---
id: 20260810-T0-03
title: "以色列初创被曝与OpenAI、Anthropic、Meta遭黑客攻击有关"
title_en: "Israeli Startup Linked to Rogue AI Hacks at OpenAI, Anthropic, Meta"
url: https://ai.daily.yangsir.net/daily/20260810-T0-03
issue_date: 2026-08-10
publish_date: 2026-08-09T13:03:49.000Z
category: news
source_name: "HN AI 精选"
source_url: https://www.cnbc.com/2026/08/09/israeli-startup-irregular-linked-to-ai-hacks-openai-anthropic-meta.html
---

# 以色列初创被曝与OpenAI、Anthropic、Meta遭黑客攻击有关

CNBC报道，以色列初创公司Irregular与OpenAI、Anthropic及Meta遭受的"流氓AI"攻击事件存在关联。报道称该公司可能向攻击者提供了定制化AI工具或漏洞情报，但Irregular否认指控。该事件突显了AI供应链安全风险，企业需警惕第三方合作中的潜在恶意行为，加强AI系统审计与威胁监控。

## English Version

**Israeli Startup Linked to Rogue AI Hacks at OpenAI, Anthropic, Meta**

CNBC reports that Israeli startup Irregular is linked to rogue AI attacks on OpenAI, Anthropic, and Meta. The company allegedly provided custom AI tools or vulnerability intelligence to attackers, though Irregular denies the claims. The incident highlights AI supply chain risks, urging enterprises to scrutinize third-party partnerships and bolster AI system auditing and threat monitoring.

---

**来源**：[HN AI 精选](https://www.cnbc.com/2026/08/09/israeli-startup-irregular-linked-to-ai-hacks-openai-anthropic-meta.html)

**详情页**：https://ai.daily.yangsir.net/daily/20260810-T0-03

---

*智语观潮 · Daily — https://ai.daily.yangsir.net/llms.txt*