Enterprise data teams are under pressure to move faster, govern better, and prepare trusted data for AI without adding more manual work.
That is exactly where the latest CluedIn release makes a difference.
CluedIn 2026.02 is not just a collection of technical updates. It is a significant step forward in making Agentic Master Data Management more reliable, scalable, controllable, and practical for real enterprise data environments.
Improved ingestion, mapping, dataset processing, delete workflows, stuck submission recovery, health monitoring, and worker resilience.
Better Azure AI Foundry support, Claude support through Azure AI Foundry, richer Copilot metadata, improved search, and natural language rule updates.
New survivorship actions, auto-incrementing values, unique value collection from version history, and more flexible rule logic.
Custom enricher icons, user-defined schedules, manual cache control, preview improvements, and better enricher management.
Tag monitoring folders, better audit logs, improved manual data entry, refresh actions, and Delete Workflow moving from alpha to beta.
The most important update in this release is also the least flashy: CluedIn has made a significant investment in platform stability, resilience, and performance.
That matters because enterprise data management is not a demo-room problem. It has to run across large data volumes, complex mappings, multiple data sources, background processes, enrichment jobs, workflows, and downstream consumers.
CluedIn 2026.02 improves the reliability of ingestion, mapping, dataset processing, delete workflows, background workers, and submission handling. It also introduces better recovery for stuck submissions, lower memory consumption during ingestion and data source processing, and stronger health monitoring across key platform services.
Why its Important
AI agents cannot safely manage enterprise data if the underlying platform is brittle. They need a stable operational layer that can process, monitor, recover, and scale.
CluedIn’s AI and Copilot capabilities also take a clear step forward.
The release adds broader support for Azure AI Foundry deployments, including support across multiple base URL formats. Claude models hosted through Azure AI Foundry can now be automatically detected and routed correctly, reducing configuration effort and improving model flexibility for customers using Microsoft-aligned AI infrastructure.
Copilot also becomes more useful in day-to-day data management work. List skills now return richer metadata across cleaning projects, datasets, search results, streams, vocabularies, and vocabulary keys. Search skills now include a lightweight Simple Search option and paginated results, which makes it easier to navigate larger result sets.
Copilot can also list, view, and update existing processing rules, allowing users to modify previously created rules through natural language.
The point is not “more AI features”.
The point is that CluedIn is making AI more operational.
Instead of using AI only to generate text or answer questions, CluedIn is embedding AI into the actual work of data management: searching, understanding, updating, governing, and improving data.
Golden records only create value when teams can control how data is selected, trusted, merged, and maintained.
CluedIn 2026.02 introduces new survivorship and rule capabilities that give data teams more precision over those decisions.
The release adds Min/Max survivorship actions, which can be useful when selecting the highest or lowest value from available records. It also introduces auto-incrementing rule actions, allowing CluedIn to generate unique values for records within defined boundaries.
Golden record rules can now collect unique values from version history and populate another vocabulary key using a configurable delimiter.
Improved PowerFx support, additional comparison operators, and more flexible mapping-chain actions.
Teams can reduce manual intervention while maintaining clear governance over how data changes.
Enrichment is one of the areas where data teams can quickly create business value, but it becomes difficult to manage when enrichment processes multiply across domains, systems, and use cases.
CluedIn 2026.02 makes enrichment easier to organise and operate.
Users can now upload custom icons for enrichers, making them easier to distinguish visually across the platform. Enricher runs can also be scheduled using user-defined schedules, allowing enrichment jobs to run automatically at the right time for the organisation’s workflow.
The release also improves manual enrichment control by allowing entity page enrichment to use or bypass the enricher cache when triggered manually.
CluedIn 2026.02 also improves the experience for users managing larger operational environments.
Tag Monitoring now supports folders, allowing teams to group monitored tags into a logical structure. This helps users keep monitoring configurations organised as the number of monitored tags grows.
Refresh actions have also been added across many list views, including properties, audit logs, stream logs, relationships, golden record lists, search, and preview views.
These are exactly the kinds of changes that reduce friction for stewards, data owners, and technical teams using the platform every day.
Governance depends on evidence. If data changes, teams need to know what changed, who changed it, when it changed, and why.
This release includes improvements to workflow audit logs, making user attribution more accurate. It also includes ongoing improvements to Delete Workflow, which has now moved from alpha to beta in the UI layer.
For enterprise buyers, this matters because data management is not just about fixing records. It is about fixing records safely.
The most important thing about CluedIn 2026.02 is not any single feature. It is the direction of travel.
CluedIn is becoming more resilient at the platform level, more useful through AI and Copilot, more flexible in rules and survivorship, more manageable in enrichment, and easier to operate through improved monitoring, workflows, and UI reliability.
Most enterprises are not struggling because they lack data. They are struggling because their data is fragmented, duplicated, inconsistent, poorly governed, and too dependent on manual effort to keep clean.
AI makes that problem more urgent. If the data foundation is weak, analytics, copilots, automation, and AI agents inherit the same quality issues. Faster technology simply exposes bad data faster.
CluedIn’s role is to help enterprises move from reactive data quality work to continuous, governed data operations.
Read the full CluedIn 2026.02.00 release notes
See how CluedIn helps enterprise data teams create trusted, governed, AI-ready data across complex systems, without relying on endless manual stewardship.
Book a discovery call