News & Blog

Here’s something worth sitting with for a second.

Somewhere in your organization, a manager has a hard conversation coming up. Underperformance, a missed deadline, a promotion they have to explain to the person who didn’t get it. They’ve got less time and less coaching than any generation of managers before them.

So they open a general-purpose AI tool and ask it how to handle the conversation.

This is happening now, in most companies, and largely unsupervised. The folks at The Predictive Index have been tracking it, and they’ve given the failure mode a name that stuck with me: the copy-paste problem.

The issue isn’t that managers are using AI. The issue is what the AI has to work with. A general-purpose model knows nothing about the person on the other side of that conversation. Nothing about how they take criticism, what motivates them, whether silence from them means agreement or shutdown. So it produces something confident, well-structured, and completely generic… the same advice it would give about anyone.

The output looks like leadership. It just isn’t aimed at anyone in particular.

You see the results downstream: feedback that lands wrong, one-on-ones that get skipped because they feel like theater, promotions that miss what somebody actually contributed, etc. PI’s research ties 63% of preventable employee exits back to weak management support. Generic AI doesn’t close that gap. It makes the wrong answer arrive faster and sound better than it actually is.

If your organization isn’t working from behavioral data, this is the gap you’re living with right now, whether or not anyone has named it.

The mainframe moment

I spent a few days last week at our annual summer off-site meeting with PI consultants, PI clients, and the team from PI headquarters. Lance Neuhauser, who took over as CEO of The Predictive Index in March, came down to talk about where the business is headed.

His framing is the most useful thing I’ve heard for what’s actually changing in our world right now.

Think about what happened to data generally over the last thirty years. In the mainframe era, organizations had data. But getting an answer out of it meant filing a request with a specialist, waiting, and receiving a report that was slightly stale by the time it reached your desk. Data was a department. It was something you asked for, and the asking was expensive or time-intensive enough that most people didn’t bother.

Then, along came cloud computing… The cloud didn’t change what the data said. It changed who could reach it and how fast. The answer became a query away, available to anyone with a reason to ask. Data stopped being a department and became a habit.

Behavioral data is having that exact moment.

For seventy years, PI’s science has been validated, rigorous, and — outside of the organizations that committed to it — largely unavailable to the people who could use it. The insight was never the bottleneck. The friction was. The distance between “I have a situation” and “here’s what the behavioral data says about it” was too wide for a manager with twelve other things on their plate.

AI collapses that distance. That’s what I mean when I talk to people about “behavioral intelligence”: it’s not new science, it’s the same validated science we’ve always delivered, but at the exact moment of decision. And we’re doing this now with PI’s built-in assistant, Obi, which is designed to answer the question a leader actually has… “how do I approach this specific conversation with this specific person?”,  in the ninety seconds before they walk into the room. It’s still officially in beta, and as of August it’s available to every new PI client.

The distinction that matters isn’t AI versus no AI. That argument is over. It’s AI grounded in real behavioral data about your actual people versus AI making a confident guess about someone it’s never met.

What this looks like on an ordinary Tuesday

Here’s where I think the next two or three years will go for organizations that move now. Behavioral intelligence stops being an annual workshop and becomes something closer to checking your email.

Before a high-stakes interview. Not just a candidate’s pattern, but the three questions that will reveal whether this person can handle the ambiguity in this particular role — and what a dodge sounds like when it comes.

Before a difficult conversation. A manager who knows, going in, that their direct report needs processing time and will read rapid-fire questions as an interrogation. Same conversation, different outcome, no guesswork.

Standing up a special projects team. Six names and a deadline. Behavioral intelligence surfaces the friction points before they cost you three weeks, and tells you who’s going to need explicit permission to push back.

When a team stalls and nobody can say why. Behavioral data doesn’t diagnose everything. But it answers a question most leaders never think to ask: is this a capability problem or a wiring problem? Those require completely different responses, and guessing wrong is expensive.

None of those are annual events. They’re Tuesday.

Why the order of operations matters

Let me push back on some of the noise in the market, because this is where organizations get hurt.

AI makes behavioral data accessible. It does not make your people capable of using it well.

An assistant will hand a manager a confident, articulate answer whether or not that manager understands what they’re looking at. If your team hasn’t been trained in the fundamentals (what the factors actually measure, why fit is contextual, where the data’s limits are) then all you’ve bought is the ability to move faster in the wrong direction. Speed without judgment isn’t an advantage. It’s exposure.

So the investment that matters right now is upskilling your people.

The organizations that win with this will be the ones that build the foundation first: a trained core who understand the science, leaders who’ve done the work of defining what their culture actually requires, and a clear picture of the business strategy that the people strategy is meant to serve. Then you add the tooling, and the tooling compounds.

Foundation, then acceleration. Reverse the order and you end up with a fast, expensive tool nobody trusts.

That’s also why now is a genuinely good moment to start. Building the foundation takes time you still have. The companies that begin the training work this year will be the ones positioned to use what’s coming next year, while their competitors are still asking a chatbot how to give feedback to a person it’s never heard of.

Where we come in

This is the part of the job I’m most excited about.

PI Midlantic has been doing this work since 1985. Forty-one years, the largest PI partner in the world, with 650-plus clients across the United States and in more than twenty-five other countries. What we’ve never been is a software reseller. Our job is the translation layer: helping leaders understand what their data means, what to do about it, and how to build the habits that make it stick.

That job doesn’t shrink as the technology improves. It gets more important, because the questions leaders can now ask are far more interesting than the ones they could ask before. Not “what’s this person’s pattern” but “what’s the behavioral shape of the leadership bench we’ll need in three years, and who do we already have?” Not “did this candidate score well” but “where are the systematic blind spots in how we’ve been hiring?”

Those are strategy conversations. They need somebody in the room who has had them a few hundred times before.

When people ask what I do for a living, I’ve started telling them I’m a behavioral intelligence consultant. Not because it sounds better, but because it’s a more accurate description of the job than it was two years ago. The work isn’t administering assessments. It’s making sure the right insight reaches the right leader at the moment they can actually use it — and making sure that leader knows what to do with it.

If you’re watching the AI conversation and trying to sort out how much of it is real, this part is real, and it’s closer than you think. I’d welcome the chance to show you what it looks like in practice.

Clayton Ramsey, Talent Optimization Advisor

PI’s September newsletter and the companion ebook, The Copy-Paste Problem, go deeper on the research above. Lance also wrote about the other half of it for Raconteur — why AI makes the human element matter more, not less.

Clayton Ramsey is a behavioral intelligence consultant and Senior Talent Optimization Advisor with PI Midlantic. A Certified PI Master Trainer and Certified Professional Trainer, he has spent more than fifteen years in recruiting, sales, and talent optimization. Reach him at caramsey@pimidlantic.com or (410) 562-0512.