沃特世推出 Waters AI Link，将 Anthropic MCP 接入 waters_connect 并加入 MHS 研究计划

摘要 · 原文要点 Highlights

Accelerates research and development by enabling AI-augmented workflows and automated lab-in-the-loop cycles, already a new and fast-growing source of demand for Waters' instruments and consumables

通过启用 AI 增强的工作流与自动化的“实验室在环”循环来加速研发，这已成为沃特世仪器与耗材一项新的、快速增长的需求来源。

Leverages Waters' unique ability to take an AI-directed experiment from the research bench into development and into Empower™ Software, carrying its audit trail as the molecule moves into downstream, regulated settings

发挥沃特世独有的能力，把 AI 主导的实验从研究台面带入开发环节、再带入 Empower™ 软件，并在分子进入下游受监管环节时同步携带其审计追踪。

Joins Anthropic's Model Hardware Standard (MHS) research program designed to enable AI agents to operate Waters instrument hardware directly and autonomously, cutting instrument integration from weeks to hours

加入 Anthropic 的 Model Hardware Standard（MHS，模型硬件标准）研究计划。该计划旨在让 AI 智能体能够直接、自主地操作沃特世仪器硬件，把仪器集成时间从数周缩短到数小时。

Brings Anthropic's Model Context Protocol (MCP) to waters_connect™ Software, enabling scientists to direct Waters instruments through proprietary agents linked to large language models (LLMs), as well as Claude Science

将 Anthropic 的 Model Context Protocol（MCP，模型上下文协议）引入 waters_connect™ 软件，使科学家可以通过对接大语言模型（LLM）的专有智能体，以及 Claude Science，来指挥沃特世仪器。

沃特世 LC-MS 平台与 waters_connect 定量软件界面（图片来源：Waters 官方博客）

MILFORD, Mass., Oct. 13, 2026 /PRNewswire/ -- Waters Corporation (NYSE: WAT) today announced that it has launched Waters AI Link, a software capability that brings Anthropic's Model Context Protocol (MCP) to its informatics platform and has joined Anthropic's Model Hardware Standard (MHS) research program.

马萨诸塞州米尔福德 2026 年 10 月 13 日电（PRNewswire）：沃特世公司（NYSE: WAT）今日宣布，已推出 Waters AI Link。这是一项软件能力，将 Anthropic 的 Model Context Protocol（MCP）引入其信息学平台，同时公司已加入 Anthropic 的 Model Hardware Standard（MHS）研究计划。

"AI can generate hypotheses faster than any lab can test them, so the bottleneck has moved from computation to the bench," said Dave Leitham, Vice President, Informatics, Waters Corporation. "Every hypothesis an AI model generates becomes a physical sample that has to be separated and measured, multiplying the sample volume running through our instruments and columns."

“AI 生成假设的速度，比任何实验室验证假设的速度都快，所以瓶颈已经从计算端转移到了实验台。”沃特世公司信息学副总裁 Dave Leitham 表示。“AI 模型生成的每一个假设，都会变成一个必须分离和测量的实物样品，这让我们仪器和色谱柱上跑的样品量成倍增加。”

Leitham continued, "Waters has spent years building the compliant data foundation that regulated laboratories depend on. Our strategy is simple. In research, we connect AI agents to both the data and the instruments that generate it, so scientists can move at the speed of AI. That work carries forward on the same platform through development, then into Empower Software, the compliant system of record, as a molecule reaches quality control. No other company can support an AI-directed experiment from the research bench through to validated release, for small molecules and biologics alike, in a single continuous workflow. That is Waters' advantage, and it is why we are becoming a leader in the AI transformation of the laboratory."

Leitham 继续说道：“沃特世花了多年时间，构建受监管实验室所依赖的合规数据基础。我们的策略很简单。在研究阶段，我们把 AI 智能体同时接到数据以及产生数据的仪器上，让科学家能够以 AI 的速度前进。这项工作会在同一个平台上延续到开发阶段，并在分子进入质量控制环节时进入 Empower 软件，也就是合规的记录系统。没有任何其他公司能够在单一连续工作流中，支持从研究台面直到验证放行的 AI 主导实验，无论是小分子还是生物药。这就是沃特世的优势，也正是我们在实验室 AI 转型中正在成为领导者的原因。”

Together, Waters AI Link and MHS will enable scientists to direct Waters instruments and connected automation through Claude, including Claude Science, Anthropic's AI workbench for scientific research, and allow AI agents to operate Waters analytical instrumentation directly and autonomously. Waters continues to deliver new AI capabilities across its informatics portfolio as laboratories move toward AI-directed and increasingly autonomous operation, an area that is already generating new demand for Waters instruments and consumables.

