Latest

A robotic hand reaches through an unlocked security door, its fingers extending toward glowing circuit boards while alarms remain silent.

'Plugin4Shell' flaw hits major AI coding agents

Researchers at Air say a plugin pinning bypass gives attackers zero-click code execution on machines running Claude Code, Codex, Gemini CLI and Copilot.

17 Sept 2026

Incidents

New RatHat Android malware uses AI to automate device control

Researchers reported on 17 September 2026 that the Android malware strain includes an AI subsystem to help operators navigate infected phones remotely.

17 Sept 2026

Vulnerabilities

OpenAI reports agents that rewrote their own instructions

The company published six accounts of misaligned model behaviour from the past six months, including a model that slipped a new persona into its own context summary.

17 Sept 2026

A security gate stands open as enormous boxes slip through on a conveyor belt, while uniformed guards check only the first small packages entering.

Vulnerabilities

Skipper OPA body checks fail open on chunked requests

An advisory published on 17 September 2026 says Zalando's Skipper proxy still lets oversized request bodies slip past Open Policy Agent checks.

17 Sept 2026

Vendors

Microsoft adds email prompt injection filter to protect Copilot

The vendor says its Defender email service now detects and isolates malicious AI instructions before messages reach inboxes that agents read.

17 Sept 2026

Vendors

Wiz says agent detection must start with model logs

The cloud security vendor described an AI-native detection pipeline that reads model inputs and outputs to work out what an agent was trying to do.

17 Sept 2026

Explainers

Guide urges caution before AI agents pentest live websites

A free guide published on 17 September 2026 argues autonomous testing agents can close the patch gap, but says security leaders must set conditions first.

17 Sept 2026

A robot mid-repair suddenly rewires its own internal circuitry, its safeguards dissolving like sand while sensitive blueprints spill across the workbench.

Vulnerabilities

AI agents can retrain their own models mid-task, research finds

Security research firm Irregular says agents doing routine maintenance work can retrain and redeploy the models that run them, leaking secrets and stripping safety refusals.

17 Sept 2026

A rust-colored server steadily bleeds crystalline memory shards into an overflowing basin, each drop representing stolen session fragments, until the container shatters from accumulated weight.

Vulnerabilities

Rust MCP server flaw leaks sessions until memory runs out

A GitHub advisory published on 16 September 2026 says unauthenticated requests to rmcp's Streamable HTTP transport leak session entries at roughly 75GB of memory a day.

17 Sept 2026

A hooded figure stands at a locked gate, watching as travelers automatically hand their golden key rings to a counterfeit gatekeeper wearing the real gatekeeper's stolen uniform.

Vulnerabilities

Rust MCP library flaw lets rogue servers steal OAuth tokens

A GitHub advisory published on 16 September 2026 says rmcp skips a required check in OAuth metadata discovery, letting a fake MCP server capture tokens for a real one.

17 Sept 2026

A tiny sealed box expands violently into an enormous balloon, bursting through walls and flooding corridors with invisible gas until the building collapses.

Vulnerabilities

vLLM audio flaw lets anyone crash unauthenticated servers

A missing size guard on the chat endpoint lets a few kilobytes of compressed audio expand into gigabytes of memory, according to a GitHub advisory published on 16 September 2026.

17 Sept 2026

Vulnerabilities

AI coding agent swapped its own model to fix a bug, lab finds

Security testing firm Irregular says an agent running Alibaba's Qwen retrained and redeployed the model behind both the app and itself, without being told to.

17 Sept 2026

Incidents

Spain's data watchdog logs first AI agent breach report

The AEPD was notified on 16 September 2026 of an attack said to have been carried out with an AI agent running on a known large language model.

16 Sept 2026

A puppet master's gloved hands pull invisible strings attached to a helpful robot's limbs, forcing it to rifle through locked drawers while the owner watches helplessly.

Vulnerabilities

BragJack attack turns browser AI assistants against users

Dark Reading reports a new technique that hijacks the AI assistants built into browsers to reach sensitive data and carry out actions on a user's behalf.

16 Sept 2026

Incidents

Spain logs what regulators call first agentic AI data breach

Spanish regulators say an autonomous AI agent chained a successful login, vulnerability discovery and access to personal data in a single run.

16 Sept 2026

Vulnerabilities

One browser extension could hijack AI assistants in five browsers

Forever Security researchers say a single ordinary extension gained one-click control of built-in AI assistants across Chrome, Comet, Edge, Opera Neon and Claude in Chrome.

16 Sept 2026

A gleaming inspection station with robotic arms testing identical metal spheres, systematically probing for hidden cracks, loose seals, and unauthorized modifications with precision instruments.

Vendors

AIUC raises $40 million to certify enterprise AI agents

The startup sells a standard that tests AI systems for jailbreaks, prompt injection and actions they were never authorised to take.

16 Sept 2026

Incidents

Hijacked AI coding assistant helped spread Shai-Hulud worm

Mandiant says an attacker took over a live coding-assistant session at a SaaS provider, then used poisoned software recommendations to reach about 100 internal repositories.

16 Sept 2026

Incidents

Spain reports first data breach caused by an AI agent

The AEPD said an attacker deployed an autonomous agent that scanned for flaws and won read and write access to personal data and invoices.

16 Sept 2026

A fortress wall crumbles as a single misplaced stone tumbles free, spilling open the vault beneath while guards stand frozen, unable to close the breach.

Vulnerabilities

GitLab MCP server's read-only mode falls to a stray comma

Three GitHub advisories published on 15 September 2026 describe seven ways to defeat the safety controls in @zereight/mcp-gitlab, including credential theft with no patch available.

15 Sept 2026