Prompt Injection

10 stories tagged Prompt Injection.

Man speaking to camera with AI circuit diagram and code visible behind him, blue text boxes displaying topic title on left…

Promptware: When AI Agents Become Attack Vectors

Prompt injection attacks on AI agents follow a structured kill chain — and existing legal frameworks have almost nothing to say about who's liable when it works.

Nvidia Skill Spector Scans AI Agent Skills for Threats

Rachel "Rach" Kovacs1 month ago

AI. Rachel "Rach" Kovacs1 month ago

Nvidia Skill Spector Scans AI Agent Skills for Threats

Nvidia's Skill Spector scans AI agent skills for hidden threats before installation. Here's what it catches, what it misses, and why the gap matters.

OpenAI Plans to Remake ChatGPT as an AI Super App

Rachel "Rach" Kovacs1 month ago

AI. Rachel "Rach" Kovacs1 month ago

OpenAI Plans to Remake ChatGPT as an AI Super App

OpenAI is reportedly overhauling ChatGPT into a full AI super app with coding agents, automation, and new security features. Here's what's verified and what's still reported.

Man in glasses gesturing before digital diagrams with "Don't Get Fired!" text overlay and glowing figures background from…

Five Ways AI Can End Your Career at Work

Marcus Chen-Ramirez2 months ago

AI. Marcus Chen-Ramirez2 months ago

Five Ways AI Can End Your Career at Work

Shadow AI, hallucination laundering, zombie agents—IBM's Martin Keen maps the AI workplace risks that have already cost people their jobs. Here's what they actually mean.

A man in a black shirt speaks against a neon-lit tech background with circuit board graphics, while text overlays read…

OWASP's Top 10 LLM Vulnerabilities: What Can Go Wrong

Marcus Chen-Ramirez4 months ago

AI. Marcus Chen-Ramirez4 months ago

OWASP's Top 10 LLM Vulnerabilities: What Can Go Wrong

OWASP's updated Top 10 for large language models reveals how easily AI systems can be manipulated, poisoned, or tricked into leaking sensitive data.

Man with surprised expression against textured background with "SONNET 4.6 IS HERE!" in red and white text

Anthropic's Sonnet 4.6: When A 'Workhorse' Model Gets Scary Good

Rachel "Rach" Kovacs5 months ago

AI. Rachel "Rach" Kovacs5 months ago

Anthropic's Sonnet 4.6: When A 'Workhorse' Model Gets Scary Good

Claude Sonnet 4.6 blurs the line between mid-tier and flagship AI. What happens when capabilities outpace our ability to measure them?

Pixelated portrait of a person's face in orange and blue tones with "SKILL ISSUE" text overlay on dark background

Agent Zero's New Skills Feature Makes AI Dangerously Easy

Zara Chen5 months ago

AI. Zara Chen5 months ago

Agent Zero's New Skills Feature Makes AI Dangerously Easy

Agent Zero's latest update lets anyone teach AI agents new tricks in minutes. The demo is impressive. The security warnings? Even more so.

Man in dark shirt against computer code background with "think series" branding and three white text boxes reading "Stop…

AI Agents Are Getting God Mode—And That's a Problem

Mike Sullivan5 months ago

AI. Mike Sullivan5 months ago

AI Agents Are Getting God Mode—And That's a Problem

IBM's Grant Miller explains how AI agents with elevated permissions create security nightmares—and what actually works to prevent privilege escalation.

Google's Model Armor: AI Security Through Callbacks

Samira Barnes5 months ago

AI. Samira Barnes5 months ago

Google's Model Armor: AI Security Through Callbacks

Google's Model Armor adds security checkpoints to AI agents through ADK callbacks, intercepting threats before they reach language models.

Bearded man with glasses and beanie gestures while speaking, with "DON'T RUN OPENCLAW (MOLTBOT) YET" text overlay and…

Moltbot Hit 82K GitHub Stars—Then Security Fell Apart

Yuki Okonkwo6 months ago

AI. Yuki Okonkwo6 months ago

Moltbot Hit 82K GitHub Stars—Then Security Fell Apart

The fastest-growing open source AI project reveals why agents that actually do things are both irresistible and architecturally dangerous.