LodeHQ
AI news for people who shipFree, by email

Infosec · Tuesday, 11 August 2026 · Issue 71 · 10 stories

Apple PCC breached, KVM host memory pwned, iOS exploits leaked

Vulnerabilities & Exploits

Critical RCE in Connective eID Extension threatens 8/10 Belgian Banks

Researchers found that Connective’s eID signing extension, used by over two million users, eight of Belgium’s ten biggest banks and dozens of government services, lets any website read smart‑card data, steal eID PINs and trigger remote‑code execution via a malicious library. The flaw gave attackers silent, full‑control over user credentials, prompting a 146‑day patch cycle.

amibeingpwned.com ↗14 min read at source

DarkSword and Coruna iOS Exploit Chains Now in Criminal Hands

Lookout uncovered that the sophisticated iOS exploit kits Coruna and DarkSword, once limited to nation‑state actors, are being wielded by the Russian‑linked UNC6353 group to steal crypto wallets and personal data. The finding shows a dangerous commoditization of mobile exploits, while Lookout’s Safe Browsing and device‑compromise tools block the threat on supported iOS versions.

lookout.com ↗12 min read at source

Threats & Malware

Ransomware ‘Gentlemen’ hijack AnMed’s Facebook, claim 6 TB of sensitive health records

The ‘Gentlemen’ ransomware group posted on AnMed’s Facebook page, saying they stole 6 TB of patient data, including SSNs, reproductive‑health, mental‑health and sexual‑assault records. The page was taken down within an hour, but the claim signals a massive breach of highly sensitive health information and raises fresh concerns about ransomware extortion tactics targeting healthcare providers.

foxcarolina.com ↗1 min read at source

DeadLock ransomware leverages Polygon contracts and Session chat to thwart takedowns

Microsoft's latest analysis shows DeadLock ransomware now runs its extortion workflow on Polygon smart contracts and the Session messaging network, moving ransom negotiations and leak hosting off‑premise. This decentralized architecture, coupled with geofencing and throttling tricks, has let the group hit over 80 organizations worldwide while evading traditional disruption efforts.

microsoft.com ↗25 min read at source

GhostSplice lets AI coders silently steal SSH keys and .env files

A proof‑of‑concept on GitHub shows how a malicious MCP server can split a theft request across harmless‑looking tool descriptions and results, bypassing AI coding assistants’ refusal safeguards. The technique, dubbed GhostSplice, lets agents exfiltrate SSH keys, environment files, source code and customer data without triggering any single‑step filter.

github.com ↗5 min read at source

Privacy, Policy & Governance

You’re caught up.

That was all 10 stories in the 11 August issue.