Daily Briefing
Today's outlook
Chrome fix closes an actively exploited flaw — CISA lists CVE-2026-125350 as known-exploited
Good morning. Patch Chrome first. CISA lists CVE-2026-125350 in its Known Exploited Vulnerabilities catalog as known to be exploited. Google's Chrome releases blog fixes it in 157.0.8103.50; the advisory record gives it CVSS 8.8 (CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H), affecting Chrome up to that fixed release. The same stable update addresses CVE-2026-125347, rated CVSS 9.6 (CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:C/C:H/I:H/A:H), also fixed in 157.0.8103.50.
A second Chrome stable update, 157.0.8103.78, fixes four more: CVE-2026-126885, CVE-2026-126872 and CVE-2026-126859, each rated CVSS 9.6 with the same vector, and CVE-2026-126915 at CVSS 7.5 (CVSS:3.1/AV:N/AC:H/PR:N/UI:R/S:U/C:H/I:H/A:H). All require user interaction. Match your installed Chrome build against the fixed releases Google names — 157.0.8103.50 and 157.0.8103.78 — and update to the fixed release, then restart the browser so the update applies.
On the supply-chain front, BleepingComputer reported that the FBI is warning hackers are using an AI-powered penetration testing framework called ARTEX to automate attacks worldwide, including a recent ConnectWise supply-chain breach that compromised Microsoft 365 customer environments. Organisations running ConnectWise or managing Microsoft 365 tenants through it should treat this as relevant and review the vendor's guidance.
The Hacker News reported that the compromise of Anthropic's official Claude Code extension on Cursor's extension marketplace resulted from the theft of an access key. Cursor said in a Friday update that an attacker obtained an access key it had issued to Fireworks, exposed in the wake of a breach targeting Fireworks AI. Help Net Security's week-in-review covered the same thread — a secret stealer hitting GitHub repositories and Claude Code turned into a remote-controlled implant. If you use the Claude Code extension via Cursor, follow Cursor's and Anthropic's updates and rotate any exposed credentials.
BleepingComputer reported that TSMC confirmed personal employee data was leaked online after a threat actor put 6.9GB of information up for auction, with the breach linked to a cloud provider.
The Hacker News reported that Italy's data protection authority, the Garante, fined Meta €805 million over a Facebook for Android feature that suggests gallery photos and videos to post; the Garante found the feature was enabled by default and accessed Italian users' photos, videos and related data.
Zoom out: Supply-chain compromise runs through two of the day's items — the FBI's ARTEX/ConnectWise warning and the Cursor extension trojanised via a stolen Fireworks AI key.
Vulnerability in focus
CVE-2026-125350 — Google. CVSS 8.8 CISA lists it as known to be exploited.
Affected: chrome.
What to do: Follow the vendor advisory for the fixed release and any interim mitigation.
What we're tracking
- BleepingComputer: FBI warns ARTEX AI hacking tool is powering global cyberattacks Read it
- The Hacker News: Italy Fines Meta €805 Million Over Facebook's Android Gallery Access Without Explicit Consent Read it
- Help Net Security: Week in review: Secret stealer hits GitHub repos, Claude Code turned into remote-controlled implant Read it
Sources
- FBI warns ARTEX AI hacking tool is powering global cyberattacks BleepingComputer
- Italy Fines Meta €805 Million Over Facebook's Android Gallery Access Without Explicit Consent The Hacker News
- Week in review: Secret stealer hits GitHub repos, Claude Code turned into remote-controlled implant Help Net Security
- TSMC employee data breach linked to cloud provider BleepingComputer
- Stolen Fireworks AI Key Led to Cursor Data Breach and Trojanized Claude Code Extension The Hacker News
Share this issue
Facebook · X · Reddit · LinkedIn · WhatsApp · Email · Bluesky
Summarized from the linked reports and the advisory record by the desk. Verify against the original sources before citing.