Operational File Data Platform for the AI Era

Nasuni is the unstructured data foundation for enterprise teams and AI, powering the operational file layer where work happens. One platform manages, protects, and activates enterprise file data: it shows what is inside the files across the environment, governs what is sensitive, gets files to people wherever they work at local-disk speed, and gives the AI platforms a business already uses permission-aware access to the live files.

They are the same platform described at two levels. The Nasuni File Data Platform page covers the architecture: UniFS, edge caching, a single global namespace, and versioning and recovery built into cloud object storage rather than bought separately. This page covers what now runs on that foundation — Resilio for high-performance distributed access, DryvIQ for content intelligence and governance, and Nasuni AI Activate for governed AI access — organized as manage, protect, activate. Nothing was replaced. The architecture is what makes the rest possible.

One foundation and three products. UniFS provides the operational file data foundation. Resilio extends high-performance access to distributed teams and the enterprise edge. DryvIQ adds cross-environment content intelligence, classification, and governance across more than 40 repositories. Nasuni AI Activate makes the file estate available to the AI platforms and agents a business already chose, as one governed namespace, with existing permissions enforced before any query returns.

• Manage is understanding and simplifying: knowing what an organization has, what is inside the data, what is valuable and what is risky, across one global namespace instead of separate systems at every site.
• Protect is governing and recovering: continuous governance of sensitive data, immutable file history, and rapid ransomware recovery.
• Activate is making the data usable: high-performance access for teams wherever they work, and current, governed context for the AI that supports them.

None of those on its own. Nasuni consolidates storage, backup, and disaster recovery into one cloud-native platform, so it replaces the separate systems most enterprises run for each, and capacity grows in object storage rather than on an array. It is not an AI model or an AI application: it is the file data foundation that the AI platforms a business already uses connect to, with existing permissions enforced before any query returns.

As Nasuni Chief Product Officer Nick Burling described in August 2026, with UniFS as the foundation, no other solution manages, moves, classifies, governs, and activates enterprise file data in a single platform. In practice it means no second copy of the estate, no parallel permission model, and no migration project standing between an enterprise and the value of the data it already owns.

No. File data stays in the organization’s own storage, cloud account, and region. Nasuni AI Activate maintains a purpose-built index for AI, based on the contents of the live files, so there is no second copy of the estate inside a third-party AI or analytics platform and no second permission model to maintain.

The platform stores file data in the object storage an organization chooses, including AWS, Azure, Google Cloud, and Oracle Cloud Infrastructure, inside a single unified namespace. At general availability, Nasuni AI Activate works with AWS- and Azure-backed volumes.

Google Cloud, Oracle Cloud Infrastructure, on-premises, and other object stores are currently out of scope for AI Activate.

UniFS is the global file system underneath the Nasuni platform. Its patented architecture fuses cloud object storage with enterprise file services — permissions, versioning, and a single global namespace — so those are properties of the architecture rather than features added on top. Resilio, DryvIQ, and Nasuni AI Activate all run against that same foundation.

Through Nasuni AI Activate, which connects AI applications and agents to the file estate over the open Model Context Protocol. One query reaches the activated estate rather than one site at a time, existing permissions are checked before anything is returned, and the AI works from the live file rather than a disconnected copy. Volumes are activated for AI individually and answer as one estate once activated.

Nasuni AI Activate is built on the open Model Context Protocol and is designed to work across Anthropic, OpenAI, Google, and Microsoft, and with whatever comes next. It connects the AI applications a business has already licensed rather than requiring it to adopt a new one, so there is no lock-in to a single AI vendor. Nasuni has verified and tested multiple AI Apps, including Anthropic Claude, OpenAI ChatGPT, Google Gemini, Amazon Q, as well as Microsoft Copilot and Copilot Studio.

Permissions are checked before the model opens the file, using the same directory permissions already in place. If a person could not open a file, the AI cannot see it either. There are no privileged service accounts, so an agent’s view is identical to the user’s view, and permission changes are reflected in query results in near real time.

No. Nothing is ever used to train any model. The data stays in the organization’s own storage, cloud account, region, and AI app, and is not duplicated into a separate AI or analytics platform.

Not their contents. Standard document formats are activated (content-indexed). Other files in unsupported formats, including CAD and BIM, are indexed by name and metadata, so AI can find them and reason across the documents that surround them — proposals, specifications, correspondence, project records — but it does not read what is inside the file itself.

Not today. Every version is preserved and nothing is overwritten, so earlier states remain available to people and to recovery workflows. Nasuni AI Activate answers from the current version of a file. Reasoning across historical versions is not a current capability.

No. Nasuni AI Activate serves governed AI access to the operational file estate running on Nasuni. DryvIQ is the product that analyzes content across more than 40 repositories, and it does a different job: it tells an organization what is in the wider environment, including where sensitive content sits, without anything having to move.

AI Activate is retrieval-based, not a database or reporting tool, so it does not guarantee complete lists, exhaustive counts, or authoritative numbers the way a database query would. For anything that must be complete, such as a compliance count, use File IQ or verify against the source file or system directly.

Yes, and it should be expected, as each AI application, model, and version synthesizes its own answer. AI Activate returns the same structured, permission-checked document chunks to every AI application, but each one performs its own reasoning and writes its own answer from those chunks, the same way two people reading the same report might summarize it differently.

DryvIQ scans content across more than 40 repositories without moving any of it, detecting PII, PHI and PCI across more than 175 languages, with coverage of 550+ file formats, 100+ entities, and 1,000+ business document types. Policy rules then quarantine high-risk content, apply retention and sensitivity labels, remediate permissions, and support defensible deletion, mappable to GDPR, HIPAA and PCI-DSS.

PSYCHIC is a vendor-neutral AI data readiness framework developed by Nasuni that defines seven properties enterprise file data needs before AI can use it: Proprietary, Secure, Yours, Current, Hybrid, Indexed, and Continuous. As it is vendor-neutral, any enterprise can apply it to any vendor, including Nasuni. It names what to look for in AI data context solutions, rather than what one company provides.

Resilio enables distributed teams with a real file-system mount in Explorer or Finder at local-disk speed, from anywhere, with no VPN and no new hardware. It delivers LAN-speed access at up to 200 Gbps and supports any-size shares, thousands of identity-provisioned users, and global file locking on Windows and Mac. It deploys on the storage an organization already owns, including Nasuni.

DryvIQ adds cross-environment content intelligence, classification, and governance across more than 40 repositories, so it can report on environments with or without Nasuni. Deeper integration with the Nasuni platform and its components is on the roadmap.

• Resilio’s latest release has been generally available since July 31, 2026.
• DryvIQ has been available since it was announced on August 27, 2026.
• Nasuni AI Activate has been generally available since September 30, 2026.

No change to the estate is required. The consolidation, governance and versioning already running are exactly what make AI app access possible, which is why customers who deployed Nasuni years ago can activate for AI quickly, in weeks, without data projects first. Volumes can be activated for AI individually, and folders can be added and removed from the scope, so an organization can start with the data that matters most.