Around the Agency – May 2026 Edition

An inside look at our latest explorations and undertakings.

Beth spent some time this week at the Nodes AI conference, where a lot of very smart people are trying to solve the same problem: how to make AI actually understand what it’s working with. Right now, most of the focus is on context: connecting data, decisions, and systems into structured “knowledge graphs” so AI can pull from something more meaningful than the internet’s greatest hits. That’s important. It’s also happening fast.

But here’s the catch: even with better context, AI still tends to default to safe, generic thinking. It knows more but it doesn’t necessarily think better. And that’s where things start to get interesting.

What we’re working on goes a step further, not just giving AI better information, but shaping how it interprets and applies that information. Especially when it comes to brand, messaging, and decision-making, where “technically correct” is not the same as actually good.

Because if all AI does is make it easier to produce more average content faster, we’ve just scaled the problem. We’re more interested in making it harder to be generic. More to come as we keep building.