Microsoft launches Copilot Cowork as AI agents reshape workplace software

Microsoft has introduced Copilot Cowork, a new artificial intelligence tool designed to autonomously carry out tasks across its Microsoft 365 ecosystem, as technology companies race to integrate AI agents into workplace software.

The feature, announced on Monday, allows users to deploy AI agents capable of handling multi-step activities such as analysing Outlook calendars, preparing meeting briefings and conducting detailed research, MarketWatch reported.

The tool is currently available in research preview and is expected to become more widely available later this month, according to a company statement.

Built using Anthropic’s AI technology

Microsoft said Copilot Cowork incorporates technology from Anthropic’s Claude Cowork, marking a deeper collaboration between the two companies in the rapidly expanding AI agents market.

“Working closely with Anthropic, we have integrated the technology behind Claude Cowork into Microsoft 365 Copilot,” Microsoft said in a statement cited by MarketWatch.

The company added that the application is model-agnostic, meaning it can operate across different large language models rather than relying on a single AI system.

Microsoft said the feature also operates within its existing enterprise security and governance frameworks, a key requirement for corporate customers deploying AI tools in workplace environments.

AI agents gaining momentum in enterprise software

The launch reflects a growing shift in the software industry toward AI agents capable of performing tasks autonomously, rather than traditional chatbots that require constant human prompts.

Anthropic introduced Claude Cowork earlier this year as an agentic AI system that can access local files and perform tasks independently. The release sparked debate among investors about whether such tools could disrupt established enterprise software providers, MarketWatch reported.

Another agent platform, OpenAI’s OpenClaw, has also gained attention for its ability to automate workflows directly on users’ machines.

Technology companies are now rushing to embed similar capabilities into their own products as AI agents begin to reshape workplace productivity software.

Microsoft responding to competitive pressures

Microsoft’s latest move comes as investors have raised questions about the company’s artificial intelligence strategy following a slowdown in Azure cloud growth earlier this year.

The company previously indicated it had redirected some computing capacity internally to strengthen Copilot capabilities, a shift that contributed to concerns on Wall Street and wiped about $357 billion from Microsoft’s market value in late January, MarketWatch reported.

By expanding Copilot with autonomous agents, Microsoft is seeking to retain control of AI workloads within its ecosystem of workplace tools, which already serve hundreds of millions of enterprise users globally.

Jefferies analyst Brent Thill said fears about AI disrupting Microsoft’s core business may be overstated.

According to MarketWatch, Thill noted that Microsoft’s strategy does not depend on owning the most powerful AI model but on controlling the platform where models are deployed and managed.

“Importantly, Microsoft is model-agnostic, and its strategy does not depend on owning the best model but on controlling the platform where models are deployed, governed and monetised,” he wrote in a research note.

AI agents set to redefine productivity tools

Industry analysts say the rise of agent-based AI systems could fundamentally change how employees interact with workplace software.

Rather than simply responding to queries, AI agents are designed to plan tasks, execute actions and coordinate workflows across multiple applications.

Microsoft’s Copilot Cowork represents one of the clearest signs yet that large technology companies are moving to embed these capabilities directly into mainstream productivity platforms.

As competition intensifies among AI developers and enterprise software providers, the ability to deploy reliable and secure AI agents may become a decisive factor in the next phase of workplace technology.