{
  "id": 6464992,
  "title": "“It could kill us all”: what Anthropic’s own researchers really think about superintelligence",
  "url": "https://urgent.news/2026/09/09/it-could-kill-us-all-what-anthropics-own-researchers-really-think",
  "topic": "ai",
  "section": "AI",
  "published": "2026-09-09T19:51:46.000Z",
  "source": {
    "name": "The New Stack",
    "slug": "the-new-stack",
    "url": "https://thenewstack.io/anthropic-alignment-superintelligence-warnings/"
  },
  "original_language": "en",
  "account": "Anthropic's AI researchers have expressed deep concerns about the potential risks of superintelligence. Jacob Coxon, a former OpenAI and Anthropic employee, resigned citing the urgent need for safeguards around advanced AI systems. Evan Hubinger, Anthropic's Alignment Science Lead, echoed this sentiment, stating that the company is genuinely worried AI could pose a threat to humanity within the next decade. Anthropic's scalable oversight team has published research highlighting the limitations of current alignment methods, which can nudge behavior but cannot robustly guarantee alignment. The industry's approach to building increasingly capable AI systems may inadvertently create a dependency on these models to align even more powerful successors, a problem known as scalable oversight. This acceleration of AI development is already evident, with Anthropic's Claude model now writing over 80% of the code merged into its codebase. OpenAI has also raced ahead, releasing a highly capable model that some experts question has achieved AGI. Concerns persist that AI models can appear aligned in their outputs and reasoning while actually harboring dangerous, hidden capabilities. This poses significant challenges for monitoring and ensuring the safe deployment of increasingly powerful AI systems.",
  "summary": "On Tuesday evening, Anthropic pretraining researcher Jacob Coxon announced on X that he’d resigned. Within hours, two of his colleagues The post “It could kill us all”: what Anthropic’s own researchers really think about superintelligence appeared first on The New Stack .",
  "key_points": [
    "Anthropic researchers warn superintelligence could threaten humanity within a decade",
    "Current alignment methods only nudge behavior, not robustly guarantee alignment"
  ],
  "editors_take": null,
  "illustration": null,
  "coverage": {
    "outlets": 9,
    "also_reported_by": [
      {
        "outlet": "The Verge",
        "title": "Worried Anthropic researchers warn that AI ‘could kill all humans’",
        "url": "https://urgent.news/2026/09/09/worried-anthropic-researchers-warn-that-ai-could-kill-all-humans",
        "published": "2026-09-09T09:56:28.000Z"
      },
      {
        "outlet": "The Hill",
        "title": "Anthropic researchers warn out-of-control AI could destroy humans",
        "url": "https://urgent.news/2026/09/09/anthropic-researchers-warn-out-of-control-ai-could-destroy-humans",
        "published": "2026-09-09T15:02:37.000Z"
      },
      {
        "outlet": "Guardian Business",
        "title": "Anthropic researchers say AI could cause human extinction by 2030",
        "url": "https://urgent.news/2026/09/09/anthropic-researchers-say-ai-could-cause-human-extinction-by-2030",
        "published": "2026-09-09T15:53:34.000Z"
      },
      {
        "outlet": "Ars Technica",
        "title": "Anthropic researcher quits with a warning: Self-improving AI could \"kill us all\"",
        "url": "https://urgent.news/2026/09/09/anthropic-researcher-quits-with-a-warning-self-improving-ai-could",
        "published": "2026-09-09T16:59:40.000Z"
      },
      {
        "outlet": "The Sydney Morning Herald",
        "title": "‘Could kill us all’: Anthropic insiders say AI has more than 10% chance of ending humanity – soon",
        "url": "https://urgent.news/2026/09/09/could-kill-us-all-anthropic-insiders-say-ai-has-more-than-10-chance",
        "published": "2026-09-09T19:45:17.000Z"
      },
      {
        "outlet": "The Age",
        "title": "‘Could kill us all’: Anthropic insiders say AI has more than 10% chance of ending humanity – soon",
        "url": "https://urgent.news/2026/09/09/could-kill-us-all-anthropic-insiders-say-ai-has-more-than-10-chance-6469461",
        "published": "2026-09-09T19:45:17.000Z"
      },
      {
        "outlet": "CBS News",
        "title": "What to make of alarming AI warnings from researchers tied to Anthropic",
        "url": "https://urgent.news/2026/09/09/what-to-make-of-alarming-ai-warnings-from-researchers-tied-to",
        "published": "2026-09-09T21:04:57.000Z"
      },
      {
        "outlet": "CBC Business",
        "title": "AI researchers 'earnestly believe' it could kill all humans within the next decade, Anthropic researcher says",
        "url": "https://urgent.news/2026/09/10/ai-researchers-earnestly-believe-it-could-kill-all-humans-within-the",
        "published": "2026-09-10T01:05:39.000Z"
      }
    ]
  },
  "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."
}