Microsoft launches Copilot Cowork with Anthropic to automate work tasks without human supervision

AI Agents at Work: Microsoft Copilot Is Getting Its Own Version of Claude Cowork

Microsoft Copilot probably isn't something you think about a lot, unless your company pays for you to use it at work. Microsoft has been fighting for consumers whose hearts and minds were quickly captured by other AI tools like ChatGPT and Gemini. The company's latest wave of agentic updates, announced Monday, is its sharpest weapon yet. The biggest new feature is Copilot Cowork, built in collaboration with Anthropic. If you've heard of or used Anthropic's Claude Cowork, Microsoft's version will feel similar. Copilot Cowork can use information in your files, email and calendar to independently complete assignments, no human supervision needed. It can create spreadsheets, run reports and do research for you. "Cowork is the new chat. It's the new way of interacting with AI," said Charles Lamanna, Microsoft president of business apps and agents. Instead of overseeing AI and chatting with it, we can now entirely delegate tasks to it like a fellow team member. "With chat, you're babysitting every step -- this is much more like 'fire and forget' with Cowork to get the job done." For example, Lamanna said he used Cowork to analyze his meeting calendar for the next three months. The AI used his email and calendar history to understand what upcoming meetings may not be necessary for him to attend, and it pulled together its recommendations in an easy-to-view chart. After Lamanna reviewed it, Cowork declined some meetings, with AI-written meeting notes attached if needed. The AI's 40-minute "delightful and practical" process saved him and his executive assistant hours worth of time so they could focus on more important duties. Cowork is rolling out now on a limited basis as it's a research preview concept. Microsoft also announced it will be making its AI agent platform, Agent 365, generally available on May 1. Agent 365 is a way for companies to oversee and manage all of the agents, or bots, that employees are using for their work. Microsoft itself has created more than a half-million AI agents using Agent 365, the company shared in a statement. New AI models from Anthropic and OpenAI will also be made available in Copilot. Smartly, the company isn't picking a side in the growing feud between the AI startups. Read More: AI Isn't Human and We Need to Stop Treating It That Way, Says Microsoft AI CEO Agentic AI tools like the ones Microsoft is building are extremely popular for workers. Despite only being a research preview, Claude Cowork has garnered a lot of fans -- and sparked a lot of worry on Wall Street. Major tech stocks fell at the end of January as Anthropic's AI developments cast doubts on the future of work. New AI tools like Cowork, Claude Code and even OpenAI's Codex are becoming increasingly capable of replacing traditional software products, like the kind Microsoft is known for. So it makes sense that Microsoft would want to bring that agentic prowess to its own AI. Agentic AI has been a major area of focus for AI companies recently. OpenClaw, an open source agentic project that went viral this year, is one of many examples of why tech execs think 2026 will be the year of agentic AI. Lamanna said that "the shape of what we do on a day-to-day basis will change," but AI ought to give time back to people to focus on high-value tasks. We're entering a new arc, going from having a human use AI to do a task quicker to delegating it entirely to an AI agent, he said. As this tech becomes more available, there are a lot of questions about the best way to integrate AI into our work lives. Many workers are worried about having their jobs replaced by AI, fueled by AI-centric layoffs at Amazon and Block. For those who manage to keep their jobs, one study found that AI may actually make their work days longer and less enjoyable. Like any new tech, the implementation of AI will determine how effective it is -- and how much it actually helps you.

Microsoft taps Anthropic for Copilot Cowork in push for AI agents

March 9 (Reuters) - Microsoft (MSFT.O), opens new tab is adding Anthropic's AI technology to its Copilot service to tap growing demand for autonomous agents, weeks after the startup's new tools sparked a selloff in software stocks. The company on Monday unveiled Copilot Cowork, a tool based on Anthropic's viral Claude Cowork offering, which has captivated Silicon Valley with its ability to handle complex tasks such as creating apps, building spreadsheets and organizing large volumes of data with limited human oversight. Microsoft is betting that its long-standing ties with enterprise customers and its ⁠focus on security and data controls will help it win business from companies interested in AI agents but wary of deploying them without safeguards. "We work only in a cloud environment and we work only on behalf of the user. So you know exactly what information it (Copilot Cowork) has access to," Jared Spataro, who leads Microsoft's AI-at-Work efforts, told Reuters. Cloud Cowork only works locally on the device and most companies feel "very uncomfortable" with that, he said. "We're the opposite." The launch comes weeks after Anthropic ⁠introduced new tools for Claude that intensified investor concerns about the threat AI agents could pose to traditional software companies, triggering to a selloff in the sector. Microsoft's own shares fell nearly 9% in February. Copilot Cowork tool is currently in testing and will be available to early-access ⁠users later this month, Microsoft said. The company did not disclose pricing, but said some usage would be included in its $30-per-user, per-month M365 Copilot offering for enterprises, with additional usage available for ⁠purchase. Microsoft also said it is making Anthropic's latest Claude Sonnet models available to M365 Copilot users. The service had previously relied only on OpenAI's GPT ⁠models. The move deepens Microsoft's ties with Anthropic at a time when investors have questioned its dependence on OpenAI, which accounts for nearly 45% of Microsoft's cloud business contract backlog. Reporting by Aditya Soni in Bengaluru; Editing by Tasim Zahid Our Standards: The Thomson Reuters Trust Principles., opens new tab

