If you’ve been anywhere near a data team, you already know that there’s an existential crisis going on right now. Here are just a few questions that data leaders and our partners have shared with us:
- Why does data management still feel like a chore?
- Can AI fix it, or is it making it worse?
- How do we move from governance as a hindrance to governance as an enabler?
These were the big questions tackled in this year’s Big Data Debate, where an executive panel of data and AI leaders took a deep dive into how governance must evolve.
Meet the experts
This discussion brought together industry leaders with deep expertise in data management, automation and artificial intelligence:
Tiankai Fengdirector of data strategy and artificial intelligence at ThoughtWorks, advocates human-centered management and explores this philosophy in his book Humanizing data strategy.
Sunil Soaresfounder and CEO of Your Data Connect, specializes in AI governance and compliance, and navigates the challenges of large language models in modern data strategies.
Sonali BhavsarGlobal Data & Management Lead at Accenture, drives governance strategies for enterprise AI and emphasizes the importance of onboarding governance from the ground up.
Bojan CiricA Technology Fellow at Deloitte, he focuses on driving automation in highly regulated industries, particularly financial services and AI-driven transformation.
Brian Amestransformation and enablement leader at General Motors, ensures data trust as GM evolves into an AI-powered, software-driven company.
The three biggest problems with data management – and how to solve them
If anything is clear, it is that governance is at a crossroads. The old way—heavy documentation, rigid policies, and reactive fixes—simply doesn’t work in an AI-driven world. Organizations struggle to keep pace and governance teams are often seen as obstacles rather than enablers.
But why does governance keep failing? And more importantly, how do we fix it? Panelists focused on three main issues – and the practical steps organizations need to take to get management right.
1. Data Governance is always an afterthought
“Governance usually only becomes important when it’s a little late. Something breaks, the data is bad, and suddenly everyone realizes, ‘Oh, we should have done this with governance’.” – Tiankai Feng
Let’s be honest: nobody cares about governance until something breaks. It’s the kind of thing that gets ignored—until a bad decision, compliance failure, or AI disaster forces management to pay attention.
This reactive approach is a losing game. When managing things is tackled as a last-minute fix, the damage is already done. The challenge is therefore to move management from an afterthought to an integral part of the functioning of organizations.
How to make management proactive, not reactive
- Make governance an enabler, not a cleaning crew. Instead of reacting to problems, management should be built into processes from the start. Brian Ames explained how GM is reframing management as “consume with confidence” rather than imposing top-down rules. Target? Ensuring teams can trust the data they rely on.
- Start small and win early. Instead of implementing governance across the entire organization, focus on a single, highly visible problem where governance can bring immediate value. As Tiankai said, “Data management takes time, but management expects immediate results. You need to demonstrate impact quickly.”
- Connect governance to business results. If governance is only about compliance, it will always be underfunded and deprioritised. Sunil Soares explained that successful governance programs are directly tied to revenue, risk reduction or cost savings. If the administration doesn’t make or save money, no one will care.
2. AI exposes – and amplifies – bad governance
“Governing AI is exponentially harder than managing data. Not only do you need good data, but now you have to navigate regulations, explainability and the risks of automation.” – Sunil Soares
The moment AI entered the chat, driving became even more difficult. AI models don’t just exploit data – they amplify its flaws. If your data is biased, incomplete, or missing a line, AI will magnify these problems and make unreliable decisions at scale.
AI governance is not just about ensuring quality data – it is also about managing entirely new risks:
- Data distortion: AI models make bad decisions when trained on bad data. If your data has dead spots, so will your AI.
- Lack of explainability: Many AI models act like “black boxes”, making it impossible to understand why they make certain predictions or recommendations.
- Automated chaos: AI agents now make decisions autonomously, sometimes without human supervision. As Sunil warned, “Regulations still talk about the ‘human in the loop,’ but AI agents are actively working to remove humans from the loop.”
How to control artificial intelligence before it controls you
- Take a proactive approach to managing AI. Management teams must anticipate risks rather than trying to fix them after an AI-driven failure. This means aligning AI governance principles with existing regulatory frameworks and internal risk management strategies.
- Automate management wherever possible. AI can actually help fix governance by automatically documenting metadata, pedigrees, and policies. “If the driving is too manual, people won’t do it,” noted Bojan Ciric. “Automating metadata generation and anomaly detection saves time and makes management sustainable.”
- Define AI railings before you need them. Organizations must create clear policies describing what AI can and cannot do. This includes monitoring AI-driven decisions, enforcing retention policies, and ensuring accurate and explainable AI outputs. Brian Ames described GM’s approach: “We need to define what our AI ‘voice’ can and cannot say. What is its kindness metric? What things must it never do? Management must ensure that the AI is aligned with the company’s brand and values.”
3. Nobody wants to “do” government – so make it invisible
“If you lead with the word ‘governance,’ you’re going to run into resistance. The history of governance is that it’s painful, bureaucratic, and frustrating. We need to reframe it as something that enables people, not slows them down.” – Brian Ames
No one wants to be a data manager if it means spending half the time documenting rules in Excel. The biggest reason for governance failure? It’s too manual, too slow, and too disconnected from the tools people actually use.
The reality is that governance cannot rely on manual processes. People don’t want to fill out spreadsheets or sit on governance forums feeling disconnected from their day-to-day work.
How to build governance that works without anyone noticing
- Let the administration run in the background. Management should be automatic – things like pedigree tracking, metadata collection and policy enforcement should be built into workflows, requiring no extra effort.
- Bring government to where people already work. Instead of having teams log into a separate management platform, integrate management into the tools they already use – Slack, BI platforms, engineering workflows. If the administration is not incorporated, it will not be accepted.
- Use AI to take the burden off people. AI can generate metadata, detect anomalies, and automate compliance tasks so humans don’t have to. As Sunil said, “People don’t want to manage things by hand anymore – they expect artificial intelligence to do it for them.”
Final takeaways: How to make governance really work
Governance is at a turning point. As AI transforms the way organizations use data, the old ways—manual, rigid, and siloed—will not survive. The Great Data 2025 Debate has made one thing clear: good governance is not only necessary – it is a competitive advantage.
The key to making it work?
- Integrate management into daily work processes. Governance cannot be a stand-alone process – it must be woven into the tools people already use, with automated compliance processing, line tracking and policy enforcement in the background.
- Let AI rule AI. As AI adoption grows, it will take on a greater role in monitoring policies, detecting breaches, and providing transparency – reducing the burden on data teams while preventing AI from making uncontrolled and important decisions.
- Connect governance with measurable business impact. Rather than being viewed as a cost, management will be judged on its ability to protect revenue, improve efficiency and ensure AI reliability. Organizations that demonstrate governance delivers financial value gain management buy-in, while others struggle to secure buy-in.
- Invest in AI management – now. Companies that fall behind will face increasing risks – regulatory, reputational and operational. As Brian Ames said: “Managing AI is not optional – it’s the foundation of everything we do going forward.”
The future of governance is not just about compliance – it’s about scaling AI responsibly and unlocking the full potential of data.
Are you ready to build AI-ready governance?
Atlan makes driving seamless, automated and built for the AI era. Book a demo today to see how Atlan can help your organization scale frictionless management.