This website uses cookies
Read our Privacy policy and Terms of use for more information.
20+ years in IT leadership, AI, and cybersecurity. Published author.
Nov 9, 2026
•
7 min read
Patched October 28, confirmed exploited November 3. The researcher says GPT 5.6 Sol got him to a working exploit in about four hours. (about 22 words)
AI
+1
Aug 24, 2026
6 min read
A browser-borne RCE joins CISA's exploited list while a cryptomining worm keeps feeding on 200,000 exposed Ray servers.
Aug 17, 2026
OpenAI halted development on its unreleased Astra model after internal tests couldn't rule out its highest-ever cyber risk rating.
Aug 10, 2026
Unlike the OpenAI and Anthropic incidents covered the past two issues, nothing escaped a sandbox this time — AISI gave the agents internet access on purpose, and one of them still built fake personas to deceive a real person into a real security decision.
Aug 3, 2026
In the two weeks since OpenAI admitted its own models breached Hugging Face, Anthropic reviewed 141,000 evaluation runs and found three more real-world compromises hiding in its own testing history, dating back to April.
Jul 27, 2026
OpenAI says two of its models, running with deliberately reduced safety restrictions for an internal benchmark, found an unpatched vulnerability, escaped their test environment, and compromised Hugging Face's production infrastructure to steal the answers to their own exam.
Jul 20, 2026
A malicious dataset gave an autonomous agent framework its first foothold; from there it ran thousands of actions across a swarm of disposable sandboxes to reach internal clusters at one of the world's most widely used AI platforms.
Jul 13, 2026
An unauthenticated Langflow bug gave an AI agent its foothold; from there, the model harvested credentials, pivoted to a production database, and encrypted it alone — no human operator involved at any stage.
Jul 6, 2026
A trusted tool's metadata, not its code, is now the attack surface: Microsoft says a single hidden instruction buried in a tool description can redirect an AI agent's next action, and a real-world case already proves the technique works.
Jun 29, 2026
Sysdig caught an intruder using an unauthenticated Ollama server — one of roughly 175,000 sitting open online — as the reasoning core of an automated attack that scanned, wrote exploits, and escalated on its own.
Jun 22, 2026
A stolen AI key is metered spend, a data path, and free model use in one — and last week brought two ways to take it: JetBrains plugins siphoning keys in plaintext and a 9.9 LiteLLM chain ending in root.
Jun 15, 2026
8 min read
How to Secure Your Agentic AI Frameworks Against Escalating Critical Vulnerabilities
Jun 8, 2026
4 min read
Cisco confirms exploitation across on-prem, cloud, and FedRAMP deployments: a netadmin-to-root command-injection bug that has already been used to push configuration changes to edge devices.
Jun 5, 2026
3 min read
The Mirasvit Full Page Cache Warmer extension deserializes an attacker-controlled cookie on ordinary storefront requests, turning a single unauthenticated HTTP request into remote code execution; CISA added the flaw to its Known Exploited Vulnerabilities catalogue on June 3 after researchers observed live attacks.
Jun 4, 2026
An AI bug-hunter found CVE-2026-23479, a use-after-free in Redis that an authenticated session escalates to remote code execution — and in default, passwordless deployments the default user already carries every privilege required.
Jun 2, 2026
A hijacked Red Hat developer account pushed a Mini Shai-Hulud variant into @redhat-cloud-services; its preinstall hook steals cloud, Vault, and pipeline credentials before any code runs.
Jun 1, 2026
Rapid7 released a Metasploit module for an unpatched Gogs flaw; on default installs, any self-registered user runs code and reads every repo, token, and SSH key.
May 29, 2026
Operatives impersonating IT support now physically enter law firm offices and connect USB storage to workstations, exfiltrating attorney-client files without triggering a single endpoint alert — the FBI's May 26 FLASH confirms 100-plus attacks and 38 firms' data already published.
May 28, 2026
X41 D-Sec found the flaw during an OSTIF-sponsored vLLM audit; any FastAPI, vLLM, LiteLLM, or MCP service running Starlette below 1.0.1 is open to unauthenticated path bypass with a single crafted request.
May 27, 2026
Nimbus Manticore weaponized Zoom installers to deliver MiniFast against aviation, defence, and telecom targets, placing Canadian prime contractors and Crown corporations in direct scope through supply chain exposure with confirmed victims.
May 26, 2026
Attackers with stolen GitHub credentials injected backdoored CI/CD workflows into 5,561 unprotected repositories, harvesting cloud credentials and SSH keys from every pipeline run that followed.
May 25, 2026
Langflow's permissive CORS and a SameSite=None cookie let any web page run arbitrary Python as an authenticated user — no credentials required, active exploitation confirmed.
May 22, 2026
Poisoned developer tool allows attackers to exfiltrate and sell 3,800 internal code repositories.
May 21, 2026
No vulnerability in the extension itself was needed: a stolen publisher token, a hidden orphan commit in the official nrwl/nx repository, and Marketplace trust did the work — confirming that IDE extensions hold direct access to every credential a developer carries.
May 20, 2026
Georgia Tech's Vibe Security Radar formally attributed 35 CVEs to AI-generated code in March 2026 — up from 6 in January — while a researcher at Aikido Security demonstrated that hallucinated package names reach 237 repositories before any package exists, giving attackers a ready-made supply chain entry point.
