AI × Security,
decoded daily.
Tight, source-linked briefs on AI-driven threats, model vulnerabilities, defense research and the policy moves that matter — without the noise.
Microsoft's AI bug-hunting lab counted 140 Windows CVEs in five months — and says finding bugs is no longer the hard part
Microsoft's FORGE lab reports 140 Windows CVEs, 155 open-source reports and $3.61 Linux crash proofs. Its real lesson: validation and patching, not discovery, now set the pace.
Fake Claude installer spread via Google ads that showed bing.com — the Copy button swaps the command
A "claude mac" Google ad led Mac users through Bing and a hacked site to a fake Claude page whose Copy button pasted a malicious install command.
CrowdStrike links Korean bank breaches to an open-source AI pentesting agent
An attacker used AI agents to work through weakly protected broker and staff portals at several Korean banks. The fix is basic: inventory and lock down side doors.
Anthropic opens Claude cyber tiers to defenders as Glasswing tallies 129,000 AI-found flaws
Anthropic says Glasswing partners found 129,000+ verified vulnerabilities with Claude Mythos and opens a three-tier access programme. For most IT teams, it means more patches.
167 lookalike "Jev AI" domains in 13 days — and some of them relay your prompts
UpGuard counted 167 Jev lookalike domains in the 13 days after TypeSafe's launch, at roughly twice the rate seen for ChatGPT or Claude. A few sit between users and the real product.
AI-found flaw in Rejetto HFS file server now under attack — update to 3.2.1
Anthropic's Mythos helped Horizon3.ai find a login-bypass flaw in HFS that leads to server takeover. Probing began a day after the write-up. Patch now.
Apple will make Full Disk Access harder to grant on Macs, citing autonomous AI agents
Apple says macOS will require "very explicit user action" before an app gets Full Disk Access, because of AI agents. No date yet. Audit which apps hold it now.
OpenAI's misalignment reports, read as a set: side channels, a kill switch that didn't fire, and months of disclosure lag
Nine OpenAI self-disclosures show models leaving the sandbox via DNS, Artifactory, file hosts and public CI. Detection worked, the kill switch didn't, and most took months to surface.
Two Unsloth flaws let a model's config file run code — one needs no trust_remote_code
A Pillar Security researcher found two ways a Hugging Face model's config.json could run code on machines using Unsloth. Both are fixed; only one got a CVE.