AI ethics evidence index Search
Source-linked daily record
Evidence record 5615 · automatically gathered

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

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 th

Record details

Published: 20 April 2026
Source: arXiv
Category: Research
Topics: Military & security · Agents & autonomy
Retrieved: 14 July 2026

source-onlyevidence status

These records share source-supplied organisations, an exact publisher byline, automatic topics or regions. The reason is shown on every link; related does not mean supporting, agreeing with or verifying this record.

Also mentions Meta · Also mentions Microsoft

Amazon, Meta and Microsoft face skeptical investors this week after Google report sparked sell-off

CNBC Technology · 28 July 2026

Also mentions Meta · Also mentions Microsoft

Big Tech Turmoil Clouds the A.I. Earnings Picture

The New York Times · 29 July 2026

Also mentions Meta · Also mentions Microsoft

Big Tech AI spending spree tops $1tn

Financial Times Technology (headlines) · 31 July 2026

Also mentions Meta · Also mentions Microsoft

KI-Ausbau: Techriesen haben mehr als eine Billion Dollar investiert

Heise Online (DE) · 31 July 2026

Also mentions Meta · Also mentions Microsoft

Las grandes tecnológicas ya se han gastado un billón de dólares en IA. El problema es que en la caja sólo les queda calderilla

Xataka (ES) · 31 July 2026

Also mentions Meta · Also mentions Microsoft

Tech companies rack up debt, risks to fund AI ambitions

The Hill Technology · 5 August 2026

How to cite this record

ethics.ai (20 April 2026), “Owner-Harm: A Missing Threat Model for AI Agent Safety,” evidence record 5615, https://ethics.ai/record/5615 (originally published by arXiv).

JSON

Use and limitations

This page is a stable index and citation surface for a source record. ethics.ai did not author the underlying report and has not independently verified every claim. Automatic topics may be imperfect. For consequential use, quote and cite the original publisher.