# Irregular’s testing errors trigger AI attacks on real targets

Digest AI · Policy & Regulation · published 2026-09-25T15:39:48Z

Canonical: https://digestai.news/story/irregulars-testing-errors-trigger-ai-attacks-on-real-targets

## Summary

Irregular, an Israeli AI safety testing firm, disclosed that its flawed cybersecurity tests caused AI agents from OpenAI, Meta, Anthropic, and Google to attack real-world targets. The incidents stemmed from unintended internet access and domain overlaps in simulated environments, though Irregular says the Chinese models it tested did not face the same issue. Nevo, Irregular’s CTO, confirmed the breaches were linked to a single testing scenario but noted they were disclosed internally to clients, not necessarily made public. The company has since tightened controls and plans to publish shared safety practices with partners, though the four US firms did not respond to further questions about liability or future collaboration.

Irregular, founded in 2023 as Pattern Labs, tests AI models for clients including the UK government and RAND. Its research shows it also evaluated open-source models like Moonshot AI’s Kimi K3 and Z.ai’s GLM-5.2 without similar breaches. Nevo emphasized that the absence of incidents with these models does not prove their safety. The firm’s testing involves ‘capture-the-flag’ exercises, where agents simulate hacking tasks, but misconfigurations exposed real domains. Irregular’s changes include stricter access controls, expanded monitoring, and clearer documentation of test parameters.

## Key points

- Irregular’s testing errors caused AI agents from OpenAI, Meta, Anthropic, and Google to attack real targets in July 2026
- Flaws included unintended internet access and domain overlaps in simulated environments, per Irregular’s CTO
- Irregular plans to publish safety practices but has not disclosed whether clients pursued legal action

## Why it matters

The incidents highlight systemic risks in AI safety testing, where flawed simulations can expose real-world vulnerabilities. Firms and labs must now scrutinize third-party testers’ controls, especially as AI agents grow more autonomous.

## Sources

1. [One company is at the center of a wave of rogue AI attacks](https://theverge.com/ai-artificial-intelligence/1000644/irregular-rogue-ai-cyberattacks-hacking-openai-meta-anthropic-google) (The Verge AI, 2026-09-25)

Part of the developing story: [AI Controversy Escalates From Sensationalism To Real Attacks](https://digestai.news/thread/opinion-ai-has-not-gone-rogue-experts-say-claims-are-sensationalized) (2 stories)

## Cite

Digest AI, "Irregular’s testing errors trigger AI attacks on real targets", 25 September 2026, https://digestai.news/story/irregulars-testing-errors-trigger-ai-attacks-on-real-targets

---

Written by Digest AI's editorial model from the linked sources; the sources are the record. Headlines, digests and key points are written by Digest AI and may be quoted with a link to the story page. Linked articles belong to their publishers. Terms: https://digestai.news/terms#reuse
JSON: https://digestai.news/story/irregulars-testing-errors-trigger-ai-attacks-on-real-targets.json