Anthropic’s new hardware standard lets AI agents control the physical world

Anthropic 推出全新硬件标准，助力 AI 智能体掌控物理世界

For all the interest in and uptake of agentic AI systems over the past year or so, the world of automated AI has thus far been primarily limited to text, images, code, and other data and actions that take place inside a computer. 尽管过去一年左右的时间里，人们对智能体 AI（Agentic AI）系统的兴趣和应用激增，但自动化 AI 的世界目前主要仍局限于文本、图像、代码以及其他在计算机内部进行的数据处理和操作。

Anthropic is now aiming to change that somewhat with what it’s calling the Model Hardware Standard (MHS), a set of standardized drivers designed to let AI agents easily interface with and control arbitrary devices. Anthropic 现在正试图通过其所谓的“模型硬件标准”（Model Hardware Standard，简称 MHS）来改变这一现状。这是一套标准化的驱动程序，旨在让 AI 智能体能够轻松地与各种设备进行交互并实现控制。

For now, the “research preview” of the MHS effort is being sold mainly as a way to help scientists streamline the arduous process of creating the custom software integrations that are often needed to get disparate components of an experiment working in concert. 目前，MHS 项目的“研究预览版”主要旨在帮助科学家简化繁琐的定制软件集成过程——通常为了让实验中不同的组件协同工作，往往需要进行此类集成。

MHS can provide a common interface and common format for data sharing between these devices, Anthropic says, allowing them to talk to each other across a network “without needing a bespoke ‘translator’ program in between.” The standardized system could reduce weeks or months of exacting experimental setup down to “hours or minutes,” Anthropic writes. Anthropic 表示，MHS 可以为这些设备之间的数据共享提供通用的接口和格式，使它们能够在网络上相互通信，“而无需在中间编写专门的‘翻译’程序”。Anthropic 写道，这种标准化系统可以将原本需要数周或数月的严苛实验设置时间缩短至“几小时或几分钟”。

In a video posted alongside the announcement, Anthropic Technical Staffer Alek Kemeny says the MHS effort was inspired by observing neuroscientist Arco Bast work through an experiment on memory formation in the brain at the HHMI Janelia Research Campus in Ashburn, Virginia. Kemeny said Bast had worked out an interface to get the rotating laser beams, microscopes, cameras, and myriad other components of the experiment to coordinate through a common interface. “This idea could be used to have AI run any science experiment in the world,” Kemeny recalls thinking at the time. 在随公告发布的一段视频中，Anthropic 技术人员 Alek Kemeny 表示，MHS 项目的灵感源于观察神经科学家 Arco Bast 在弗吉尼亚州阿什本的 HHMI Janelia 研究园区进行的一项关于大脑记忆形成的研究。Kemeny 说，Bast 当时设计了一个接口，让实验中的旋转激光束、显微镜、摄像头以及无数其他组件能够通过一个通用接口进行协调。“当时我就想，这个想法可以用来让 AI 运行世界上任何科学实验，”Kemeny 回忆道。

There’s nothing about a common machine interface language that requires the use of AI models, of course. And Anthropic says MHS devices can be controlled directly in real time via command-line prompts and API code files. But integrating an MHS system with an AI model through the Model Context Protocol lets scientists interact with devices using natural language, and lets models “reason through each step in an experiment, update parameters in real time, and, in some cases, recover from hardware errors without intervention,” Anthropic writes. 当然，通用的机器接口语言并不一定非要使用 AI 模型。Anthropic 表示，MHS 设备也可以通过命令行提示符和 API 代码文件直接进行实时控制。但 Anthropic 写道，通过“模型上下文协议”（Model Context Protocol）将 MHS 系统与 AI 模型集成，可以让科学家使用自然语言与设备交互，并使模型能够“推理实验的每一步，实时更新参数，并在某些情况下无需人工干预即可从硬件错误中恢复”。

Anthropic gave the example of a model like Claude adjusting a laser, checking the results via a separate camera, then repeating the process to automatically calibrate the whole system. MHS could also allow an AI model to focus a microscope, analyze the results, decide what part needs more observation, then automatically move the microscope to the relevant section to continue the experiment. Anthropic 举例说明，像 Claude 这样的模型可以调整激光，通过独立的摄像头检查结果，然后重复该过程以自动校准整个系统。MHS 还可以让 AI 模型对焦显微镜、分析结果、决定哪些部分需要进一步观察，然后自动将显微镜移动到相关区域以继续实验。

In a video, Anthropic also showed Claude reasoning how to get a robotic arm to pick up an aluminum can even though it had not been specifically trained on the required steps. And rather than reasoning through each step each time, Anthropic says MHS-enabled models can sequence steps across instruments by writing API scripts and adjusting them as conditions require. 在视频中，Anthropic 还展示了 Claude 如何推理出让机械臂捡起铝罐的方法，尽管它并未针对所需的步骤进行过专门训练。Anthropic 表示，MHS 支持的模型无需每次都重新推理每一步，而是可以通过编写 API 脚本并根据需要调整条件，从而实现跨仪器的步骤序列化。

Anthropic says MHS also includes a standardized tagging system to describe hardware’s real-world constraints for models that may have been trained more in the virtual world. That includes encoded information about the hardware’s physical characteristics (e.g., the weight and range of a robot arm) as well as its adjustable parameters, measurement options, and enforced safety limits. These tags can then be integrated into a reference file that can quickly provide an AI model with crucial information about a device it has no previous training experience with. Anthropic 表示，MHS 还包含一套标准化的标签系统，用于为那些可能更多是在虚拟世界中训练的模型描述硬件的现实约束。这包括有关硬件物理特性的编码信息（例如机械臂的重量和活动范围），以及其可调参数、测量选项和强制安全限制。这些标签随后可以集成到参考文件中，从而快速为 AI 模型提供有关其此前未接触过的设备的关键信息。

For now, Anthropic says it is working with “a first group of scientific research labs and advanced manufacturers” during an MHS preview period, including Amazon Web Services (Strands Robots), Hugging Face (LeRobot), Raspberry Pi, Automata, and Universal Robots. These partners will help Anthropic “build safety evaluations and develop best practices for AI systems operating physical equipment,” the company writes. After that, the plan is for MHS to eventually become an open source and “agent agnostic” standard for integrating AI and physical systems. 目前，Anthropic 表示在 MHS 预览期间，正与“第一批科学研究实验室和先进制造商”合作，包括亚马逊云科技（Strands Robots）、Hugging Face（LeRobot）、树莓派（Raspberry Pi）、Automata 和优傲机器人（Universal Robots）。该公司写道，这些合作伙伴将帮助 Anthropic “建立安全评估体系，并为操作物理设备的 AI 系统制定最佳实践”。此后，MHS 的计划是最终成为一个开源且“与智能体无关”（agent agnostic）的标准，用于集成 AI 和物理系统。

In early testing with scientific partners over the last year, Anthropic says it “saw MHS reduce the time it took to integrate devices, mak[ing] it possible to iterate faster in a variety of experimental settings.” “If you can test hypotheses faster, you could create general technologies faster,” Kemeny said in a promo video alongside the announcement. “This is how a century of progress can condense into a decade.” 在过去一年与科学合作伙伴的早期测试中，Anthropic 表示“MHS 缩短了集成设备所需的时间，使得在各种实验环境中进行更快速的迭代成为可能”。Kemeny 在随公告发布的宣传视频中说道：“如果你能更快地验证假设，你就能更快地创造通用技术。这就是如何将一个世纪的进步浓缩到十年之内。”