When the Governor Becomes the Disturbance: Control-Generated Disturbance and Cost-Aware Backoff in Governed Tool-Using Agents
We study this possibility in a controlled file-recovery environment where increases in regulatory intensity trigger experimentally imposed tool failures.
ProofPaper ↗
Key points
- Supervisory governors can interfere with the tool-using agents they regulate.
- A cost-blind governor can turn these failures into persistent blocking that prevents task completion.
- On a hand-coded stochastic-policy agent, the failure pattern appears under both result replacement and execution of corrupted tool arguments.
- For the persistent policy, adaptive backoff improves completion relative to a fixed weak governor with approximately matched intervention frequency.
Sources (1)
- [1]When the Governor Becomes the Disturbance: Control-Generated Disturbance and Cost-Aware Backoff in Governed Tool-Using AgentsarXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 6, 07:39 PM
We study this possibility in a controlled file-recovery environment where increases in regulatory intensity trigger experimentally imposed tool failures.
Supervisory governors can interfere with the tool-using agents they regulate.
Extractive summary: sentences quoted from the sources.
Before this
- Oct 6, 2026EmbeddingGemma 2: an open, lightweight multimodal embedding model
- Sep 30, 2026Gemini 4 Argon: our next era of frontier intelligence
- Sep 30, 2026[AINews] OpenAI DevDay 2026: Dots, 6.1 Sol, Ultrafast, Decisions API, Agents API, Spaces, Marketplace, and 1.2 Billion ChatGPT WAU
- Sep 2, 2026Proactive cyber defense for governments and enterprises
- Jun 30, 2026Start building with Nano Banana 2 Lite and Gemini Omni Flash
- Jun 9, 2026Introducing Gemma 4 12B: a unified, encoder-free multimodal model