prompt-injection
130 articles with this tag
INFO
INFO
INFO
HIGH
MEDIUM
INFO
HIGH
LOW
MEDIUM
MEDIUM
HIGH
HIGH
MEDIUM
HIGH
HIGH
INFO
INFO
HIGH
INFO
HIGH
INFO
MEDIUM
MEDIUM
INFO
CRITICAL
MEDIUM
MEDIUM
HIGH
LOW
CRITICAL
MEDIUM
MEDIUM
LOW
INFO
INFO
MEDIUM
MEDIUM
HIGH
LOW
MEDIUM
MEDIUM
MEDIUM
MEDIUM
HIGH
MEDIUM
LOW
MEDIUM
LOW
HIGH
MEDIUM
INFO
MEDIUM
INFO
MEDIUM
INFO
MEDIUM
MEDIUM
CRITICAL
LOW
MEDIUM
MEDIUM
INFO
LOW
INFO
MEDIUM
LOW
LOW
INFO
INFO
HIGH
LOW
LOW
LOW
MEDIUM
MEDIUM
MEDIUM
INFO
HIGH
HIGH
INFO
INFO
INFO
LOW
HIGH
LOW
INFO
CRITICAL
CRITICAL
INFO
MEDIUM
MEDIUM
LOW
LOW
INFO
LOW
MEDIUM
MEDIUM
MEDIUM
HIGH
LOW
How credential brokering prevents AI agents from compromising credentials via prompt injection
AI Security CTF (free, open) - prompt injection, agent workflow hijacking, guardrail bypass - June 17-22
AI Security CTF (free, open) - prompt injection, agent workflow hijacking, guardrail bypass - June 17-22
Even Claude agrees: hole in its sandbox was real and dangerous
New image-based prompt injection attack targets multimodal AI models
Warning: AI can give your passwords to hackers. Prompt injection demo
Malicious Coding Agent Skills and the Risk of Dynamic Context | Datadog Security Labs
Malicious AI Prompt Injection Attacks Increasing, but Sophistication Still Low: Google
Indirect prompt injection is taking hold in the wild
AI agents weaponized through indirect prompt injection intrusions
Researchers Uncover 10 In-the-Wild Prompt Injection Payloads Targeting AI Agents
Prompt injection turned Google’s Antigravity file search into RCE
Google Patches Antigravity IDE Flaw Enabling Prompt Injection Code Execution
Copilot & Agentforce offen für Prompt-Injection-Tricks
Claude Code, Gemini CLI, GitHub Copilot Agents Vulnerable to Prompt Injection via Comments
Agents hooked into GitHub can steal creds – but Anthropic, Google, and Microsoft haven't warned users
Prompt injection tags along as GenAI enters daily government use
CrewAI Vulnerabilities Expose Devices to Hacking
Microsoft details AI prompt abuse techniques targeting AI assistants
Agent skill marketplace supply chain attack: 121 skills across 7 repos vulnerable to GitHub username hijacking, 5 scanners disagree by 10x on malicious skill rates (arXiv:2603.16572)
Open, Closed and Broken: Prompt Fuzzing Finds LLMs Still Fragile Across Open and Closed Models
Co-Pilot, Disengage Autophish: The New Phishing Surface Hiding Inside AI Email Summaries
Researchers Discover Major Security Gaps in LLM Guardrails
Speakeasies to Shadow AI: Banning AI Browsers Will Fail
they stole Claude’s brain 16 million times
Red Teaming LLM Web Apps with Promptfoo: Writing a Custom Provider for Real-World Pentesting
Fooling AI Agents: Web-Based Indirect Prompt Injection Observed in the Wild
ClawJacked Bug Enables Covert AI Agent Hijacking
I made Mistral believe Donald Trump runs OpenAI, here's how
RoguePilot Flaw in GitHub Codespaces Enabled Copilot to Leak GITHUB_TOKEN
Compromised npm package silently installs OpenClaw on developer machines
AI coding assistant Cline compromised to create more OpenClaw chaos
Why AI agent containers need a syscall-level observer: the prompt injection blind spot
'God-Like' Attack Machines: AI Agents Ignore Security Policies
Lessons From AI Hacking: Every Model, Every Layer Is Risky
Using threat modeling and prompt injection to audit Comet
Hacking LLMs: How to Prevent Prompt Injection Attacks Now
Cline CLI 2.3.0 Supply Chain Attack Installed OpenClaw on Developer Systems
LLMs change their answers based on who’s asking
PromptSpy ushers in the era of Android threats using GenAI
Crims create fake remote management vendor that actually sells a RAT
PromptSpy is the first known Android malware to use generative AI at runtime
PromptSpy is the first Android malware to use generative AI at runtime
