Culture everywhere you work: what culture means in the age of AI

When I started studying culture 20 years ago, it was progressive to try to measure and make sense of workplace culture at all. Culture was a thing you talked about at an all-hands or tried to inspire with posters on the wall. It wasn't a thing you managed.

Now, every leader knows that “culture eats strategy for breakfast,” but every culture is undergoing a major upheaval at the same time that our mechanisms for understanding and directing it are transforming. 

AI has plenty of intelligence and almost no workplace culture context

Something interesting happened over the last two years. AI got very good at reasoning and stayed very bad at knowing anything about your company and the people within it.

Ask any of the tools your leaders now use every day what to do about a struggling team, and you will get a competent, confident, generic answer. But it doesn’t know that their manager is new, that their engagement scores dropped two points on the questions that matter most, or that the last three people who left the group all cited the same reason on the way out.

That gap isn’t a small technical inconvenience. A leader making a people decision with a smart tool and no context is still guessing, only faster and with more conviction.

Most of the industry has focused on getting data into AI. We think that’s the easy half. The harder part is ensuring that data is relevant, validated, and grounded in the science of how people work.

Culture Amp is everywhere you already work

Leader and manager development is the top CHRO priority for 2026, according to Evanta's annual CHRO survey report. Yet the insight those leaders need mostly sits in an annual survey, a review cycle, or a dashboard somebody opens once a quarter. 

So we're taking the always-on intelligence layer we have spent fifteen years building, and putting it where leaders already work.

Using Model Context Protocol (MCP) connectivity, an open standard for connecting systems to AI securely, Culture Amp will be reachable from ChatGPT, Claude, and other AI workspaces. No new tab. No new login. A leader can ask for their team's engagement trends, pull the thread from their last 1-on-1, review goals and feedback ahead of a performance conversation, or combine what Culture Amp knows with what the rest of their stack knows.

Two things make that different from piping a dataset into a chat window.

The first is what sits underneath. Every answer is grounded in fifteen years of People Science research and 1.6 billion data points. It’s the reason we can tell a leader what a two-point drop actually means, and which of the four obvious responses tends to work.

The second is what stays put. Our benchmark data and People Science models remain inside Culture Amp. We share the answer and the workflow, never the underlying confidential data. Trust runs in both directions or it doesn’t run at all.

The part that hasn’t changed

Culture is still your operating system, the connective layer that decides whether your strategy survives contact with reality.

What’s changed is that leaders no longer have to go somewhere to understand it. AI made a lot of things faster. Our job is to make sure the fast things are also the right things, and that requires context nobody can synthesize from the open internet.

That context is what Culture Amp has been building since 2010. Now it goes everywhere you do.

And now it’s ready. You can register here.