{"id":62439,"date":"2025-08-06T14:56:14","date_gmt":"2025-08-06T12:56:14","guid":{"rendered":"https:\/\/www.schneider.im\/?p=62439"},"modified":"2025-08-06T18:01:05","modified_gmt":"2025-08-06T16:01:05","slug":"microsoft-azure-ai-foundry-agent-service-deep-research-in-public-preview","status":"publish","type":"post","link":"https:\/\/www.schneider.im\/lu\/microsoft-azure-ai-foundry-agent-service-deep-research-in-public-preview\/","title":{"rendered":"Microsoft Azure AI Foundry Agent Service: Deep Research in public preview"},"content":{"rendered":"

Effective July 7, 2025, Microsoft has introduced a powerful new capability<\/strong> in Azure AI Foundry: Deep Research<\/strong>, now available in public preview (registration necessary). You can automate web-scale research tasks<\/strong> using programmable AI agents<\/strong> with this new service.<\/p>\n

With Deep Research<\/strong>, you can build agents that plan, analyze, and synthesize information<\/strong> from across the web<\/strong>, generate transparent and auditable outputs<\/strong>, and integrate <\/strong>into your business workflows<\/strong>.<\/p>\n

 <\/p>\n

\"\"<\/p>\n

 <\/p>\n

What is Azure AI Foundry Agent Service?<\/strong><\/h2>\n

Azure AI Foundry Agent Service<\/strong> is a platform<\/strong> from Microsoft that helps you run AI agents<\/strong>. It connects models, tools, and frameworks<\/strong> into a unified system for operating AI agents<\/strong>. These agents can make decisions, invoke tools, and complete tasks independently or in coordination with other agents or humans.<\/p>\n

Reminder<\/strong>: Microsoft now differentiates between \u201cAI agents\u201c\/\u201dagents\u201d, meaning autonomous agents, and \u201cCustomer service representative\u201c\/\u201dservice representative\u201d\/\u201drepresentative\u201d, meaning humans: https:\/\/www.schneider.im\/microsoft-entra-agent-id-a-new-era-of-identity-for-ai-agents\/<\/a>.<\/p>\n

 <\/p>\n

What is Deep Research?<\/strong><\/h2>\n

While classic large language models (LLMs) focus on generating fluent responses based on statistical patterns in text, reasoning LLMs<\/strong> are designed to break down complex tasks, follow multi-step logic, and interact with tools<\/strong>.<\/p>\n

Deep Research<\/strong> is a programmable agent<\/strong> that can be embedded into apps, workflows, or other agents to perform research automatically<\/strong>. It uses a reasoning LLM, called o3-deep-research<\/strong>.<\/p>\n

It is built for tasks that require deep analysis, synthesis, and traceable insights<\/strong> from online sources<\/strong>.<\/p>\n

 <\/p>\n

Key Features<\/strong><\/h2>\n