Agentic AI is a rising buzzword, capturing attention of enterprise tech companies, but the definition and execution remains in question. At Qlik Connect this week, the gap between expectation and reality was clearly on display. While vendors push new agent-powered capabilities into data platforms, customers are still trying to figure out what it all means, and how to apply it without disrupting trust, quality, and governance of data.
Enterprises know they need to make better use of their data. They’re investing in data lakes and modern analytics platforms, but these assets are often underutilized or too complex for business users to navigate. AI promises to help, but customers remain unsure how agentic AI fits into their workflow, and they’re wary of relying on tools they don’t fully understand. What they want is clear: Simpler access to insights, more value from existing data assets, and tools that bridge the gap between data teams and decision makers. That’s exactly where agentic AI could help, and why I’m optimistic about the direction we’re heading.
From Interface to Infrastructure: Qlik’s Measured Approach
Qlik’s new Agentic Experience is designed to meet enterprises where they are. Instead of chasing the latest AI trend, Qlik integrates AI agents directly into its analytics pipeline, adding natural language interfaces, automation, and context-aware workflows without sacrificing control. These agents are purpose-built: The Discovery Agent scans for business risks and opportunities; the Pipeline Agent builds data pipelines based on business outcomes; and Qlik Answers enables conversational interaction with structured and unstructured data.
What sets Qlik apart is its deliberate emphasis on governance. The company is cautious with generative models, prioritizing data quality and trust over flashy automation. The customers I spoke with at Qlik Connect shared this concern, noting that while they’re excited about AI’s potential, they’re not ready to hand over critical decisions to autonomous systems. They want AI to make their data lakes more useful, to help ingest and prepare data faster, and to simplify analytics, not to replace human staff.
Qlik’s platform reflects this mindset. The Agentic Experience is powered by the same engine that underlies Qlik Cloud Analytics, now enhanced by the Qlik Open Lakehouse, a fully-managed, high-performance platform built on Apache Iceberg and integrated into the Qlik Talend Cloud. With near real-time ingestion, native support for standards, and interoperability with leading analytics engines such as Snowflake, Spark, Athena, and SageMaker, Qlik is emphasizing openness and flexibility. The company’s collaboration with AWS, including Bedrock for generative AI support, further strengthens the foundation for agentic use cases that are both innovative and enterprise-grade.
Multiple Paths to Agentic AI
Other vendors are converging on the same goal from different angles. I spoke to a few of them in the last week or two, and it was interesting to hear how their perspectives are aligned with the end users I spoke with. Each comes at the agentic AI question from a different angle, yet each is focused on real-world customer solutions.
Starburst, building on its Trino heritage, sees agentic AI as a way to unlock the potential of data lakes. Its agents enhance Iceberg environments with search, classification, and summarization, and its support for air-gapped deployment appeals to customers with strict compliance needs. Starburst also has its eye on emerging frameworks like Google’s Agent2Agent and future MCP (Model Context Protocol) integrations to support multi-agent coordination at scale. Their perspective is different from Qlik’s but the end result is surprisingly similar, and both companies are committed to open interoperability.
CData approaches agentic AI as a connectivity challenge. Its MCP Servers enable secure, SQL-like access to SaaS and NoSQL systems by translating native APIs into a language LLMs can understand. This opens new possibilities for AI-driven applications to query and act on distributed enterprise data without compromising on permission boundaries or data governance. CData’s real value lies in maintaining up-to-date connectors across hundreds of APIs, ensuring stability in dynamic environments. I intend to write more about this solution soon, since it really opens the door to derive value from agentic AI to any data application.
Why Agentic AI Matters Now
Across Qlik Connect and recent briefings, one message is consistent: AI is becoming an essential part of the data pipeline, not just a bolt-on feature. Agentic AI, defined not by autonomy, but by intelligent, context-aware functionality, offers real value by making data more accessible, discoverable, and actionable. According to a recent survey by The Futurum Group, nearly 90% of CIOs consider agent-based AI to be a strategic priority, and these leaders are turning to trusted providers to help make this a reality. Companies like Qlik are wise to meet their customers with ready-to-implement solutions.
But education remains critical. Many customers are still unsure how to evaluate agentic features, or even what the term really means. There’s an industry responsibility here, to move beyond hype and help customers understand how AI can safely enhance analytics, improve data quality and accelerate workflows. Tools that tag, clean and query data using natural language aren’t replacing jobs, they’re expanding the reach of data-driven decisions across every business unit.
The Agents are Coming
Agentic AI isn’t a speculative concept, it’s already being implemented to solve concrete problems. Companies like Qlik, Starburst, and CData are embedding agentic capabilities in ways that align with their customers’ operational realities. From helping business users “talk to their data” to automating the preparation and structuring of that data behind the scenes, agentic AI is driving a quiet but powerful transformation in enterprise analytics.
For enterprises, the question isn’t whether to adopt agentic AI, but how and how soon. The case is real. The technology is ready. Now it’s about bridging the last mile between capability and confidence. For more detail on the new agentic AI features from Qlik, watch their Tech Field Day presentations from last week!
