# Study finds privilege-confusion attacks feasible on six of seven AI accelerators

_Published Tuesday, September 22, 2026 at 7:08 PM EDT · Security, AI, Science · Latest · Tier 2 — Notable_

Researchers reported that malicious applications could trick six of seven tested edge AI accelerators into performing privileged operations, potentially affecting more than 128 system-on-chips and 100 million devices. The preprint examined hardware from Google, Nvidia, Hailo, Texas Instruments, NXP, AWS and Rockchip using DeputyHunt, a framework that combines static and dynamic analysis with a large language model.

The vendors acknowledged the findings, which received CVE-2025-66425. A proposed on-demand validation defense produced roughly 15% average runtime overhead in a Gem5-salam simulation, compared with roughly 56% for traditional defenses; the evaluation was simulated rather than a production deployment.

## Sources

- [cs.CR updates on arXiv.org](https://arxiv.org/abs/2605.17707)

---
Canonical: https://techandbusiness.org/newswire/AZjzM-XR7IkVQrrYH0PMNS
Published: 2026-09-22T23:08:54.096Z
Story chronology: 2026-09-22T04:00:00.000Z
Retrieved: 2026-09-23T00:38:11.097Z
Publisher: Tech & Business (techandbusiness.org)