News

Your Copilot agent options for Microsoft 365: Extend or Build?

In today’s AI-driven world, businesses are increasingly looking for smarter ways to integrate artificial intelligence (AI) into their workflows. A critical decision they face is determining which AI approach will best enhance their copilot experiences: to Extend existing systems or Build custom solutions from scratch. The image above illustrates these two distinct paths and how they align with different use cases and capabilities in Microsoft 365 environments.

The Extend Path (Declarative Agent)

The Extend path is for those who want to leverage pre-existing AI capabilities already built into Microsoft 365. This path is ideal for organizations that prioritize ease of integration and access to Microsoft's powerful ecosystems without building custom AI from the ground up.

Here’s what this path offers:

  • LLM Built into Microsoft 365: Microsoft’s large language models (LLM) are pre-integrated into its products, giving you advanced language processing tools right out of the box.
  • Orchestration Built into Microsoft 365: Coordination between different systems and services is handled seamlessly with Microsoft's built-in orchestration capabilities.
  • Rich Integration: Leverage native Microsoft 365 integrations with Word, Excel, Teams, and more, ensuring smooth communication across platforms.
  • Graph Connectors, SharePoint, and OneDrive: Tap into knowledge repositories through Microsoft’s Graph connectors, SharePoint, and OneDrive for an efficient flow of information.
  • Licensing via Microsoft Copilot: Simplify payment and usage through Microsoft 365 Copilot licensing, avoiding the need for custom AI contracts or specialized support.

For companies seeking a plug-and-play AI solution that builds on the infrastructure they already have, Extend path is a logical choice.

The Build Path (Custom Engine Agent)

On the other hand, if your business requires more tailored AI capabilities, the Build path allows you to create Custom Engine Agents using LLM/SLM (large or small language models) tailored specifically to your unique needs.

The Build path enables:

  • Custom AI Models (LLM/SLM): Develop AI models designed specifically for your organization’s unique tasks and challenges. This approach is perfect for companies with complex workflows or niche requirements.
  • Custom Orchestration Planners: Create planners that orchestrate AI agents in more specialized ways than what's available out-of-the-box in Microsoft 365.
  • Teams AI Library UI Enhancements: Implement enhanced user experiences, including features such as:
    • Feedback Loop: Improve your AI’s performance through continuous feedback from users.
    • Generated by AI Labels: Clearly mark AI-generated content to ensure transparency.
    • Sensitivity Labels: Protect sensitive information with AI-driven sensitivity checks.
    • Customized Citations: Provide citations tailored to your specific knowledge sources, improving trust and accountability.

For organizations looking for flexibility and customization, Build path offers the tools to develop bespoke AI solutions that can scale and evolve with your business.

Choosing the Right Path: Extend or Build?

Both approaches have their merits, depending on your organization’s current needs, resources, and long-term vision.

  • If you’re looking for a streamlined, lower-effort solution that integrates well with Microsoft 365’s existing ecosystem, the Extend approach may be your best option. This path allows for quicker deployment, minimal customization effort, and takes advantage of Microsoft’s robust AI offerings.
  • On the flip side, if your organization has specialized workflows, requires more control over AI’s behavior, or seeks to push the boundaries of what’s possible with custom AI solutions, then Building your own custom agents offers you greater flexibility and deeper integration potential.

Conclusion

The decision between extending existing AI capabilities or building custom solutions is a critical one for organizations aiming to enhance their copilot experiences. Whether you choose to Extend and harness Microsoft 365’s built-in power or Build a tailored AI that fits your specific needs, the key is aligning your choice with your business goals. The future of copilot experience is flexible—choose the path that suits you best.