Waters AI Link 与 MHS 结合起来，将使科学家能够通过 Claude（包括 Anthropic 面向科学研究的 AI 工作台 Claude Science）指挥沃特世仪器和与之相连的自动化设备，并让 AI 智能体能够直接、自主地操作沃特世分析仪器。随着实验室走向 AI 主导、并日益自主化的运行方式，沃特世持续在其信息学产品组合中交付新的 AI 能力，而这一领域已经在为沃特世的仪器与耗材带来新的需求。

一、把 AI 工作流贯通整个实验室，从研究到放行

Connecting AI Workflows Across the Laboratory, From Research to Release

waters_connect Software is Waters' cloud-enabled informatics platform for instrument control, data acquisition, and scientific workflows for liquid chromatography and mass spectrometry (LC-MS). The platform serves research and development laboratories, and that work transitions into Empower Software as it advances into regulated environments. For customers in research settings, these capabilities will enable the seamless transfer of workflows between dry labs and wet labs, allowing customers to accelerate outcomes via AI-augmented workflows or via fully automated lab-in-the-loop cycles. This includes autonomous physical testing of AI-generated hypotheses, followed by measurement capture that feeds directly back to the model, which hones its predictions and designs the next experiment. Because this work runs on waters_connect Software, those same workflows can then move seamlessly into development as the molecule advances.

waters_connect 软件是沃特世的云化信息学平台，用于液相色谱与质谱（LC-MS）的仪器控制、数据采集和科学工作流。该平台服务于研发实验室，而当工作推进到受监管环境时，这些工作会过渡到 Empower 软件。对于研究场景的客户，这些能力将实现工作流在干实验室与湿实验室之间的无缝迁移，让客户能够通过 AI 增强的工作流，或通过全自动的“实验室在环”循环更快获得结果。这包括对 AI 生成假设的自主实物测试，随后采集测量数据、直接反馈给模型，模型由此改进其预测并设计下一个实验。由于这些工作运行在 waters_connect 软件上，随着分子的推进，同样的工作流可以无缝进入开发阶段。

For customers in development settings, these capabilities can enable workflow optimization and efficiency gains, unlocking AI-enhanced method development, characterization, and testing with significantly less manual intervention, compressing timelines from weeks to days.

对于开发场景的客户，这些能力可以实现工作流优化和效率提升，以显著更少的人工干预解锁 AI 增强的方法开发、表征和检测，把周期从数周压缩到数天。

二、Waters AI Link 与 Anthropic MCP

Waters AI Link and Anthropic MCP

Model Context Protocol 官方架构示意：以统一协议连接 AI 应用与数据源、工具（图片来源：modelcontextprotocol.io）

The Model Context Protocol is an open standard that connects AI models and agents to software and data through a common interface. Waters AI Link brings MCP to waters_connect Software. Customers can link waters_connect Software to Claude and other large language models and AI agents through a natural language interface, pairing Waters' data handling and integration algorithms with the speed of AI-driven workflows, including the orchestration of connected devices such as sample preparation systems. Agents can create methods, run acquisitions, and process results without custom integration.

Model Context Protocol 是一项开放标准，通过统一接口把 AI 模型与智能体连接到软件和数据。Waters AI Link 把 MCP 引入 waters_connect 软件。客户可以通过自然语言界面，把 waters_connect 软件连接到 Claude 以及其他大语言模型和 AI 智能体，将沃特世的数据处理与集成算法同 AI 驱动工作流的速度结合起来，其中包括对样品前处理系统等已连接设备的编排。智能体可以创建方法、运行采集、处理结果，无需定制集成。

三、沃特世与 MHS

Waters and MHS

沃特世质谱实验室（图片来源：Waters 官方博客）

The Model Hardware Standard is an open specification, developed by Anthropic with scientific research institutions, that is designed to let AI agents safely operate physical laboratory devices such as instruments, liquid handlers, and robotic arms. Connecting proprietary instruments typically takes weeks or months of custom software engineering, because most devices do not speak to one another. MHS cuts that integration to hours or minutes. Once enabled, agents can run experiments overnight, adjust parameters as results arrive, recover from instrument errors without intervention, and scale research around the clock in automated lab-in-the-loop cycles.

Model Hardware Standard 是一项开放规范，由 Anthropic 与科研机构共同制定，旨在让 AI 智能体能够安全操作仪器、液体处理工作站、机械臂等物理实验室设备。连接专有仪器通常需要数周或数月的定制软件工程，因为大多数设备彼此之间无法通信。MHS 把这一集成时间缩短到数小时或数分钟。一旦启用，智能体可以在夜间运行实验、在结果到达时调整参数、在无人干预的情况下从仪器错误中恢复，并在自动化的“实验室在环”循环中全天候扩展研究。

Waters AI Link is available now for early access, with general availability in November for waters_connect Software customers. Waters' participation in the MHS research program is effective immediately.

Waters AI Link 目前已开放早期访问，并将于 11 月面向 waters_connect 软件客户正式发布。沃特世参与 MHS 研究计划即刻生效。

Waters, waters_connect, and Empower are trademarks of Waters Corporation or its affiliates. Anthropic and Claude are trademarks of Anthropic, PBC. Model Context Protocol is a trademark of LF Projects, LLC. All other marks are the property of their respective owners.

Waters、waters_connect 和 Empower 是沃特世公司或其关联公司的商标。Anthropic 和 Claude 是 Anthropic, PBC 的商标。Model Context Protocol 是 LF Projects, LLC 的商标。所有其他标识均为其各自所有者的财产。

资料来源：Waters Corporation 官方新闻稿，2026 年 10 月 13 日由 PRNewswire 发布（prnewswire.com/apac/news-releases/waters-launches-ai-link-advancing-ai-enabled-lc-ms-workflows-from-research-to-release-302906480.html）。英文原文与中文译文逐段对照，文字照原文保留，未作删改；中文译文如有出入，以英文原文为准。
配图来源：沃特世公司官方博客（waters.com/blog，2026 年 10 月 13 日）与 Model Context Protocol 官方文档（modelcontextprotocol.io）。图片版权归原权利人所有，此处仅作新闻转述配图使用。