Panzura Nexus and the Copilot AI Data Gap: Why Experts Like 451 Research by S&P Global Are Paying Attention

The File Data Microsoft 365 Copilot Can’t See Is the Data Enterprises Most Need It to Know – Panzura Nexus Fixes That, and the Broader Industry Has Taken Note

Key Takeaways:

• Microsoft 365 Copilot only sees SharePoint and OneDrive, leaving the file systems where most corporate knowledge actually lives (project files, contracts, drawings, design history) invisible to it. That gap is the single biggest reason Copilot deployments under-deliver.
• Panzura Nexus is a permission-preserving retrieval layer that surfaces CloudFS file content into Copilot, with no migration, no data lakes, and no copies to secure. Original file permissions follow through to every query and revoked access takes effect in near real time, which file system connectors cannot match.
• Independent coverage is building, with Henry Baltazar of 451 Research by S&P Global publishing coverage of Panzura Nexus, and early AEC use cases (precedent search, proposal response, field operations, succession planning) showing the payoff as won proposals, better-staffed projects, and time returned to skilled people.

Panzura Nexus is here, and it answers a question enterprises have been asking with growing frustration. They want to know why Microsoft 365 Copilot doesn’t know what their files know. The answer is deceptively simple, and the solution is now at hand.

Organizations have spent heavily on Copilot expecting it to draw on everything they know. The promise was an assistant fluent in the full body of company knowledge. The reality has been narrower. Copilot indexes SharePoint and OneDrive, for example. It does not see the file systems where most corporate knowledge actually lives including the project files, contracts, drawings, and design history that accumulates over years of work. The result is an AI assistant answering confidently from a fraction of the picture, while the most valuable institutional knowledge in the organization stays invisible to the tool that was supposed to unlock it.

That gap is the single biggest reason Copilot deployments fall short of the return that was expected and needed. The licenses are bought, and the tool is rolled out, but the answers come back thin because the system is reasoning over a sliver of the data repository itself. The knowledge people rely on to do their jobs is sitting on file shares to which the AI isn’t connected.

Panzura Nexus closes the gap, and we are not the only ones who see the size of this problem. Industry analysts who track enterprise data and AI have begun examining Panzura Nexus and what it means for organizations trying to get real value out of Copilot.

Among them is Henry Baltazar of 451 Research by S&P Global, who has published coverage of Panzura Nexus in his recent report, Panzura targets Copilot gap with file-to-AI retrieval layer Nexus¹. 451 Research by S&P Global subscribers can access the full report here,

We see thoughtful independent analysis as useful in understanding how Panzura Nexus answers questions our customers are genuinely wrestling with. It helps organizations more clearly see the requirements before they acquire solutions.

What Panzura Nexus does and why it matters

Panzura Nexus is a permission-preserving retrieval layer that makes file data visible to Copilot without moving it. It requires no data lakes and no complex, expensive consulting projects.

Here’s how it works. With the initial integration, Panzura Nexus surfaces Panzura CloudFS file content into Microsoft 365 Copilot through Microsoft’s Graph connector framework. There is no migration to SharePoint or OneDrive or copying terabytes of project history into a new repository.

Table 1. Approaches to bringing file data to Microsott 365 Copilot  

There are also no new applications for anyone to learn. People ask Copilot natural-language questions and have intelligent conversations with their own file data inside the interface they already have open. Copilot answers from the files. The difference is that now it is reasoning over the real corpus of company work, not just the documents that happened to be saved in the right place.

The approach matters as much as the outcome. Migration-based alternatives ask an organization to relocate its file estate before AI can touch it, a project that can take a long time, disrupt established workflows, and create a second copy of sensitive data to secure and govern. Panzura Nexus leaves the data where it is. The files stay in place, under the same controls, in the same systems, and Panzura Nexus does the work of making them legible to Copilot.

Trust is the whole game

The part that matters most for regulated work is permission fidelity, and this cannot be understated. When file content is indexed for AI, a hard question follows as to exactly who is allowed to see what comes back. An AI assistant that surfaces a sentence, or location or insights from a document that was not meant to be seen or known by a particular user has created a breach or at least an unacceptable risk.

Panzura Nexus preserves the original file permissions all the way through to the Copilot query. It translates the access controls that already govern the file estate into the identity claims Microsoft Entra enforces on every request, so a person who cannot open a file cannot surface knowledge derived from it. Access is checked at query time and never assumed. And when permissions change, the change follows the user in near real time rather than waiting for a periodic crawl to catch up – this is a clear and crucial differentiator against the vast majority of Graph connectors. With Panzura Nexus, a revoked permission means revoked access to context and content almost immediately. Not tomorrow morning.

For industries where migration projects and re-permissioning exercises are non-starters, that distinction is everything. The architecture, engineering and construction (AEC) and legal industries are top of mind in this regard. Financial services, energy, healthcare, and public sector are also organizations where permission integrity is often a condition of doing business. Panzura Nexus was built so that bringing AI to file data does not mean loosening the controls that keep that data safe.

Built for control, not guesswork

Panzura Nexus is designed for administrators who need to know exactly what AI can and cannot reach before anything goes live. Teams scope what gets indexed by share, by path, by file type, by size, and by date range, with fine-grained filtering for the edge cases every real file estate contains. Before activating a policy, they can run it as a “dry run” simulation to preview precisely what would be ingested. That way, there are no surprises and nothing is exposed by accident. A dashboard shows what has been indexed, at what rate, and under which policies, giving teams a clear, auditable view of Copilot’s reach into their file estate. You don’t deploy until you’re completely satisfied with the result.

