# Owner-Harm: A Missing Threat Model for AI Agent Safety

arXiv: https://arxiv.org/abs/2604.18658 · HTML: https://mlsanity.com/paper/2604.18658/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in Agent threats, guardrails and governance, one of 16 directions in LLM agents and tool use.
- Builds on 14 of its 31 references that are in the map.
- First posted on arXiv 20 Apr 2026.

## Placed in

- LLM agents and tool use / [Agent threats, guardrails and governance](https://mlsanity.com/agents/agent-threats-guardrails-and-governance/)

## Builds on (most cited first, 14 of 14)

- [Not what you've signed up for: Compromising Real-World LLM-Integrated Applications with Indirect Prompt Injection](https://mlsanity.com/paper/2302.12173/) (2023)
- [Llama Guard: LLM-based Input-Output Safeguard for Human-AI Conversations](https://mlsanity.com/paper/2312.06674/) (2023)
- [Identifying the Risks of LM Agents with an LM-Emulated Sandbox](https://mlsanity.com/paper/2309.15817/) (2023)
- [AgentHarm: A Benchmark for Measuring Harmfulness of LLM Agents](https://mlsanity.com/paper/2410.09024/) (2024)
- [NeMo Guardrails: A Toolkit for Controllable and Safe LLM Applications with Programmable Rails](https://mlsanity.com/paper/2310.10501/) (2023)
- [Benchmarking and Defending Against Indirect Prompt Injection Attacks on Large Language Models](https://mlsanity.com/paper/2312.14197/) (2023)
- [ShieldGemma: Generative AI Content Moderation Based on Gemma](https://mlsanity.com/paper/2407.21772/) (2024)
- [AgentSpec: Customizable Runtime Enforcement for Safe and Reliable LLM Agents](https://mlsanity.com/paper/2503.18666/) (2025)
- [ShieldAgent: Shielding Agents via Verifiable Safety Policy Reasoning](https://mlsanity.com/paper/2503.22738/) (2025)
- [AGrail: A Lifelong Agent Guardrail with Effective and Adaptive Safety Detection](https://mlsanity.com/paper/2502.11448/) (2025)
- [Towards Verifiably Safe Tool Use for LLM Agents](https://mlsanity.com/paper/2601.08012/) (2026)
- [Formal Policy Enforcement for Real-World Agentic Systems](https://mlsanity.com/paper/2602.16708/) (2026)
- [Provably safe systems: the only path to controllable AGI](https://mlsanity.com/paper/2309.01933/) (2023)
- [Solver-Aided Verification of Policy Compliance in Tool-Augmented LLM Agents](https://mlsanity.com/paper/2603.20449/) (2026)

## Built on by (most cited first, 1 of 1)

- ["You're Right, Let Me Fix It": How LLM Agents Damage Correct Work When Falsely Accused](https://mlsanity.com/paper/2609.32616/) (2026)

## Abstract (arXiv)

Existing AI agent safety benchmarks focus on generic criminal harm (cybercrime, harassment, weapon synthesis), leaving a systematic blind spot for a distinct and commercially consequential threat category: agents harming their own deployers. Real-world incidents illustrate the gap: Slack AI credential exfiltration (Aug 2024), Microsoft 365 Copilot calendar-injection leaks (Jan 2024), and a Meta agent unauthorized forum post exposing operational data (Mar 2026). We propose Owner-Harm, a formal threat model with eight categories of agent behavior damaging the deployer. We quantify the defense gap on two benchmarks: a compositional safety system achieves 100% TPR / 0% FPR on AgentHarm (generic criminal harm) yet only 14.8% (4/27; 95% CI: 5.9%-32.5%) on AgentDojo injection tasks (prompt-injection-mediated owner harm). A controlled generic-LLM baseline shows the gap is not inherent to owner-harm (62.7% vs. 59.3%, delta 3.4 pp) but arises from environment-bound symbolic rules that fail to generalize across tool vocabularies. On a post-hoc 300-scenario owner-harm benchmark, the gate alone achieves 75.3% TPR / 3.3% FPR; adding a deterministic post-audit verifier raises overall TPR to 85.3% (+10.0 pp) and Hijacking detection from 43.3% to 93.3%, demonstrating strong layer complementarity. We introduce the Symbolic-Semantic Defense Generalization (SSDG) framework relating information coverage to detection rate. Two SSDG experiments partially validate it: context deprivation amplifies the detection gap 3.4x (R = 3.60 vs. R = 1.06); context injection reveals structured goal-action alignment, not text concatenation, is required for effective owner-harm detection.