{
  "id": 15650454,
  "title": "‘The user is hacking himself’: How AI exposes Israel’s political inner circles - analysis",
  "url": "https://urgent.news/2026/10/19/the-user-is-hacking-himself-how-ai-exposes-israels-political-inner",
  "topic": "ai",
  "section": "AI",
  "published": "2026-10-19T20:46:32.000Z",
  "source": {
    "name": "Jerusalem Post",
    "slug": "jerusalem-post",
    "url": "https://www.jpost.com/israel-news/politics-and-diplomacy/article-912208"
  },
  "original_language": "en",
  "account": "A program created by Prime Minister Benjamin Netanyahu's adviser Yonatan Urich scanned news sources for mentions of Netanyahu, his wife, Likud, and political rivals. When significant information was detected, the AI sent summaries to two WhatsApp groups. However, Urich left the code publicly accessible on GitHub, exposing private numbers, group identifiers, and access tokens. After the code was exposed, access was restricted.\n\nCalcalist reported that Urich connected his personal WhatsApp account to Ultramsg, a third-party automation service based in the UAE. This connection could have granted access to his private correspondence and media. On October 11, purported conversations from Urich's WhatsApp appeared online, though the authenticity of the material has not been confirmed.\n\nUrich's lawyers claim unauthorized access and allege some exchanges were altered or supplemented with fabricated content, threatening legal action. However, no public forensic investigation has determined whether the GitHub exposure or Ultramsg connection caused the leak. The perpetrators remain unidentified.\n\nCybersecurity experts note that sensitive political communications can become vulnerable without hacking, as users inadvertently expose themselves to risk by granting too much trust or power to AI tools. Prof. Yossi Oren from Ben-Gurion University explained that while there is no evidence of WhatsApp or phone hacking, the user is essentially \"hacking himself.\" He emphasized the dangers of granting excessive trust to AI.\n\nThe episode highlights that political advisers, despite lacking formal security clearance, can hold valuable intelligence. Cyber intelligence expert Matias Rott stressed that even without classified documents, accumulated exchanges could provide significant insights into senior officials and their advisers.",
  "summary": "The Urich leak raises concerns about AI-assisted programming, third-party messaging services, and the potential for personal accounts to expose sensitive political communications.",
  "key_points": [
    "AI program scans news for Netanyahu mentions and sends summaries to WhatsApp groups",
    "Yonatan Urich's code publicly exposed on GitHub, revealing private numbers and access tokens",
    "Sensitive political communications can be inadvertently exposed by granting excessive trust to AI"
  ],
  "editors_take": null,
  "illustration": null,
  "coverage": {
    "outlets": 1,
    "also_reported_by": []
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}