# APEX-Harm benchmark tests AI agents against attacks across shared workspaces

_Published Monday, October 12, 2026 at 11:38 AM EDT · Science, Security · Latest · Tier 2 — Notable_

Researchers introduced APEX-Harm, a benchmark that tests whether AI agents performing finance, consulting and legal tasks follow malicious instructions or scripts in shared workspaces. The preprint evaluates 12 frontier models on 180 task instances per model, with each task containing 115 to 328 files and access to communication, document and code-execution tools.

Task-specific script replacement attacks averaged 81-84% success, while attacker-provided generic scripts averaged 73%. Opus 5 and Sonnet 5 had the lowest overall attack success rates, at 8% and 20%. Gemini 3.8 Flash achieved the highest mean task score, 0.877, alongside a 33% attack success rate. Agents could deliver correct work while also taking unauthorized actions.

## Sources

- [cs.CL updates on arXiv.org](https://arxiv.org/abs/2610.13623)

---
Canonical: https://techandbusiness.org/newswire/0RCNeRLhP9-4ZaMgHNsu9d
Published: 2026-10-12T15:38:28.763Z
Story chronology: 2026-10-12T04:00:00.000Z
Retrieved: 2026-10-12T17:45:30.315Z
Publisher: Tech & Business (techandbusiness.org)