2026 Trends: Predicting the Future Has Never Been Harder
(So Obviously, Here Are My Predictions)
Welcome to FullStack HR, and an extra welcome to the 48 people who have signed up since last week.If you haven’t yet subscribed, join the 10100+ smart, curious leaders by subscribing here:
The Swedish version of the article is available here.
Happy Thursday,
107 lectures/workshops for 52 organisations in 7 continents. That’s what I’ve done this fall. A bit too much, in all honesty, but it surely has taught me a lot about how to do and not do AI adoption in organisations. 62 of them have been for leaders/ lead teams, 23 for HR teams, and the rest for a broad mix of departments (but mostly finance/legal)
This spring, I will work with fewer organisations but for a more extended period.
But if we’ve worked together this fall - thank you.
If not? Reach out, and we’ll see if we are a great fit for each other. :)
That said, let’s get to it!
The whole “trends and predictions” thing has always felt a bit sketchy to me. There’s something about making bold statements about the future that makes me a bit uncomfortable. Because who can really tell the future? No one.
But screw it. Let’s do this anyway.
And yes, predicting the future has never been harder than right now. And that, in itself, tells us something important.
Let’s Get Some perspective first.
A year ago, the best model we had was GPT-4o. We’d just gotten o1 in September, the first model that could actually reason and think in steps.
That was just over a year ago.
In just fourteen months, the development in AI capabilities has been absolutely insane.
So when I’m sitting here trying to predict what 2026 will look like, I’m acutely aware that I’m probably going to be wrong about half of this. But that’s fine.
(If you want to take a time machine and see very old predictions from me, click here: Social media and HR anno 2013.)
My Three Trends for 2026
I’m not going to rank these. You’ll see why they’re all connected anyway.
1. The Digital Twin (Or Whatever We End Up Calling It)
I think we’re going to see the rise of what I’ll call the “digital twin.” Not in the sense that we’re all getting AI clones that can replace us, that’s not what I mean.
What I mean is this: an external memory that we connect to. A bot that sits in on things.
Most people already have the capability for this. You’ve got it in Teams. You’ve got it in Google Meet. You’ve got it in Zoom. Everywhere we meet, there’s usually some form of transcription option available.
Most people haven’t turned it on yet.
But I think 2026 is the year when transcribing meetings becomes normal. When having an AI assistant listen in becomes as common as taking notes used to be.
And here’s where it gets interesting: we’re not just collecting data anymore. We’re starting to use it.
Right now, it’s all very individual. “Look at me, I have transcripts of all my meetings!” Cool. But you can’t do anything with them? It’s just meetings being transcribed for transcription’s sake.
What’s coming is the ability actually to use that information to make better decisions. To have something that sits in, listens, summarizes, and helps you act on what matters.
That’s what I mean by the digital twin. Not a clone. More like an always-on assistant that knows what you know because it’s been listening to what you’ve been doing.
I believe this will be one of the first truly agentic use cases that actually works. We talk a lot about “agentic AI” but there’s still very little real business being done with agents.
This, I think, is where it starts.
2. Organization Over Technology
This is the shift I’ve been shouting about for what feels like ages (but it’s only three years). But I’m going to keep shouting about it because it matters.
2026 is going to be the year when we talk more about organizational change than we talk about technology.
To be fair, this shift already started in 2025. But it’s going to accelerate.
Don’t get me wrong here, I’ve been saying this since basically the dawn of time (or at least the last three years): AI adoption is not a technical challenge. It’s an organizational challenge. It’s a behavior change challenge.
That’s why I’ve been so adamant that HR needs to be involved. The knowledge we have in HR about change management, about learning, about behavior, this is what’s being screamed for right now.
Very few HR people have actually taken the leap yet. I get it, AI still feels scary and complicated. And you haven’t made time to really understand what this means for how work gets done.
I’ve been nagging about this all autumn. You have to carve out time. I’ve been sitting with several different teams in recent weeks, just building and experimenting. And you can see it, people start making progress. They learn stuff. The joy of actually building something, of dedicating time to rethinking how things work.
You have to do this.
But it requires understanding what you can do with these tools. I’ve been nagging about this for nearly three years now: (did I mention three years enough yet?) we need to add some AI competence to our skillset. We just have to. It’s not optional anymore.
And I think, no, I know, this trend is going to stick. It’s not going to swing back to “oh, now we’re just talking about technology again.”
Because organizations are slow, and the gap between what’s technically possible and what organizations can actually adopt keeps getting wider.
McKinsey is talking about it. IT consultants are talking about it. I’ve helped several larger consulting firms this autumn add this organizational competence to their service palette because they clearly lacked it.
So again: if we dare to take this step, there’s so much to gain. But I think this is a permanent trend. The conversation has shifted from tech specs to org design, and it’s not shifting back.
3. True Agentic Organizations (Maybe? Probably?)
Okay, I’m sticking my neck out here. And this is the Hail Mary one, but hey, we need something fun to look back at in a couple of years.
I think by the end of 2026, we’re going to see the embryo of truly agentic organizations. Not everywhere. Not in large enterprises. But in smaller organizations, we’ll start seeing AI agents that can actually take on work tasks, more or less independently, and just solve them.
Now, some people will argue this is already happening. But I think what we’re seeing now is still too automation-like. Every time I ask for examples, what I get back are basically fancy automation workflows. (Pls drop viable examples in the comments should you know of any.)
What I mean by agentic is this: the agent acts independently on incoming information. It understands and interprets without manual intervention. It makes decisions and executes without constant human babysitting.
Even when I build “agentic” solutions (which I’ve been doing), they still require manual input at key points. They handle sub-tasks, but there’s still human work needed somewhere in the process.
What I’m talking about is giving an agent a task like “handle my bookkeeping” and it just... handles your bookkeeping without further instructions.
I think bookkeeping might actually be one of the first places this happens. But somewhere in that ballpark, tasks where you can say “do this” and without much additional guidance, it starts solving it.
I think we’ll see the first real examples of this by the end of 2026. Not widespread. Not in every company. But visible enough that we’ll all see it’s possible.
(I could be completely wrong about this.)
The Bonus Trend: AI Fatigue Is Coming
Here’s the overall insight I think we’ll see playing out across 2026:
Resistance to AI is going to sharpen. We’re going to see fatigue.
Especially among organizations that have been doing the “innovation theater” thing, testing a little here, piloting a little there, but not really committing the time and focus needed to actually implement this stuff.
Those organizations are going to get tired. They’ll burn out on AI.
But here’s the split: the organizations that do commit, that take the time, that build the capability, that actually go deep, they’re going to see incredible results.
So we’ll have this widening gap. Some organizations saying “we tried AI, it didn’t work” while others are quietly building genuine competitive advantages.
I also think we might see some specific vendors struggle. Microsoft Copilot is going to have a harder time keeping up. ChatGPT might find the competitive landscape getting tougher. The model race is still very much alive; we saw that with the recent releases.
So What Now?
Look, I’m not pretending I can see the future with perfect clarity. Nobody can. The pace of change is too fast, and there are too many variables.
But these are the patterns I’m seeing. These are my bets.
The digital twin becoming normal. Organization trumping technology. The first real agentic use cases emerging. And a growing split between organizations that commit and those that don’t.
We’ll see how this ages. Maybe I’ll link back to this in December 2026, and we can all laugh at how wrong I was.
But for now, these are my predictions.
What are yours?