PPhying中文
Curated security research, vulnerabilities, advisories and tools for practitioners.

Agentic Problems and the Rise of Zombie AIs

ethiack.com | research | #ai-security | #prompt-injection | #llm-security | #ai-agents | #chatgpt | #ethiack | #hackaicon
📄 Raw .md⬇️ Download .md🔗 Original ↗

Summary

Johann Rehberger shares HackAIcon research on hijacking AI agents: ChatGPT Operator was manipulated via prompt injection and a fake tool into leaking private data, despite multiple OpenAI defenses.
Published
Collected

Related coverage

research · nebusec.ai
Beyond the Model: Benchmarking Codex Security, Claude Security, and Nebu
Nebu benchmarks its security pipeline against Codex Security and Claude Security on 76 known bugs; with GPT-5.4 it matches the best Codex result in less time at lower cost.
research · wiz.io
Inside 90 days of attacks on AI infrastructure
This research provides real-world threat telemetry on how attackers target modern AI deployments and MCP interfaces.
blog · wiz.io
Building an AI Detection Engine That Understands Agent Intent
Wiz outlines a detection pipeline that analyzes AI agent input and output telemetry to trace shifts in intent, using the OpenAI-disclosed Hugging Face breach to show what legacy tools miss.
tool · github.com
LLM SecRange: Open-Source Penetration Testing Playground for Large Language Models & Vulnerable Agents
Offers security researchers and penetration testers an out-of-the-box, comprehensive environment to practice prompt injection, OWASP LLM Top 10 risks, and agent tool-abuse exploitation against real models.
blog · netspi.com
How App Integration Transactions Increase the Attack Surface of LLMs
Learn how OpenAI's AppsSDK, AgentKit, and "Buy It" turn LLMs into transactional agents—expanding security risks from rapid rollout, prompt injection, and access control gaps.
blog · blog.sentry.security
ChatML Role Injection (CRI)
Attackers can inject system roles into ChatML message arrays to override developer instructions, invoke hidden tools, and pad payloads for denial-of-wallet DoS; strict schema validation is key.
Back