Microsoft unveils 50+ AI tools for the ‘agentic web’ at Build 2025
▼ Summary
– Microsoft unveiled over 50 AI tools at its Build conference to advance autonomous “agentic web” systems that perform tasks with minimal human intervention.
– The company introduced GitHub Copilot as an autonomous coding agent capable of refactoring code, fixing defects, and collaborating with other AI agents.
– Microsoft’s Azure AI Foundry now supports multi-agent workflows, enabling businesses to deploy specialized AI agents for complex tasks and automation.
– Local AI capabilities were expanded with Windows AI Foundry, allowing on-device AI deployment for privacy, cost savings, and reduced internet dependency.
– Microsoft addressed AI security and governance with Microsoft Entra Agent ID and Purview integration to manage “agent sprawl” and ensure compliance.
Microsoft has unveiled an ambitious vision for AI-powered automation at its Build 2025 conference, introducing over 50 new tools designed to accelerate the shift toward what the company calls the “agentic web.” These innovations span the company’s entire ecosystem—from Azure and Windows to GitHub and Microsoft 365—empowering developers to build autonomous systems capable of independent decision-making and task execution.
At the heart of this strategy is a fundamental reimagining of AI interactions. Unlike traditional chatbots or assistants that merely respond to prompts, AI agents proactively initiate workflows, collaborate with other systems, and operate with minimal human oversight. Frank Shaw, Microsoft’s Chief Communications Officer, emphasized this evolution, stating that advancements in reasoning and memory are enabling AI to solve problems in entirely new ways.
From Code Assistants to Autonomous Developers
GitHub, Microsoft’s developer platform, is taking center stage in this transformation. The newly enhanced GitHub Copilot coding agent now handles more than just code suggestions—it autonomously refactors, tests, and even implements new features. For complex projects, multiple Copilot agents can collaborate across the software development lifecycle. In a move toward transparency, Microsoft is open-sourcing GitHub Copilot Chat within Visual Studio Code, inviting community contributions to refine these capabilities.
Multi-Agent Systems for Enterprise Workflows
Businesses looking to scale AI adoption will benefit from updates to Azure AI Foundry, which now supports multi-agent orchestration. Ray Smith, VP of AI Agents at Microsoft, explained that breaking tasks across specialized agents improves reliability and maintainability. The platform includes tools for debugging and managing agent interactions, along with support for open protocols like Agent2Agent (A2A) and Model Context Protocol (MCP)—critical for enterprises deploying AI at scale.
The Rise of On-Device AI
While cloud-based AI remains dominant, Microsoft is doubling down on local processing with Windows AI Foundry, a toolkit for deploying AI directly on devices. This approach reduces latency, enhances privacy, and cuts costs by minimizing cloud dependency. Steven Bathiche, a Microsoft technical fellow, noted the rapid pace of innovation in client-side AI, with predictions materializing in months rather than years.
Open Standards for Interoperability
To foster a broader ecosystem, Microsoft is championing open protocols like MCP and NLWeb, which aim to standardize agent communication much like HTML did for the web. Jay Parikh, leading Microsoft’s Core AI team, stressed the importance of interoperability in accelerating innovation across platforms.
A New Computing Paradigm
Kevin Scott, Microsoft’s CTO, likened the current excitement to the early days of the internet, where modular components enabled rapid experimentation. With its comprehensive suite of tools, Microsoft is positioning itself as a linchpin in the agentic revolution—one where AI doesn’t just assist but anticipates, decides, and reshapes how work gets done.
For enterprises, the implications are clear: automation will soon extend beyond repetitive tasks to complex decision-making, requiring organizations to adapt quickly or risk falling behind. The question isn’t if AI agents will transform industries, but how swiftly businesses can harness their potential.
(Source: VentureBeat)
