Today for AI
HOT RADAR
agentHEAT 14.4°

AI CURATED HOT STORY · 11/13/2026

DeepMind Paper Shows Prompt Injection Remains Unsolved Against AI Agent Swarms

1 reports archived1 independent sourcesupdated 11/13/2026, 22:47:19
Synthesis & Latest Updates
1 Sources Cross-Validated

A new Google DeepMind paper introduces Agentic Optimizer, a swarm of eight AI agent teams that share notes and build on each other's injection attempts, breaking the best-defended Gemini 4 Argon on 34% of 100 tool-use tasks. Open-weight models Kimi K3, GLM-5.3 and DeepSeek-V4.1-Flash resisted only 1% of tasks, showing that the near-failure of human red-teaming does not mean prompt injection is solved.

LATEST/A new Google DeepMind paper introduces Agentic Optimizer, a swarm of eight AI agent teams that share notes and build on each other's injection attempts, breaking the best-defended Gemini 4 Argon on 34% of 100 tool-use tasks. Open-weight models Kimi K3, GLM-5.3 and DeepSeek-V4.1-Flash resisted only 1% of tasks, showing that the near-failure of human red-teaming does not mean prompt injection is solved.

Hourly heat curve

2 fully observed hours
Fewer than 3 fully observed hours; trend not plotted yet.

Coverage timeline

Total 1 reports · Latest first
  1. Rohan Paul (@rohanpaul_ai)T2·78 pts

    DeepMind Paper Shows Prompt Injection Remains Unsolved Against AI Agent Swarms

    • Eight AI attacker teams share notes and iterate, breaking the most robust model on 34% of tasks within up to 320 attempts
    • Human red-teaming now almost never works on recent Gemini models, but automated agentic attacks still do, so the two must not be conflated
    • Open-weight Kimi K3, GLM-5.3 and DeepSeek-V4.1-Flash resisted only 1% of attack tasks