Licensing follows the same principle of predictability. It is based on capacity rather than on how often people query, so finance teams can plan with confidence and adoption is never penalized. It’s a capacity-based model that reflects each CloudFS deployment. The more an organization uses Panzura Nexus to get value from its file knowledge, the better the value proposition looks because they’re not penalized for using it more over time.

Faster and more intelligent work

The point of all this is not the underlying infrastructure or the implementation details, although those are incredibly important to the IT team responsible for file data and security. Mostly, it comes down to the work and the outcome Panzura Nexus creates in an organization. That’s why we like to say, “Copilot, meet your files” – because that’s how end users understand it and rightly so.

For example, in the AEC industry, where Panzura Nexus began as a pilot program among six CloudFS customers, teams sit on decades of precedent that is nearly impossible to search by hand. Altogether, four initial use cases show what changes when file data becomes answerable.

• Precedent search. Determining whether or not a firm has done a similar project in the past, and what can be learned from it, used to be a long hunt for relevant documents and tribal knowledge often abandoned in favor of redoing work the firm had already done. With Panzura Nexus it becomes a thirty-second conversation with Copilot that returns the relevant past projects, the files that matter, and the people who worked on them.
• Proposal response. Winning language from past bids is scattered across years of project folders, effectively lost the moment a proposal is submitted. Panzura Nexus lets teams pull that language into new RFP responses in seconds rather than days. Proposal effort trends down while win rates trend up.
• Field operations. Job-site staff need answers from spec sheets, drawings, and change orders while they are standing on the site, not after a phone call back to the office and a wait for someone to find the file. Panzura Nexus puts those answers in Copilot on mobile, so the field queries the project record directly.
• Succession planning. Decades of design judgment used to walk out the door when a senior engineer retired. With the firm's accumulated knowledge accessible through Copilot, that experience stays with the organization and stays available to everyone, long after the person who created it has moved on.

Certainly, the use cases are much broader than what is described here, and we talk to customers every day who are discovering applications for Panzura Nexus that surprise and delight us. These are the kinds of outcomes that move the conversation past IT and into the business realm, because they show up as won proposals, better-staffed projects, and time given back to people doing skilled work.

From file platform to AI data management

Panzura Nexus is generally available now for Panzura CloudFS customers. Support for additional file platforms is on the near-term roadmap, and the direction from there runs toward deeper agent capabilities, broader source systems, and workflows tuned to specific industries. Panzura holds Microsoft co-sell partner status, and Panzura Nexus data is built to feed the custom agents organizations are already building in the Microsoft ecosystem, so the investment compounds as those programs mature.

This is a deliberate step. For more than a decade, Panzura has managed the world’s unstructured file data and made it visible, resilient, and instantly available wherever it is needed with one source of truth. Alongside the CloudFS platform capabilities that make file data ready for AI and agentic workflows, Panzura Nexus dramatically extends that same discipline further into the AI paradigm, turning the file estate from a storage and business continuity asset into a secure intelligence asset. It’s the difference between holding an organization’s knowledge and making it smart.

Want to see what Panzura Nexus can do with your own file data?

If you’re already a Panzura customer, reach out to your Customer Success Manager. If you’re new to Panzura, schedule time with a Panzura expert and let us show you what you’re missing.

1. 451 Research by S&P Global: Panzura targets Copilot gap with file-to-AI retrieval layer Nexus, Henry Baltazar, 15 May 2026

451 Research by S&P Global is a trademark or registered trademark of 451 Research by S&P Global. All other product and company names are trademarks or registered trademarks of their respective holders. Use of those names does not imply any affiliation with or endorsement by their owners. The opinions expressed above are solely those of Panzura LLC.

Frequently Asked Questions (FAQ)

• Why can't Microsoft 365 Copilot see all of my company's files?

  Copilot indexes SharePoint and OneDrive, but it does not see the file systems where most corporate knowledge actually lives. Project files, contracts, drawings, and years of design history stay invisible, so Copilot answers from a fraction of what your organization knows.

• How does Panzura Nexus give Microsoft 365 Copilot access to file data without migration?

  Panzura Nexus is a permission-preserving retrieval layer that surfaces Panzura CloudFS file content into Copilot through Microsoft’s Graph connector framework. There is no migration to SharePoint, no data lakes, and no second copy of sensitive data. The files stay in place under the same controls.

• Does Panzura Nexus keep file permissions intact when surfacing data to Microsoft 365 Copilot?

  Yes. Panzura Nexus carries original file permissions all the way through to the Copilot query, translating existing access controls into the identity claims Microsoft Entra enforces. Access is checked at query time, and revoked permissions take effect in near real time rather than after a periodic crawl.

• How is Panzura Nexus different from migration-based approaches to AI file access?

  Migration-based alternatives require relocating your file estate before AI can use it, a long project that disrupts workflows and creates a second copy to govern. Panzura Nexus leaves data in place and makes it legible to Microsoft 365 Copilot, with permission changes that follow the user almost immediately.

• Which industries benefit most from connecting file data to Microsoft 365 Copilot securely?

  Industries where migration and re-permissioning are non-starters benefit most. Architecture, engineering, and construction (AEC) and legal are top of mind, along with financial services, energy, healthcare, and public sector, where permission integrity is often a condition of doing business.

• How can administrators control what Panzura Nexus indexes for Microsoft 365 Copilot?

  Teams scope indexing by share, path, file type, size, and date range, with fine-grained filtering for edge cases. A dry-run simulation previews exactly what would be ingested before anything goes live, and a dashboard shows what has been indexed, at what rate, and under which policies.

• How is Panzura Nexus licensed?

  Panzura Nexus uses a capacity-based licensing model tied to each Panzura CloudFS deployment rather than query volume. Because pricing does not depend on how often people query, finance teams can plan with confidence and adoption is never penalized as usage grows.