Informatica Introduces AI Agents but Does Not Become an Agentic AI Platform Yet
Image source: Informatica, 2025
The fall 2026 release of Informatica’s IDMC (Intelligent Data Management Cloud) significantly expands CLAIRE Copilot from a prompt-based helper into a more agentic, goal‑driven AI assistant across data integration, automation, MDM, and data quality management. So, what’s been enhanced?
1. CLAIRE Copilot for Data Integration:
• Natural language–driven pipeline creation: Users can describe ingestion, replication, and integration tasks in plain language, and Copilot generates the mappings and flows.
• Complex expression generation using natural language or pseudo‑code, reducing the need for deep technical skills.
• In‑context transformation generation inside mappings so users don’t have to switch tools or manually wire logic.
• Smarter table selection rules driven by natural language plus user‑provided context (e.g. lists of tables or business intent).
• Auto‑documentation and summaries for integration assets, improving maintainability and collaboration.
2. CLAIRE Copilot NL to XQuery:
• Translates natural language descriptions directly into XQuery code for application integration and process automation.
• Eliminates the need to handwrite transformation logic.
3. CLAIRE GPT: Agentic Data Management Upgrade:
• CLAIRE GPT evolves into an agentic system with:
◦ Planning and reasoning capabilities (not just Q&A).
◦ Improved natural language understanding without requiring product‑specific syntax.
◦ Human‑in‑the‑loop oversight to resolve ambiguity and validate actions.
• Supports discovery, exploration, integration, data quality, and product help as specialized agents, not a single monolithic assistant.
4. Universal CLAIRE Copilot for MDM:
• Natural language access to master data for exploration, understanding, and product help directly inside Informatica MDM.
• Cross‑entity exploration and relationship visualization (including graph views).
• Date and date range queries supported conversationally.
5. CLAIRE Copilot for Data Quality:
• Converts business‑defined data quality standards into executable DQ rules using natural language.
• Helps evaluate, operationalize, and remediate data quality issues faster.
• Avoids duplicate rules using AI‑powered similarity checks.
Our Take
Although Informatica presents the above enhancements as a new platform with AI agents, it is still a far cry from becoming a platform for deploying and managing business process–oriented enterprise AI agents. It is still unclear how Informatica would integrate with SFDC Agentforce – or if it would remain as a separate platform for data management only fulfilling the “AI-ready data” goal. There is the new MDM Extension for Agentforce, but it does not seem to be positioned to become the ontological brain of Agentforce.
Admittedly, the new collection of specialized agentic copilots can:
• Understand business intent.
• Plan and reason through data tasks.
• Execute actions safely with human oversight.
• Connect AI agents to trusted enterprise data.
However, all the new agents are focused on data management roles/tasks only – and many other vendors are providing similar enhancements. (Recently, we wrote about CluedIn’s new offerings of data management agents.)
To become a platform for business process–oriented AI agents, Informatica will have to add, at the minimum:
• A semantic governance framework, including enterprise ontology and knowledge and context graphs.
• A behavioral governance framework that would track AI agents’ behavior against defined policies and rules and initiate interventions when necessary.
The recent MDM & Data Governance Summit in Toronto did not highlight any significant changes in this direction. Hopefully, the forthcoming Informatica World conference will bring more refreshing news.
Want to Know More?
• Microsoft Adopts Ontology-Based IQ Layer for Agentic AI
• CluedIn: Graph-Based MDM With Agentic Data Management
• AI Agent Development Platforms: What They Offer and How to Select the Right One