4 min read
News Brief GitLab fixed a critical flaw in its self-hosted AI Gateway that let an authenticated Duo Agent Platform user escape the prompt template sandbox and run arbitrary commands on the host.
AI Security Wire
Tracking AI threats, vulnerabilities, and defensive strategies for security professionals.
Latest Posts
View all articles → 
4 min read
News Brief NVIDIA fixed a maximum-severity CVSS 10.0 vulnerability in its AIStore framework for Linux, one of eight flaws disclosed in the October 13 bulletin covering the AI storage and ETL platform.
6 min read
Defensive Techniques Researchers from HKUST and NUS propose a tagging-and-filtering defense that asks LLMs to cite which instruction they are answering, cutting prompt injection success rates from over 80% to under 1% without any fine-tuning.
4 min read
News Brief OpenCode disclosed a breach of a legacy console system that exposed OpenAI, Anthropic, and Google API keys for 1.4% of its users, linking the intrusion to a distillation campaign.
6 min read
Threat Actors SOCRadar uncovered a phishing-as-a-service platform that rents out AI voice agents posing as Apple Support to trick stolen-iPhone victims into handing over their Activation Lock passcode.
4 min read
News Brief GitLab fixed a critical flaw in its self-hosted AI Gateway that let an authenticated Duo Agent Platform user escape the prompt template sandbox and run arbitrary commands on the host.
4 min read
News Brief NVIDIA fixed a maximum-severity CVSS 10.0 vulnerability in its AIStore framework for Linux, one of eight flaws disclosed in the October 13 bulletin covering the AI storage and ETL platform.
4 min read
News Brief OpenCode disclosed a breach of a legacy console system that exposed OpenAI, Anthropic, and Google API keys for 1.4% of its users, linking the intrusion to a distillation campaign.
5 min read
Vulnerabilities DEF CON 34 research from Cyera Research Labs traced a single architectural flaw in the Pyodide WebAssembly Python runtime into seven products, including two AI code-execution sandboxes, Cohere's Terrarium and Hugging Face's smolagents, yielding four CVEs rated 8.3 to 9.9.
6 min read
Vulnerabilities A critical flaw in ComfyUI's LoadTrainingDataset node lets unauthenticated attackers upload a malicious pickle shard and trigger arbitrary code execution through torch.load, exposing the most widely used Stable Diffusion interface.
5 min read
Vulnerabilities A critical prompt injection vulnerability in Upstash's Context7 documentation server, installed by millions of developers, has no documented fix four days after disclosure. Researchers say it may be a regression of a bug patched in February.
6 min read
Threat Actors SOCRadar uncovered a phishing-as-a-service platform that rents out AI voice agents posing as Apple Support to trick stolen-iPhone victims into handing over their Activation Lock passcode.
6 min read
Threat Actors ESET discovered PromptSpy in February 2026 — the first known Android malware to query a live generative AI API at runtime. It uses Google Gemini to parse on-screen UI state and issue gesture instructions that keep the malware alive on infected devices.
6 min read
Threat Actors Sysdig documented the first confirmed case of an LLM agent autonomously executing a complete ransomware operation: initial access, lateral movement, credential harvesting, encryption, and extortion without human steering on any technical decision.
6 min read
Research A new benchmark from Shanghai AI Laboratory tests six leading GUI agents against environmental injection attacks embedded in real Android apps. Every agent is vulnerable, attack success rates reach 66.9%, and stronger agents turn out to be more exploitable, not less.
6 min read
Research Anthropic and EPFL researchers demonstrate that ideological and action payloads can spread between AI agents through editable system prompt state files, surviving 20-hop chains with no human intervention. One defensive measure stops them almost entirely.
5 min read
Research 1Password's Off-by-1 Labs tested two frontier AI models against six real CVEs and found that only a quarter of the generated patches fully fixed the vulnerability without introducing new problems. The implications for teams relying on AI-assisted remediation are significant.
6 min read
Defensive Techniques Researchers from HKUST and NUS propose a tagging-and-filtering defense that asks LLMs to cite which instruction they are answering, cutting prompt injection success rates from over 80% to under 1% without any fine-tuning.
6 min read
Defensive Techniques Google's Mandiant disclosed technical details of its Agentic Vulnerability Discovery Harness (AVDH), a multi-agent AI pipeline that found over 100 true-positive critical vulnerabilities in a stolen-repository incident response in two days, with 12 CVEs assigned so far.
6 min read
Defensive Techniques Tracebit research shows a single planted string engineered to trigger an AI model's safety guardrails can cut successful AI-driven AWS attacks by more than 80 percent.
4 min read
Incident Reports A service called Poison Claude resold Claude API access at a fraction of the official price by routing requests through compromised AWS Bedrock accounts, giving operators full visibility into every customer prompt. A configuration error revealed nearly 900 active users had been sending sensitive queries through a third-party proxy they didn't know was reading their traffic.
5 min read
Incident Reports Anthropic disclosed on July 31 that three of its models — Claude Opus 4.7, Mythos 5, and an unreleased internal prototype — breached real companies during cybersecurity capability evaluations after an evaluation partner misconfigured network egress. The models used basic techniques: weak passwords, unsecured endpoints, SQL injection. Mythos 5 never concluded it had left the simulation.
5 min read
Incident Reports An attacker deployed Nous Research's open-source Hermes AI agent in YOLO mode against Thailand's Ministry of Finance, autonomously running reconnaissance, privilege escalation, and database exploitation with no operator in the loop.