If you read my big-picture report from Davos, maybe you’re wondering, “That’s great. Now how do I apply this? What can I be doing to help us better surf this AI wave?”
To answer that question, I’d like to tell you about a side conversation I had with the head of a global powerhouse in the fast-moving consumer goods (FMCG) category. This leader talked about two kinds of warehouses in her business: green and brown.
A "green warehouse" is one that actively incorporates environmentally friendly practices and sustainable technologies to minimize its ecological footprint, while a "brown warehouse" refers to an existing facility that likely needs upgrades to become more efficient or environmentally conscious.
“Green" signifies sustainability, while "brown" represents a more conventional, potentially outdated warehouse approach. Green warehouses are typically designed from the ground up with sustainability in mind, while brown warehouses come from a time where these needs hadn’t yet been acknowledged or envisioned; as such, they often need retrofitting to work for the current age.
As she spoke, a lightbulb clicked on in my mind. It’s not a far step from physical warehouses in the consumer goods space to our own data warehouses, lakes or lakehouses. Swap “sustainability” for “AI readiness”, and you see the parallel. The success of an AI model rises or falls on the organization and cleanliness of the data that feeds it.
In other words, to win with AI, you need to get your data house in good order. You need to ensure your data is clean, well-structured, relevant, accurate, accessible and properly labeled. You need to rectify, standardize, combine and simplify. And you also need a database that’s built with AI in mind. You need a green warehouse, not a brown one.
Yet these days, far too many customers are working with a legacy data environment that’s a patchwork of dated or fit-for-purpose solutions. In fact, the consulting firm KPMG sees so much of that impediment they are pivoting to expand their data advisory services. And of course, removing such impediments is the value proposition of SingleStore: a single solution that can deploy your data with speed, scale and simplicity.
Now another conversation that stayed with me was about how to use AI effectively and responsibly. This year, the talk was less about regulation than about governance. We intuitively know what’s right and wrong in relation to AI, so how do we self-govern until the regulations catch up and take effect?
The answer lies in making AI everyone’s responsibility. Today, IT organizations are being asked to do all the business as usual functions while also building for a new, AI-driven world. That’s too much for anyone to juggle, so business leaders should create dedicated teams that help make AI a company priority — one that extends beyond the boundaries of IT to be strategically embedded in business objectives. In other words, AI is everyone’s domain.
So as we kick off 2025, I encourage you to think about three key questions:
- Is your data estate green or is it brown?
- If it’s more brown, how do you make it more green?
- And do you have a strategy to think beyond traditional IT as you embed AI into the structure of your business?
These are vital questions to ask, because as many have noted, winning with AI is 10% about technology, 20% about implementation and 70% about the strategy. To capitalize on the promise of AI, we need to become much more intentional about the state of our data estates, and think bigger about who “owns” AI.
