{
  "id": 7288271,
  "title": "Chinas Minister für Staatssicherheit warnt vor KI-Risiken",
  "url": "https://urgent.news/2026/09/14/chinas-minister-fur-staatssicherheit-warnt-vor-ki-risiken",
  "topic": "business",
  "section": "Business",
  "published": "2026-09-14T09:34:30.000Z",
  "source": {
    "name": "NZZ Wirtschaft",
    "slug": "nzz-wirtschaft",
    "url": "https://www.nzz.ch/wirtschaft/chinas-minister-fuer-staatssicherheit-warnt-vor-ki-risiken-ld.10023667"
  },
  "original_language": "de",
  "account": "China's top security official has warned of the risks associated with artificial intelligence (AI), expressing concerns that rapid technological advancements could threaten the country's political stability and critical infrastructure. In a recent article published on a Chinese online censorship platform, Chen Yixin, the Minister for State Security, highlighted the potential dangers posed by adversaries who could exploit AI to mount \"political, institutional, and ideological threats\" to China. Chen specifically cited leading American AI models such as Claude Mythos by Anthropic and GPT-5.5-Cyber by OpenAI as examples of the escalating capabilities of AI systems. Beijing has expressed particular alarm over the possibility of \"hostile forces\" engaging in \"propaganda wars\" and \"cognitive warfare,\" employing deepfakes, AI-generated text and images, and social media bots to create political rumors, disseminate harmful information, and deepen societal divisions. The article reflects one of the most explicit statements so far on China's assessment of AI-related risks, appearing almost simultaneously with an appeal from leading American AI companies to slow down AI development. However, no link between the timing of the appeal and the planned meeting between China's President Xi Jinping and US President Donald Trump in Washington on September 24th has been confirmed, as the interests of China and the US corporations are fundamentally divergent. Chen did not respond to the calls from Anthropic and OpenAI. As the planned meeting between Chinese leader Xi Jinping and US President Donald Trump approaches, discussions about AI regulation are heating up, with both countries' delegations preparing for the summit. While China has not officially confirmed the date of the meeting, the delegations have placed AI regulation on the agenda. Approximately 90% of the leading AI models are developed in China and the US, prompting calls for cooperation between the two countries in regulating the technology. Experts have long advocated for collaboration between the two nations in AI regulation, as the competition for AI dominance has become central to the geopolitical power struggle between China and the United States. Chinese AI labs like Alibaba, Deepseek, Moonshot, and Zhipu AI have unveiled major language models in August that can match or surpass leading American models in many areas, despite the US restricting the export of advanced chips and manufacturing machines to China. This move aims to slow down China's technological catch-up. However, the US accuses Chinese companies of unlawfully leveraging American technology in developing their models, with the FBI, NSA, and CISA alleging in a joint statement that several Chinese firms, including Alibaba, Deepseek, and Moonshot, have accelerated their AI models' training using American models, likely with Beijing's knowledge. The US Ministry of Foreign Affairs dismissed the accusations as unfounded, stating that China's AI advancements are the result of its high degree of self-reliance and independent scientific and technological development, as well as its openness to collaboration. Unlike most American providers, Chinese companies often release their models freely and at relatively low costs, leading to rapid dissemination within China and globally. Experts attribute China's rapid progress in AI to this open approach, as many models build upon each other. In July, President Xi emphasized during a keynote speech at the World AI Conference in Shanghai that \"AI development should not be the sole endeavor of a single country.\" Beijing has offered other nations an invitation to collaborate in AI development and regulation, with its conditions in mind. During the recent Brics summit, Xi advocated for the China-initiated organization aimed at fostering cooperation in AI development and regulation (Waico). While the Chinese government pushes for a global spread of \"made in China\" AI, American tech giants are seeking to protect their leadership role and business model. This is evident in the recent opinions published, with Anthropic co-founder Dario Amodei arguing that AI capabilities will advance faster than safeguards can keep up, and advocating for stronger coordination between companies, the US, and other democracies. The call for slowing down the AI race faces resistance in Washington, where President Trump fears China may gain an advantage in the crucial technology.",
  "summary": "In China und den USA wächst die Sorge, dass künstliche Intelligenz ausser Kontrolle gerät. Beim geplanten Treffen von Xi Jinping und Donald Trump dürfte das Thema in den Fokus rücken.",
  "key_points": [],
  "editors_take": null,
  "illustration": null,
  "coverage": {
    "outlets": 2,
    "also_reported_by": [
      {
        "outlet": "Handelsblatt",
        "title": "KI: Chinas Geheimdienstchef warnt vor KI-Risiken",
        "url": "https://urgent.news/2026/09/14/ki-chinas-geheimdienstchef-warnt-vor-ki-risiken",
        "published": "2026-09-14T04:55:06.000Z"
      }
    ]
  },
  "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."
}