The Rise of AI Agents: From assistants to Teammates

Every technological era is marked by pivotal moments where niche technologies transition into foundational elements. Artificial intelligence stands at such a precipice, evolving from simple assistants awaiting instructions to sophisticated agents capable of functioning as workplace collaborators. Microsoft, celebrating its 50th anniversary, is actively shaping this future with azure AI foundry, a platform designed to underpin the digital economy of tommorow.

The AI landscape has undergone notable conversion in recent years. Organizations are now grappling with the complexities of integrating proprietary data, domain-specific actions, and secure environments with cutting-edge models to maximize their AI investments.Fine-tuning and distillation have emerged as critical processes for customizing models and unlocking enterprise value.

Azure AI Foundry: A Foundation for AI Advancement

Azure AI Foundry, introduced last fall, serves as a unified platform encompassing models, tools, and workflows, facilitating the design, customization, and management of AI applications and agents. It functions as a modular production system, streamlining the creation of intelligent systems through pre-built components, standardized processes, and continuous improvement driven by operational feedback. Currently,over 60,000 customers are leveraging azure AI Foundry,with access to more than 1,800 models and advanced tooling,to build their future.

AI applications are increasingly incorporating agentic capabilities. Since its public preview in late 2024,the Azure AI Agent Service has been adopted by tens of thousands of organizations. One notable example is Atom, an AI agent developed by Atomicwork, a leading agentic service management platform. Atom assists IT and HR teams in boosting productivity, resulting in significant gains, reduced operational costs, and enhanced employee satisfaction for early adopters.

Microsoft is introducing new capabilities to further assist customers in building and scaling advanced AI agents in a trustworthy manner.

Multi-Agent Systems: A New Paradigm

Microsoft is announcing the general availability of its agent framework, an extension of Azure AI Foundry’s open-source kit, semantic kernel.This framework is specifically designed to simplify the orchestration of multi-agent systems.

The Semantic Kernel agent framework, available immediately, streamlines agent coordination and significantly reduces the amount of code developers need to write. Organizations like KPMG are already utilizing Semantic Kernel to orchestrate workflows among specialized agents, substantially reducing development complexity.

At the heart of azure AI Foundry lies a feedback system that enables continuous improvement through real-time telemetry and user feedback. This is complemented by an advanced observability suite, including evaluations, tracing, and A/B testing, providing developers with unprecedented insight into agent behaviors and outcomes. These tools facilitate continuous improvements with each iteration of the AI system.

Consider Fujitsu: after leveraging Azure AI Agent Service and Semantic Kernel to develop an intelligent, scalable AI agent for sales automation, sales proposal creation became 67% more productive, allowing employees to concentrate on customer engagement and strategic planning.

we are using microsoft’s Semantic Kernel and Azure AI Agent Service to orchestrate multiple specialized AI agents and an orchestrator AI to coordinate them to answer questions as a team.

hirotaka Ito, Lead Engineer of AI, Digital System Platform unit, Fujitsu

AI Red Teaming: Scaling Trust and Mitigating Risks

As AI agent capabilities advance, managing risk becomes increasingly critical. However, a recent report by MIT Technology Review indicates that over half of businesses still rely on manual AI evaluation, an unsustainable approach for scaling.

To address this challenge, Microsoft is introducing the AI Red Teaming Agent, now in public preview. This agent systematically probes AI models to identify safety risks, integrating Azure AI Foundry’s robust evaluation systems with microsoft Security’s PyRIT (Python Risk Identification Tool) framework. The agent generates comprehensive reports, tracking improvements over time, and creating an AI safety testing ecosystem that evolves alongside the system.

At Accenture, we’re creating more agentic applications for our clients than ever before. Azure AI Foundry gives us a one-stop shop for all the right tools and services. To meet the growing demand while ensuring our applications are safe and secure, we’re looking into the AI Red Teaming Agent to automatically scan and ensure responsible development.

Nayan Paul, Managing Director, Accenture

Additionally, Microsoft’s new agentic evaluations, also available today, provide comprehensive risk and quality assessments, enabling teams to build trustworthy, compliant, and advanced AI agents from the outset. These evaluation metrics help teams systematically identify and mitigate potential risks in agent behavior, providing an additional layer of security and compliance.

Seamless Integration: Meeting Developers Where They Work

microsoft’s development ideology centers on creating technology that feels intuitive and natural.Over the past year, Microsoft has strengthened its collaboration with GitHub, starting with the launch of GitHub Models to its developer community, and more recently through open integrations with Azure AI Foundry and widely used open-source frameworks.

Microsoft is extending this approach with the new Azure AI Foundry extension for Visual Studio Code, now in public preview. Developers can now build, test, and deploy agent-based applications directly within their Integrated Development environment (IDE), eliminating the need for context switching.

Furthermore, Microsoft is enhancing Visual Studio Code by making agent mode in GitHub Copilot available to all users. Agent mode,MCP support,and code review elevate GitHub Copilot beyond code suggestions,transforming it into a peer programmer capable of performing complex,multi-step changes,running tests,and providing reasoning.

Looking Ahead: Building the Future Together

Whether streamlining business processes, transforming customer interactions, or pioneering multi-agent applications, Azure AI Foundry provides the robust infrastructure developers need to create intelligent agents at scale.

As Microsoft commemorates its 50th anniversary, the company is looking forward to the future. Join Microsoft at Microsoft Build from May 19-22, 2025, in Seattle and online, to discover the latest advancements in Azure AI Foundry. Build the next 50 years together.