コンテンツへ移動
AI Risk Research独立監視とライブ実験
← AIリスクトラッカー
生成要約英語原文を表示PUB-0A4BFA5D5A

Hackers Deploy AI-Generated Exploits Targeting Critical Infrastructure

U.S. officials reported that threat actors used an AI-generated exploitation script to target devices in critical infrastructure, contributing to cyberattacks on U.S. water systems and a U.K. power plant. In some affected towns, the attacks disrupted water pressure levels and led to precautionary boil-water advisories.

重大度上昇56/100
証拠の確信度54%1 独立した情報源
証拠状態シグナル公開済み

何が起きたか

U.S. officials reported that threat actors used an AI-generated exploitation script to target devices in critical infrastructure, contributing to cyberattacks on U.S. water systems and a U.K. power plant. In some affected towns, the attacks disrupted water pressure levels and led to precautionary boil-water advisories.

Active deployment of AI-generated exploitation scripts against critical utilities caused physical disruptions including water pressure anomalies and plant shutdowns.

証拠の抜粋

  • Threat actors actively used an AI-generated exploitation script to target critical infrastructure devices.
  • Cyberattacks on U.S. water systems affected water pressure levels and prompted precautionary boil-water advisories in some towns.
  • A cyberattack on a U.K. power plant led to a four-day shutdown.

重大度の評価軸

影響60
規模50
制御喪失45
悪用可能性65
緊急性70
不可逆性40

情報源の引用

  1. AI is making critical infrastructure easier to attackAxios · 2026-08-25
情報源、訂正、プライバシーの方法を読む