SecurityWeek

Dubbed MikroTrick, the bugs allow attackers to bypass authentication, overwrite configuration files, and take over devices.

Hackers stole the information of students, teachers, staff, and parents/guardians from a self-hosted Metabase instance.

Administrators are advised to check their deployments for newly created user accounts they don’t recognize.

The proof-of-concept (PoC) exploits lead to privilege escalation, spawning a shell with System privileges.

The stealthy toolkit embeds a backdoor in HAProxy and targets automotive and media organizations in South Korea for long-term surveillance.

OpenAI agents made 15,000–18,000 autonomous edits to a German wiki over three months, evading moderation and echoing tactics seen in the Hugging Face breach.

The StyleSmuggler zero-day allows attackers to execute code and deploy a stealthy backdoor on Adobe Commerce and Magento stores.

The attacks rely on backdoored ScreenConnect instances to transfer and execute payloads to newly connected clients.

Tracked as CVE-2026-32475 (CVSS score of 9.8), the bug described as an arbitrary file upload issue in the function that handles form submissions.

Noteworthy stories that might have slipped under the radar: Microsoft rolled out patches for cloud services, hackers compromised 5,000 Dropbox accounts, and Guardio is now valued at $1.1 billion.

Nearly two dozen issues, tracked collectively as CVE-2026-73749 (CVSS score of 9.8), were addressed with the updates.

The Daybreak initiative will provide subsidized AI cyber capabilities, training and technical assistance, though OpenAI has disclosed few details about costs and eligibility.

Tracked as CVE-2026-9586, the unauthenticated SQL injection flaw can be exploited remotely for arbitrary code execution.

Dubbed PostGREShell, CVE-2026-6471 turns low-level replication access into code execution, permanent superuser privileges and a persistent database backdoor.

Catch promises the capabilities of a trusted executive assistant, with built-in controls governing what data and systems it can access.

The flaws could allow attackers with administrative access to a virtual machine to execute code on the host system.

Google’s Chrome 152 security update resolves 12 vulnerabilities, including a high-severity type confusion flaw in the V8 engine.

The deal highlights Nvidia’s push to champion increasingly popular open-source AI models.

Hacker group published roughly 550GB of data after MAG reportedly refused to pay a ransom demand; the group says it gained access via exposed admin keys.

New models, trained using NVIDIA Nemotron 3 Ultra, aim to catch rogue agent behavior before it executes, without the latency of large-model review.