Microsoft Foundry (formerly Azure Artificial Intelligence (AI) Foundry)

Research Question

What is the complete set of features, functions, and capabilities offered by Microsoft Foundry, and how do those capabilities support the full Artificial Intelligence (AI) development lifecycle, from model selection and fine-tuning through deployment, evaluation, and production governance, in a regulated enterprise?

Findings

(Populated from §6 Synthesis above.)

Executive Summary

Microsoft Foundry already covers most of the enterprise AI application lifecycle natively, including model discovery, benchmarking, selected fine-tuning, agent development, evaluation, deployment, tracing, and Azure-native security controls.

Its strongest areas are model access, agent tooling, evaluation and observability, and deployment flexibility, while its weakest areas for regulated enterprises are not missing features so much as fragmented governance boundaries across connected Azure services, tenant administration, and preview-only control surfaces.

Prompt Flow is now a legacy transition surface rather than the forward path, and Microsoft's current strategic direction is agent-centric development through Agent Service, workflows, project endpoints, and Microsoft Agent Framework.

For a regulated enterprise, Microsoft Foundry is best treated as a strong Azure-native AI application factory, not as a self-sufficient governance plane, because identity, search, storage, compliance, and tenant-wide Copilot controls still require surrounding Azure and Microsoft 365 governance layers.

Key Findings

1. Microsoft Foundry consolidates the former Azure AI Studio and Azure AI Foundry experience into a single resource and project model, and Microsoft can upgrade Azure OpenAI resources in place without changing existing Azure OpenAI endpoints, keys, or saved state.
2. The model catalogue is broad and practically useful, because Microsoft documents more than 1,900 models, side-by-side comparison, public-benchmark leaderboards, model cards, deployment tabs, and filterable deployment and licence metadata within the same discovery surface.
3. Microsoft Foundry supports meaningful native model customisation, especially Low-Rank Adaptation (LoRA) based fine-tuning and selected optimization methods, but customization support is narrower than catalogue support and some partner or open-model paths still depend on managed compute or classic-era deployment patterns.
4. The development stack is now decisively agent-centric, combining prompt agents, workflow agents, hosted agents, tool catalogues, memory, Foundry IQ, Microsoft's managed knowledge layer, unified SDKs, and playgrounds, while Prompt Flow remains only as a legacy classic feature with retirement and migration guidance already published.
5. Evaluation and observability are first-class platform features, because Microsoft documents lifecycle evaluation from benchmark-driven model selection through agent testing, continuous evaluation, dashboard monitoring, OpenTelemetry tracing, and automated red-teaming, even though some runtime-inspection paths remain preview.
6. Deployment and serving flexibility are broad for a managed platform, because Microsoft Foundry supports standard, provisioned, batch, data-zone, regional, and managed-compute serving patterns, but that same flexibility pushes architects to make explicit data-residency, quota, and cost-governance decisions.
7. Microsoft Foundry has serious enterprise-governance machinery, including Microsoft Entra identity, Foundry RBAC roles, Private Link, bring-your-own-storage, guardrails, task-adherence checks, and Control Plane fleet views, but those controls remain distributed across Foundry, connected Azure resources, and several preview-only surfaces.
8. Microsoft Foundry integrates well with Azure OpenAI, Azure Machine Learning (Azure ML), Azure AI Search, OneLake, and Microsoft Copilot surfaces, but those integrations make the platform more useful as a developer-side AI application factory than as a complete enterprise governance plane in its own right.

Assumptions

• Preview features are counted as existing capabilities but not as production-grade baseline controls because Microsoft publishes them in current documentation while also disclaiming service-level agreements and full support.
• Azure Machine Learning (Azure ML) integration is assessed only at the documented Microsoft Foundry model-catalogue and deployment surface, not at the level of general Azure ML workspace feature parity, because that broader Azure ML scope is out of scope for this item.
• This survey treats Microsoft documentation as authoritative for product-capability claims and does not infer undocumented feature parity across every region or every model family.

