From Data Rich to Insight Rich: A Cloud Charter for 2025

In 2022, the technology industry was heading straight into a wall. Enterprises were collecting everything but understanding little. Petabytes of logs were shipped off to data lakes, and teams spun up pipelines without ever asking the hard questions, like: What signal are we chasing? What outcomes are we enabling?

The problem wasn’t storage. It wasn’t compute. It was direction. We were stockpiling data like it was strategy.

To address the challenges we were facing, I wrote Trolling the Data Rich and published it on Dev.to.

Fast forward to today. There’s been real progress. Cloud-native platforms are more automated. FinOps has earned a seat at the executive table. Generative AI is everywhere. But some habits are hard to break. Too many organizations are still chasing dashboards after the fact. Data is still treated like an asset to be hoarded instead of activated.

The real opportunity isn’t just more data — it’s better thinking. And that means stepping back, sharpening intent and building smarter. Here’s what that shift looks like now and what will matter in the next two to five years.

A different kind of velocity

Flexera’s 2025 State of the Cloud report states that 80% of enterprises are already using or piloting AI workloads. Further, 72% are engaging with generative AI — a jump from under 50% the year prior.

These aren’t hobby experiments. This is the new normal. And it’s forcing a new kind of velocity: not just speed, but precision.

The shift underway now is no longer about moving fast and breaking things. It’s about scaling fast and building things that last.

Three priorities for data leaders right now

1. AI governance becomes real — and visible.
Governance can’t live in the background anymore. As AI systems make higher-impact decisions — from healthcare risk models to financial forecasting — we need guardrails that are transparent, testable and auditable.

Here’s what I see: technical teams are starting to treat governance like performance engineering. Not as an afterthought. A design constraint. A core layer of the architecture.

This shift is already playing out in regulated industries. But it won’t stop there. Governance frameworks will become mandatory across sectors. And the companies that treat this proactively — versus reactively — will win.

2. Privacy-first analysis becomes mainstream.
Secure multiparty computation, federated learning and anonymized data pipelines aren’t experimental anymore. These techniques are unlocking collaboration across partners — without exposing sensitive raw inputs.

This is going to be table stakes in any ecosystem with shared data, including finance, health, retail and energy. If you can’t co-analyze data without leaking it, you’ll be left behind.

3. Policy must follow the data — even to the edge.

As compute shifts to on-premises and edge environments, policies can’t live in PDF binders or static slide decks. They must travel with workloads — written as code, versioned like any other dependency and enforced automatically. Retail chains are already embedding policies into edge containers to govern customer data capture in-store — helping to ensure that privacy and compliance aren’t lost in transit.

This is where we see policy-as-code moving from a DevOps niche to a business necessity. And it’s one of the most powerful unlocks for AI at the edge — from retail stores to factories to distributed fleets.

The shift that’s coming next

Let me be blunt — AI governance is about to be regulated. Not just in the EU or inside compliance teams, but globally. And when that moment hits, companies that treat governance like technical debt will be on the defensive.

This isn’t about fear. It’s about confidence. AI systems don’t just need to be fast — they need to be trusted. This means building governance into systems from day one versus hoping to bolt it on later.

AI-ready enterprises — the ones we help create at Rackspace Technology through the Foundry for AI by Rackspace (FAIR™) — don’t wait for the audit to get it right. They build transparency, trust and accountability directly into the architecture. Governance is treated like observability, like security and like performance — because it is.

And that’s the pivot we all need to make. Governance isn’t the cost of AI. It’s the system that lets AI work — and last.

Final thought

We don’t need more frameworks. We need more teams that treat AI like a product and governance like a core feature.

Done right, governance isn’t friction. It’s how you move faster with clarity. It’s how you scale safely with trust. And it’s how you ensure your decisions and your models are something you can stand behind.

The winners in this next chapter of enterprise AI won’t be the ones with the biggest models. They’ll be the ones who know how to steer them.

And for that, governance isn’t a checkbox. It’s the compass.