Microsoft debuts Copilot Cowork built on Anthropic's tech and E7 product suite | Fortune

Microsoft has announced a new set of products to get enterprise customers to build AI agents on its platform, including a new Copilot Cowork product built on top of Anthropic's AI product Claude Cowork and a new business productivity software bundle that includes its own AI offerings. The new products, which Microsoft is calling "Wave 3 of Microsoft 365 Copilot" journey, come as the software giant seeks to combat increased competition in the AI agent space both from rival business productivity software companies, such as Salesforce, and frontier AI companies such as OpenAI and Anthropic, which received billions of dollars in strategic investments from Microsoft but are nonetheless increasingly pursuing the U.S. tech giant's traditional customer base. Microsoft also faces competition for AI agents from open source offerings like OpenClaw. Microsoft hopes to assuage investors concerned that AI agents will reduce companies' need to rely on traditional software-as-a-service providers. The company's shares have fallen more than 14% since Anthropic debuted its Claude Cowork product in mid-January. At the heart of Microsoft's announcement is Copilot Cowork, a new feature built in close collaboration with Anthropic. Copilot Cowork is designed to handle long-running, multi-step tasks -- such as preparing for a customer meeting by assembling a presentation, pulling together financials, emailing the team, and scheduling prep time -- all from a single request. "We really believe right now is an inflection point," Jared Spataro, Microsoft's chief marketing officer for AI at Work, told Fortune. "The inflection point for us is Copilot taking on these agentic capabilities and going from assistance to real doing." Spataro said that Copilot Cowork uses Anthropic's Claude model as the AI powering its reasoning and uses the same "agentic harness" -- the system that allows the AI model to use other software tools and the guardrails around how it functions -- as Anthropic's Claude Cowork, but that Copilot Cowork includes features that make it easier to build the kinds of agents that companies need. For instance, Microsoft's implementation runs in the cloud within a customer's Microsoft 365 tenant, meaning it is covered by the company's enterprise data protection and integrated with what Microsoft calls "Work IQ" -- a layer of intelligence drawn from a user's emails, files, documents, meetings, and chats. Anthropic's Claude Cowork, by contrast, runs locally on a user's device. "We actually don't work locally, and that's a feature, not a bug," Spataro said. He described Anthropic's offering as "a fantastic tool" but one with "limitations" in a corporate environment, noting it lacks access to cloud-based enterprise data and raises security concerns when deployed at scale. "What Anthropic has done is demonstrate the value of these agentic capabilities and show us practically what it could look like," he added. "Microsoft is all about commercialization." Copilot Cowork feature is currently being piloted with select customers and will become available as a research preview in March through Microsoft's new Frontier Worker product suite. The company also announced that Anthropic's Claude model is now available across the full Copilot Chat experience, not just in Microsoft's Researcher and Excel features where it was previously offered. While Microsoft initially built all of its Copilot offering around OpenAI's models, it has now shifted to a flexible approach that allows customers to pick any model to power their AI assistants and agents. "Every 60 days at least, there's a new king of the hill," Spataro said. "There's so much demand for a platform that doesn't feel like I have to skip over to the next vendor." The company said that its Agent 365 product -- which is a so-called control plane or "orchestration platform" for AI agents, allowing IT and security teams to monitor, govern and secure agents, including those created using other vendors' software, across an organization -- will be generally available from May 1, priced at $15 per user per month. Spataro said the key insight behind Agent 365 came from recognizing that the same management infrastructure used for human employees -- tools like Entra, Defender, Purview, and Intune -- could be extended to manage AI agents as well. "AI agents are as subject to phishing attacks as people are," he said. "As soon as an AI agent has an email address, they get spam too, and they can respond to it." Microsoft said that in just two months of preview, tens of millions of agents had appeared in the Agent 365 registry. Internally, Microsoft said it now has visibility into more than 500,000 agents across the company, with the most widely used focused on research, coding, sales intelligence, customer triage, and HR self-service. Finally, Microsoft announced the Microsoft 365 E7 Frontier Worker Suite, also available from May 1, priced at $99 per user per month. The bundle combines Microsoft 365 E5 -- long the company's premium business productivity suite -- with Microsoft 365 Copilot and Agent 365. It also includes the Microsoft Entra Suite and advanced Defender, Intune, and Purview security capabilities. The $99 price is below what customers would pay if they purchased these capabilities separately, according to Microsoft. The component pricing of the constituent parts -- E5 at $60, Entra Suite at $12, Copilot at $30, and Agent 365 at $15 -- adds up to $117 per user. While many analysts have speculated that AI agents will eventually force software-as-a-service companies such as Microsoft to shift away from per-user pricing toward consumption-based models, Spataro said Microsoft's customers are not currently demanding this. "I think I have the most data points of anyone in the industry -- customers want per user right now," he said. "Doesn't mean they always will, but that's what they want currently." Microsoft said Copilot paid seats have grown more than 160% year over year, with daily active usage up tenfold. The number of customers deploying Copilot at significant scale -- more than 35,000 seats -- has tripled year over year, and 90% of the Fortune 500 now use Copilot, according to the company. It also said that 80% of the Fortune 500 are now using Microsoft AI agents in some capacity.