{
  "id": 7270829,
  "title": "Sam Altman reveals ‘two ways AI progress could go very badly’ days after backing ‘slowdown’ call",
  "url": "https://urgent.news/2026/09/14/sam-altman-reveals-two-ways-ai-progress-could-go-very-badly-days",
  "topic": "ai",
  "section": "AI",
  "published": "2026-09-14T07:47:08.000Z",
  "source": {
    "name": "Hindustan Times - World News",
    "slug": "hindustan-times-world-news",
    "url": "https://www.hindustantimes.com/world-news/open-ai-sam-altman-reveals-2-ways-ai-progress-could-go-very-badly-days-after-backing-slowdown-call-101789367179972.html"
  },
  "original_language": "en",
  "account": "Sam Altman, CEO of OpenAI, has warned that artificial intelligence (AI) progress could go \"very badly\" in two major ways if the development of advanced AI models is not properly managed. The entrepreneur, who recently called for a \"pace\" in AI development, emphasized that this does not mean stopping AI progress, but rather slowing the pace enough to ensure that safety and monitoring systems can keep up.\n\nAltman highlighted two significant risks associated with AI: the potential for humans to lose control of increasingly powerful AI systems, and the risk of too much power ending up in the hands of one person, company, or country. In a statement on X, Altman made clear that the goal of \"pacing\" AI development is to ensure that the capabilities of AI do not move faster than humans' ability to understand and control the associated risks.\n\nTo address these risks, Altman proposed several measures. He suggested that companies developing the most advanced AI models should prepare explicit safety cases before certain \"frontier reinforcement-learning runs,\" which could significantly increase a model's capabilities. OpenAI has already incorporated safety work into its model release process and supports the idea of independent auditors assessing AI systems. Additionally, Altman advocated for the development of shared standards for areas such as misalignment, monitoring, and safety.\n\nThe entrepreneur also expressed support for a federal framework establishing consistent safety requirements for frontier AI companies in the United States, arguing that companies should not wait for legislation to improve safety practices. Altman acknowledged the need for international coordination, as AI development is not limited to a single country.\n\nTwo other prominent figures in the AI industry, Anthropic CEO Dario Amodei and billionaire Elon Musk, had previously called for a similar slowdown in AI development. In a detailed essay, Amodei proposed that AI companies give safety systems more time to catch up with rapidly improving capabilities by using independent evaluators with employee-level access, industry-wide coordination, and international cooperation. Musk also backed the calls for a slowdown, stating that Dario was \"right\" in his concerns.",
  "summary": "Sam Altman said OpenAI prepares explicit safety cases before some frontier reinforcement-learning runs that are expected to increase a model’s capabilities.",
  "key_points": [],
  "editors_take": null,
  "illustration": null,
  "coverage": {
    "outlets": 3,
    "also_reported_by": [
      {
        "outlet": "Times of India",
        "title": "Billionaire David Sacks to Sam Altman and Dario Amodei's call to slowdown AI development",
        "url": "https://urgent.news/2026/09/14/billionaire-david-sacks-to-sam-altman-and-dario-amodeis-call-to",
        "published": "2026-09-14T02:13:48.000Z"
      },
      {
        "outlet": "Free Malaysia Today",
        "title": "OpenAI’s Altman voices support for ‘pacing’ AI development",
        "url": "https://urgent.news/2026/09/14/openais-altman-voices-support-for-pacing-ai-development",
        "published": "2026-09-14T05:33:46.000Z"
      }
    ]
  },
  "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."
}