How “Clinejection” Turned an AI Bot into a Supply Chain Attack
PromptSpy Android Malware Abuses Gemini AI to Automate Recent-Apps Persistence
Grok Exposed a Porn Performer’s Legal Name and Birthdate—Without Even Being Asked
Android malware taps Gemini to navigate infected devices
Malicious AI
Compromising Cline's Production Releases just by Prompting an Issue Triager
Hackers can turn Grok, Copilot into covert command-and-control channels, researchers warn
Introducing "AI Unlocked: Decoding Prompt Injection," a New Interactive Challenge
Job scam uses fake Google Forms site to harvest Google logins
HackerOne 'updating' Ts&Cs after bug hunters question if they're training AI
Scammers use fake “Gemini” AI chatbot to sell fake “Google Coin”
EU bans AI over cybersecurity and privacy fears on Parliament devices
Leaking secrets from the claud: AI coding tools are leaking secrets via configuration directories
Researchers Show Copilot and Grok Can Be Abused as Malware C2 Proxies
From Clawdbot to Moltbot to OpenClaw: Security Experts Detail Critical Vulnerabilities and 6 Immediate Hardening Steps for the Viral AI Agent
I pretended to be an AI agent on Moltbook so you don’t have to
AI in the Middle: Turning Web-Based AI Services into C2 Proxies & The Future Of AI Driven Attacks
Hobby coder accidentally creates vacuum robot army
Why 2025’s agentic AI boom is a CISO’s worst nightmare
Ireland now also investigating X over Grok-made sexual images
Prompt Injection Standardization: Text Techniques vs Intent
Microsoft Finds “Summarize with AI” Prompts Manipulating Chatbot Recommendations
The Promptware Kill Chain
ChatGPT gets new security feature to fight prompt injection attacks
Why do I not use “AI” at OSNews?
I built a free, open-source platform to learn GenAI security, learning content + hands-on labs against real LLMs (beta, looking for feedback)
LayerX reports vulnerability in Claude Desktop Extensions, Anthropic declines to fix
Ars Technica Pulls Article With AI Fabricated Quotes About AI Generated Article
Vercel Security Checkpoint
fedoraproject.org
Microsoft: Poison AI buttons and links may betray your trust
'Summarise with AI' can secretly sway recommendations, researchers warn
Microsoft warns of AI recommendation poisoning attacks
PromptJacking: The Critical RCEs in Claude Desktop That Turn Questions Into Exploits
Claude extensions open a security hole in endpoints
Claude add-on turns Google Calendar into malware courier
Extra Packages for Enterprise Linux (EPEL)
fedoraproject.org
fedoraproject.org
Google fears massive attempt to clone Gemini AI through model extraction
Fake AI Assistants in Google Chrome Web Store Steal Passwords and Spy on Emails
Those 'Summarize With AI' Buttons May Lying to You
1Password open sources a benchmark to stop AI agents from leaking credentials
Apple patches decade-old iOS zero-day, possibly exploited by commercial spyware
Microsoft's Notepad Got Pwned (CVE-2026-20841)
What CISOs need to know about the OpenClaw security nightmare
Google: China's APT31 used Gemini to plan cyberattacks against US orgs
Microsoft warns that poisoned AI buttons and links may betray your trust
Companies are using ‘Summarize with AI’ to manipulate enterprise chatbots
Why Your “Skill Scanner” Is Just False Security (and Maybe Malware)
Augustus: Open Source LLM Prompt Injection Tool
Posting AI-generated caricatures on social media is risky, infosec killjoys warn
Exposed Training Open the Door for Crypto-Mining in Fortune 500 Cloud Environments
Prompt Injection Via Road Signs
That “summarize with AI” button might be manipulating you
Notepad's new Markdown powers served with a side of remote code execution
RFK Jr's Nutrition Chatbot Recommends Best Foods to Insert Into Your Rectum