FRST在Microsoft 2024上推出，這是Azure AI鑄造的一部分的服務，它是基於Microsoft以前在AI驅動的自動化中的工作，包括其Azure OpenAi服務，該服務使企業可以將生成的AI集成到他們的工作流程中使用內部系統，自動化決策過程並執行複雜的任務。在-apps/“> azure邏輯應用，Azure AI Foundry SDK, which provides developers with tools to design AI-driven automation without the need for extensive machine learning expertise.
Azure AI Agent Service supports large language models (LLMs) from OpenAI, Meta, Mistral, and Cohere, offering businesses the flexibility to choose models that best fit their needs. These AI agents can be used for customer service, IT support, financial analysis, and research automation, among other use cases.
“Azure AI Agent Service is a flexible, use-case-agnostic platform for building, deploying, and managing AI agents as micro-services,”Microsoft stated in its official announcement earlier this month.
Azure AI Agent Service provides the foundational platform for building AI agents, which can be orchestrated using frameworks like Microsoft’s AutoGen. In such a setup, AutoGen can serve as the orchestration layer, allowing multiple AI agents (including those created with Azure AI Agent Service) to collaborate and complete complex tasks.
Microsoft already released Magentic-One as an implementation of such a multi-agent system built using AutoGen, showcasing the practical application of these technologies in creating advanced AI solutions.
How Azure AI Agent Service Works
Unlike standalone AI tools, Azure AI Agent Service functions as a fully managed automation platform, allowing companies to integrate AI agents into their operations without requiring significant development effort.該服務提供了對企業數據源的直接訪問，並通過利用微軟的生態系統來啟用AI驅動的決策。
AI代理可以執行自動任務，例如響應客戶查詢，匯總文檔，檢索業務智能洞察力，並執行IT Worktrows。 These agents can also work alongside human employees, handing off tasks when needed to maintain efficiency and reliability.
Azure AI Agent Service Call Center Agent sample case (Image: Microsoft)
By integrating with Azure Logic Apps, AI agents can trigger automated workflows, reducing the need for manual intervention in repetitive business processes. Meanwhile, Microsoft Fabric enables agents to process structured and unstructured data for real-time analytics.
Example of an implementation of Magentic-One for AI agent orchestration (Image: Microsoft)
Security, Observability, and Compliance
Enterprise AI adoption often faces challenges related to security and data governance. Microsoft通過在Azure AI代理服務中包含強大的安全性和合規性來解決這些問題。
Microsoft還集成了基於OpentElemetry的監控，使企業可以跟踪AI代理人如何與企業系統進行交互，從而確保與內部策略和監管案例的依從性策劃Azure AI代理服務以使關鍵業務功能自動化:
Bristol Myers Squibb正在將AI代理集成到其研究操作中，在此過程中，它們有助於處理科學數據，回答員工的查詢並分析內部知識基礎。 Core42正在使用該平台來增強其AI驅動的自動化解決方案，並將AI驅動的工作流程與企業客戶端集成在一起。富士通採用了AI代理來自動化客戶互動並優化銷售分析，從而提高了客戶服務效率。 NTT數據正在利用該平台來獲得AI驅動的銷售情報，從而幫助企業獲得了對客戶需求的實時見解。 Youngwilliams正在開發AI代理商，以協助國家衛生和公共服務，使政府互動更加有效。
對於希望整合AI自動化的企業，這些早期的採用案例可以瞥見企業如何按大規模部署AI代理。
[嵌入式內容]
ai代理市場
azure ai azure ai aid Agent Service Service inters競爭激烈的公司都投入了一些大型技術公司。 Microsoft faces competition from Google which has also developed its own AI-driven agent framework, AgentSpace, which also focuses on developer-driven AI tools, allowing companies to build customized AI automation frameworks.
While competitors provide powerful AI models, Microsoft’s key advantage lies in its deep integration with enterprise applications, allowing Azure AI Agent Service to work seamlessly with Microsoft 365, Dynamics 365, and Teams.
The launch of Azure AI Agent Service represents Microsoft’s most structured approach to AI-driven enterprise automation yet, consolidating past AI initiatives into a scalable and deployable platform.
What’s Next for Azure AI Agent Service?
With the public preview now available, Microsoft plans to expand Azure AI Agent Service with additional AI models, security enhancements, and deeper integrations across its cloud ecosystem.該公司表示，未來的更新將改善AI代理的協作，使更複雜的自動化工作流程以及增強AI驅動決策的可觀察性。
隨著AI的採用，Microsoft的AI Automation方法可以在企業融入其操作中的核心作用。已經使用Microsoft雲服務的公司可能會發現在其現有工作流程中部署AI代理商，從而提供了比其他AI自動化平台的競爭優勢。
