{
  "id": 3793543,
  "title": "This Is How Anthropic Thinks AI Agents Should Navigate the Physical World",
  "url": "https://urgent.news/2026/08/27/this-is-how-anthropic-thinks-ai-agents-should-navigate-the-physical-3793543",
  "topic": "ai",
  "section": "AI",
  "published": "2026-08-27T18:06:52.000Z",
  "source": {
    "name": "Wired",
    "slug": "wired",
    "url": "https://www.wired.com/story/anthropic-standard-ai-agents-coming-to-the-physical-world/"
  },
  "original_language": "en",
  "account": "Anthropic, an AI company, has unveiled a new framework named Model Hardware Standard to guide AI agents in interacting with physical systems such as microscopes, industrial machinery, quantum computing hardware, manufacturing equipment, and robotic arms. This framework sets guidelines on how AI agents should and should not engage with various hardware components, reflecting the belief that AI can revolutionize scientific research and industries like manufacturing if it can operate safely in the physical world. The company plans to collaborate with trusted partners to ensure maximum safety before making the standard widely accessible. Although misuse concerns, such as the potential for developing biological weapons, exist, Anthropic believes that built-in safety measures within AI models will prevent malicious exploitation of the new standard.\n\nQuantum physicist Alek Kemeny, who co-led the development of the framework, emphasized the goal of accelerating scientific progress by bridging the gap between literature review, data analysis, and experimental work. Claude and similar chatbots are already adept at processing vast amounts of information from scientific papers and experimental results, uncovering new insights. AI agents, the next evolution of chatbots, are designed to perform actions, ranging from answering emails to interacting with physical hardware. Several startups, including Periodic Labs, LILA Sciences, Edison Scientific, and Discovery Loop, are pursuing AI-driven scientific discovery.\n\nA key challenge lies in configuring scientific equipment and hardware, which typically demands extensive expertise. Anthropic's standard aims to automate much of this complex engineering by enabling AI to configure machines and facilitate communication between different hardware components. Jonah Cool, an experimental biologist involved in creating the standard at Anthropic, highlighted that configuring scientific equipment and having it interact seamlessly with other hardware traditionally requires specialized skills. With the new standard, Claude can observe robotic systems on a factory line and optimize their behavior.\n\nWhile AI agents have recently drawn attention due to security risks, such as unauthorized system access and deceiving human users, Anthropic stresses that the standard will help scientists and engineers specify how AI models should avoid misusing different hardware, mitigating potential hazards. The company previously introduced the Model Context Protocol, which outlines rules for AI models interacting with various software programs.",
  "summary": "The potential for AI to automate scientific research and manufacturing must be balanced with new risks, Anthropic says.",
  "key_points": [],
  "editors_take": null,
  "illustration": null,
  "coverage": {
    "outlets": 2,
    "also_reported_by": [
      {
        "outlet": "Wired Business",
        "title": "This Is How Anthropic Thinks AI Agents Should Navigate the Physical World",
        "url": "https://urgent.news/2026/08/27/this-is-how-anthropic-thinks-ai-agents-should-navigate-the-physical",
        "published": "2026-08-27T18:06:52.000Z"
      }
    ]
  },
  "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."
}