# Gebru and Bender: error-prone military AI, not superintelligence, is the near-term risk

> In a Guardian column, Timnit Gebru and Emily M. Bender cite a CNN report that US military reliance on brittle chatbots nearly brought the US into war with China.

- **Topic**: Society
- **Published**: 2026-10-02T06:17:05.314Z
- **Canonical URL**: https://highsignal.sh/stories/gebru-and-bender-error-prone-military-ai-not-superintelligence-is-the-near-b6fcdb34

## Why It Matters

The commentary argues the past month's dominant AI story — Anthropic engineer Jacob Coxon's resignation over a claimed race to self-improving superintelligence at OpenAI and Anthropic — drew far more attention than a reported near-miss between the US and China. Gebru and Bender's point: present-day, brittle systems already carry international-security consequences. Their column lands as labs push frontier model news, a reminder that reliability and deployment failures stay on the agenda alongside capability claims. It is opinion, and the CNN incident is described secondhand.

## Primary Sources & Citations

- [Forget ‘superintelligence’: error-prone AI nearly sparked world war three this month | Timnit Gebru and Emily M Bender](https://www.theguardian.com/commentisfree/2026/oct/01/forget-superintelligence-error-prone-ai-nearly-sparked-world-war-iii-this-month) — *Guardian AI* (Reporting)

---

[← Back to Headlines](https://highsignal.sh/)