When Agents Act Unwatched: The Reduced-Supervision Paradox in Agentic AI

arXiv:2609.29547 · cs.CY, cs.AI · Submitted 2026-08-28 · Read on arXiv

cs.CY, cs.AI

Submitted: 2026-08-28

Updated: 2026-08-28

Terminology

• Trace — 47× in this paper · explained in 2 episodes
• Accountability — 43× in this paper · explained in 2 episodes
• Actor — 16× in this paper · explained in 3 episodes
• Verification — 14× in this paper · explained in 2 episodes
• Benchmark — 12× in this paper · explained in 2 episodes
• Agentic AI — 11× in this paper · explained in 1 episode
• AI Agents — 5× in this paper · explained in 2 episodes
• Agentic System — 5× in this paper · explained in 2 episodes
• Benchmarking — 4× in this paper · explained in 3 episodes
• Commentary — 3× in this paper · explained in 2 episodes
• Scaffolding — 3× in this paper · explained in 3 episodes
• Foundation Model — 2× in this paper · explained in 3 episodes
• Generative AI — 2× in this paper · explained in 2 episodes
• Human-in-the-Loop — 2× in this paper · explained in 3 episodes
• Multi-Agent System — 2× in this paper · explained in 5 episodes
• Abstraction — 1× in this paper · explained in 1 episode
• Agent Harness — 1× in this paper · explained in 1 episode
• Agentic AI systems — 1× in this paper · explained in 1 episode
• Agentic Workflows — 1× in this paper · explained in 1 episode
• Alignment — 1× in this paper · explained in 1 episode
• Failure Modes — 1× in this paper · explained in 3 episodes
• Provenance — 1× in this paper · explained in 3 episodes
• Robustness — 1× in this paper · explained in 5 episodes

Sources

• SARC: A Governance-by-Architecture Framework for Agentic AI Systems
• WorkArena++: Towards Compositional Planning and Reasoning-based Common Knowledge Work Tasks
• Building Effective AI Coding Agents for the Terminal: Scaffolding, Harness, Context Engineering, and Lessons Learned
• SABER: Small Actions, Big Errors -- Safeguarding Mutating Steps in LLM Agents
• AgentDojo: A Dynamic Environment to Evaluate Prompt Injection Attacks and Defenses for LLM Agents
• WorkArena: How Capable Are Web Agents at Solving Common Knowledge Work Tasks?
• From Governance Norms to Enforceable Controls: A Layered Translation Method for Runtime Guardrails in Agentic AI
• ToolSandbox: A Stateful, Conversational, Interactive Evaluation Benchmark for LLM Tool Use Capabilities
• Natural-Language Agent Harnesses
• Identifying the Risks of LM Agents with an LM-Emulated Sandbox
• Toolformer: Language Models Can Teach Themselves to Use Tools
• Authenticated Delegation and Authorized AI Agents
• Voyager: An Open-Ended Embodied Agent with Large Language Models
• OpenHands: An Open Platform for AI Software Developers as Generalist Agents
• AutoGen: Enabling Next-Gen LLM Applications via Multi-Agent Conversation
• OSWorld: Benchmarking Multimodal Agents for Open-Ended Tasks in Real Computer Environments
• TheAgentCompany: Benchmarking LLM Agents on Consequential Real World Tasks
• SWE-agent: Agent-Computer Interfaces Enable Automated Software Engineering
• ReAct: Synergizing Reasoning and Acting in Language Models
• $\tau$-bench: A Benchmark for Tool-Agent-User Interaction in Real-World Domains

Related papers

• Reasoning Enhances Robustness to Prompt Injection in LLM-Based Consensus
• Generative AI Purpose-built for Social and Mental Health: A Real-World Pilot
• PersonaMem-v3: Toward Omni-Platform Personal Intelligence for Holistic User Understanding, Recommendation, and Agentic Tasks
• What is an intelligent system?
• AI University: An LLM-Powered Learning Assistant for Engineering---A Finite Element Method Case Study
• Generative AI Use in Entrepreneurship: An Integrative Review and an Empowerment-Entrapment Framework