber SafeguardsThe Hacker News·Jun 10, 07:37 UTC · Jun 10, 2026Exploit / PoCCVE-2026-4747160
Flaw in Claude’s Chrome extension allowed ‘any’ other plugin to hijack victims’ AICyberScoop·May 8, 13:14 UTC · May 8, 2026Exploit / PoC in the wild60
Claude Mythos Has Found 271 Zero-Days in FirefoxSchneier on Security·Apr 29, 10:12 UTC · Apr 29, 2026Exploit / PoC60
Claude Mythos finds 271 Firefox flaws, Mozilla believes it shifts security toward defendersHelp Net Security·Apr 22, 00:00 UTC · Apr 22, 2026Exploit / PoC60
Top AI Agents Built to Catch Malicious Code Can Be Tricked Into Running ItThe Hacker News·Jul 9, 05:17 UTC · Jul 9, 2026Exploit / PoCCVE-2026-3986150
Anthropic and OpenAI Security Tools Could Fuel CyberInfosecurity Magazine·Jul 10, 13:45 UTC · Jul 10, 2026Exploit / PoC60
Week in review: Firmware-level Android backdoor found on tablets, Dell zero-day exploited since 2024Help Net Security·Feb 22, 00:00 UTC · Feb 22, 2026Exploit / PoC in the wildCVE-2026-2441CVE-2026-22769CVE-2026-2329+1 CVEs60