Analysis

Microsoft's documentation supports a direct answer that Microsoft Foundry is already broad enough to cover most lifecycle stages natively, so the platform is not limited to model hosting or playground experimentation.

The stronger competing interpretation is that Microsoft Foundry is now a complete enterprise AI control plane, but the architecture, networking, identity, storage, and publish-to-Copilot documents do not support that stronger claim because multiple essential governance surfaces remain outside the Foundry resource itself.

Another plausible rival interpretation is that the platform is still mostly Azure OpenAI with new branding, but the agent runtime, workflow builder, Foundry IQ knowledge layer, Control Plane, and unified project endpoint together show a materially broader application platform than standalone Azure OpenAI provided.

The best-supported conclusion is therefore narrower and more decision-useful: Microsoft Foundry is a strong Azure-native build, test, deploy, and operate platform for AI applications, but regulated enterprises still need explicit surrounding governance for connected data sources, tenant administration, identity, and release policy.

Risks, Gaps, and Uncertainties

• Several of the most ambitious management surfaces, including parts of Control Plane, workflow and hosted-agent support, publish-to-Copilot, toolbox, and some tracing and guardrail behaviors, are still preview, which means production commitments and operational stability are not yet equivalent across the whole stack.
• Microsoft's public documentation describes model families, regions, and quotas as moving targets, so any procurement or architecture decision still needs a live region and quota check in the target subscription before deployment.
• The survey establishes breadth of documented capability more strongly than day-two operational maturity, because many documents describe setup and feature availability rather than large-scale reference operations in tightly regulated production estates.

Open Questions

• Which Microsoft Foundry preview surfaces reach general availability first, specifically Control Plane governance panes, hosted-agent observability, and publish-to-Copilot?
• How much of current Microsoft Agent Framework functionality eventually becomes a managed Microsoft Foundry surface rather than remaining an external code framework?
• How much operational evidence will Microsoft publish on large regulated-enterprise deployments using DataZone, private networking, Foundry IQ, and preview governance controls together?

