Knowledge Graphs Are Going Mainstream: The New Foundation for AI

There’s been a noticeable shift in the enterprise data world over the past year. You can feel it. SAP now has a Knowledge Graph. Netflix just published a fascinating piece on how they’re using ontologies to “model once and represent everywhere”. ServiceNow acquired Data.World to make their customers' data “AI-ready”. Samsung embedded a knowledge graph directly into their flagship smartphone.

The pattern is clear: large organisations across industries are waking up to the power of knowledge graphs and ontologies - not as a nice-to-have, but as foundational infrastructure for modern AI and analytics.

What’s driving this? A few things stand out:

🔵 AI needs context - Everyone’s experimenting with GenAI. But if you want answers that are accurate, explainable, and grounded in reality - you need a structured, contextual foundation. A knowledge graph gives you that.

🔵 Data silos are everywhere - Most companies are drowning in fragmented schemas and duplicated logic. Knowledge graphs help unify meaning across systems - they give you one model of the business that everything else can plug into.

🔵 Business users want answers, not table joins - Knowledge graphs bring a semantic layer that makes data more usable by everyone - not just engineers. Ask a question in business terms, and let the graph do the hard work underneath.

Those of us who’ve worked with knowledge graphs for years - often in the background - always felt they made sense. Not because it was trendy, but because the logic was solid. Now that AI is pushing context and trust to the forefront, it seems that knowledge graphs are going mainstream.

And this isn’t just a tech giant thing. Pharma companies have been deploying knowledge graphs for years. Large investment banks are building out dedicated graph teams - some quite substantial. Manufacturers are adopting knowledge graphs for digital twins. Legal use cases are emerging where graphs work beautifully in neural-symbolic loops. And there’s widespread interest in retail. Cloud providers are racing to offer graph-native tooling. I’ll stop there - but it really does feel like this is becoming a standard layer in the enterprise stack, just like data warehouses in the 90s or data lakes in the 2010s.

If your organisation hasn’t started thinking about knowledge graphs yet - now is the time. Start small. Pick a domain. Define your core concepts. Link what you already have. You don’t need a large team - just a clear idea of what you want to connect.

There are pitfalls. Plenty of silly mistakes that can trip you up. Knowledge graphs aren’t a silver bullet. The idea is simple, but the devil is in the detail. That said, the foundational principles are intellectually robust. You need to connect your data. You need to organise it by business semantics. And you would be wise to insist on open standards.

Because in a world increasingly shaped by AI, it’s no longer enough to just have data.

You need knowledge.