For most organisations, the main challenge with data is not scarcity but fragmentation. Data is spread across many systems, platforms, clouds and formats. This makes it hard to connect, manage and quickly turn into insights. On a recent episode of Futurise, Suzanne EL-Moursi, co-founder and CEO of BrightHive, discussed why this issue has persisted for decades. She explained how agentic AI—AI systems designed to act independently, making context-aware decisions to manage and connect data—could bring real change rather than just adding another tool.
Suzanne’s perspective comes from real-world enterprise experience. Even before large language models, organisations struggled to bring data together in regulated fields like higher education, healthcare and the public sector. Despite big investments in data platforms, teams still have to move datasets between different systems just to answer basic business questions.
This deeper issue, which is how data workflows are managed across the enterprise, helps explain why automation often falls short and shows why a more holistic approach is needed.
From Tooling to End-to-End Data Orchestration
Data work is one of the most complex tasks in an enterprise. It includes ingestion, cleansing, governance, schema mapping, privacy controls and ongoing updates as new questions come up. Most of this work is still done by people, even though there is a global shortage of data engineers and analysts.
Suzanne estimates that nearly 80 percent of data work is repetitive and takes a huge amount of time. This slows down insights and increases risk, especially with sensitive personal data. As data volumes grow, the gap between the work and the number of skilled workers keeps getting bigger.
BrightHive’s agentic approach takes a different view of the problem. Instead of automating just a few steps, agentic data agents streamline the whole data process. For example, in a healthcare organisation with sensitive patient data, an agentic system can automate finding and standardising different data sources. This helps meet privacy rules while bringing together data from many departments. These agents work under a main agent that understands the context, rules and business goals. They help organisations cut down on manual work, get insights faster and improve data quality. By using knowledge graphs, specialised models, and integrations, agentic AI copies how human teams solve data problems, but does it faster and more reliably.
Serving the 97 Percent
A key point from the discussion is that most organisations limit who can work with data. Usually, only two to three percent of employees have the technical skills for complex data systems. The other 97 percent have to submit tickets, wait their turn, or ask others to get the insights they need.
Agentic data connectivity changes this. It allows business users to interact with data in a conversational way, while agents handle rules, integration and quality in the background. This gives faster access to insights, reduces bottlenecks and extends data skills beyond technical teams. It helps organisations move toward self-service decision-making without losing control or trust. The goal is not to replace data teams, but to lighten their workload so experts can focus on more complex problems.
Rethinking Data Architecture in an Agentic World
This change in architecture is clear as agentic capabilities challenge old ideas about centralised data management.
Suzanne describes a future where humans are not replaced but work with agentic systems. The most successful organisations will rethink their workflows, letting agents handle repetitive data tasks so people can focus on judgement, creativity and responsibility.
In this way, agentic data connectivity is more than just another use of AI. It is a major change in how enterprises manage data. Agentic systems help organisations handle data more responsibly and at a larger scale, and they encourage a culture of data-driven decision-making that values ethics, efficiency and adaptability. Still, this shift comes with possible risks. For example, relying on agentic systems can raise concerns about reliability, data security and bias in automated processes. So, while agentic data connectivity can reshape workflows and make data a strong base for strategic insight and growth, organisations need to implement it carefully and stay alert to its challenges.