sources

• [x] Microsoft Learn What is Microsoft Foundry? - product definition, evolution table, available models, and key capabilities
• [x] Microsoft Azure Microsoft Foundry product page - product composition and integrated-services summary
• [x] Microsoft Learn Microsoft Foundry documentation home - current documentation hub and feature map
• [x] Microsoft Learn Microsoft Foundry architecture - resource hierarchy, connected services, and security separation of concerns
• [x] Microsoft Learn Microsoft Foundry Models overview - catalogue structure, model categories, filters, and benchmark surfaces
• [x] Microsoft Learn Model leaderboards in Microsoft Foundry - leaderboard methodology and benchmark categories
• [x] Microsoft Learn Models sold directly by Azure - Azure Direct model category, support, billing, and model families
• [x] Microsoft Learn Deployment overview for Microsoft Foundry Models - standard versus managed compute deployments
• [x] Microsoft Learn Deployment types for Microsoft Foundry Models - global, data-zone, regional, provisioned, batch, and developer deployment types
• [x] Microsoft Learn Region support for Microsoft Foundry - project-region list and feature-specific region caveats
• [x] Hu et al. (2021) LoRA: Low-Rank Adaptation of Large Language Models - original definition of the Low-Rank Adaptation method
• [x] Microsoft Learn Customize a model with fine-tuning - fine-tuning methods, supported-model examples, permissions, and deployment flow
• [x] Microsoft Learn Plan and manage costs for Microsoft Foundry - cost planning, billing models, and fine-tuned model cost behavior
• [x] Microsoft Learn SDK overview for Microsoft Foundry - SDK choices, endpoint patterns, and authentication guidance
• [x] Microsoft Learn Quickstart: Build with models and agents - project endpoint, model call, and agent creation examples
• [x] Microsoft Learn Foundry Agent Service overview - agent types, tools, observability, publishing, and enterprise capabilities
• [x] Microsoft Learn Build a workflow in Microsoft Foundry - workflow patterns, visual orchestration, and human-in-the-loop support
• [x] Microsoft Learn Tool catalog for Microsoft Foundry Agent Service - built-in tools, custom tools, Model Context Protocol (MCP), and toolbox
• [x] Model Context Protocol Introduction - authoritative definition of Model Context Protocol (MCP)
• [x] A2A Protocol Specification - authoritative definition of the Agent-to-Agent (A2A) protocol
• [x] Microsoft Learn Memory usage in Foundry Agent Service - managed memory stores and per-user scoping
• [x] Microsoft Learn What is Foundry IQ? - knowledge-base and agentic-retrieval capabilities
• [x] Microsoft Learn Publish agents to Microsoft 365 Copilot and Microsoft Teams - publishing workflow and preview status
• [x] Microsoft Learn Microsoft Foundry playgrounds - model and agents playground capabilities
• [x] Microsoft Learn Prompt flow in Microsoft Foundry portal (classic) - Prompt Flow scope, lifecycle, and retirement status
• [x] Microsoft Learn Audit, rebuild, and validate Prompt Flow for Agent Framework migration - migration path from Prompt Flow to Microsoft Agent Framework
• [x] Microsoft Learn Observability in generative AI - evaluation, monitoring, tracing, and red-teaming overview
• [x] Microsoft Learn Evaluate your AI agents - agent evaluation flow and built-in evaluator categories
• [x] Microsoft Learn Agent tracing in Microsoft Foundry - tracing scope, OpenTelemetry, and preview status
• [x] OpenTelemetry What is OpenTelemetry? - authoritative definition of OpenTelemetry tracing and telemetry concepts
• [x] Microsoft Learn Monitor agents with the Agent Monitoring Dashboard - production monitoring, continuous evaluation, and dashboard metrics
• [x] Microsoft Learn Foundry Control Plane overview - cross-project fleet management, compliance, and security panes
• [x] Microsoft Learn Role-based access control for Microsoft Foundry - Foundry-specific roles and scope model
• [x] Microsoft Learn Authentication and authorization in Microsoft Foundry - control plane versus data plane and Microsoft Entra identity guidance
• [x] Microsoft Learn Configure network isolation with private link - private endpoints, public network access, and virtual network injection
• [x] Microsoft Learn Guardrails and controls overview in Microsoft Foundry - guardrail intervention points and supported risk categories
• [x] Microsoft Learn Task Adherence in Microsoft Foundry and Content Safety - task-misalignment detection and tool-call blocking logic
• [x] Microsoft Learn Add a new connection to your project - supported connection types and preview connectors
• [x] Microsoft Learn Bring your own Azure Storage to Microsoft Foundry - bring-your-own-storage patterns and capability hosts
• [x] Microsoft Learn Upgrade Azure OpenAI to Microsoft Foundry - upgrade behavior, preserved endpoints, and governance implications
• [x] Microsoft Learn Endpoints for Microsoft Foundry Models - deployment and endpoint model
• [x] Microsoft Learn Microsoft Foundry Models in Azure Machine Learning - Azure Machine Learning (Azure ML) integration surface for the model catalogue
• [x] Microsoft Fabric Use OneLake as a knowledge source for Microsoft Foundry - Fabric and OneLake grounding path
• [x] Azure AI Search Indexed OneLake knowledge source for agentic retrieval - Azure AI Search knowledge-source creation for OneLake
• [x] David Mitchell (2026) Vendor-agnostic enterprise Artificial Intelligence (AI) capability model: Microsoft Copilot and GitHub families vs AWS Bedrock ecosystem - prior repository synthesis on enterprise AI capability boundaries
• [x] David Mitchell (2026) Alternative Continuous Integration and Continuous Delivery pipeline platforms for governing agents built with Microsoft Copilot Studio - prior repository evidence on Microsoft-side pipeline governance boundaries
• [x] David Mitchell (2026) What identity and access management model is required for Artificial Intelligence (AI) agents and low-code artefacts operating within enterprise systems? - prior repository synthesis on workload identity and attribution