Top Technology Trends in 2025 and 2026. Part 2

Continuation of Technologies in 2025 Overview

Peter:
If we’re talking about the coming year, I think we need to be very cautious about a number of things. If I were to give a few predictions for 2026, Mark referenced governance, and I believe this will become a much bigger issue, especially around the application of artificial intelligence. We already know that the European Union has fairly stringent regulations coming into force.

We also know that many other jurisdictions are likely to mirror this playbook and introduce their own legislation around how AI is used and governed. I would personally recommend that anyone involved in technology start thinking seriously about regulation. What were some of the lessons learned with GDPR (General Data Protection Regulation)?

Many organizations chose to pretend GDPR wasn’t going to happen. They ignored it, and then it did happen, with major implications. We should now think about how AI regulation will be introduced, whether in the EU, in different U.S. states, or in Canada, Australia, New Zealand, and parts of APAC. This will be a reality, driven by the need to clamp down on certain use cases, such as the examples Mark referenced earlier in the toy industry. And that’s just one of many.

Mark:

We saw GDPR come into force a few years ago to establish data protection, but now we’re seeing AI turbocharge what organizations can do with customer data. This gives brands insights into customer behavior that we’ve never seen before.

I think we’re going to reach that uncanny valley very quickly. You’ll be interacting with a company that knows everything about you: when you buy, what you buy, what you like, what you don’t like. They’ll push offers that they claim are perfectly timed and tailored for you.

But many people will find this intrusive and unsettling. You might be having a conversation with your partner about buying a new table and chair set, and suddenly your phone pings with an ad saying, “Are you interested in this table and chairs from IKEA?” That’s the kind of thing we’re going to see.

This happens because AI models are being used to create deeper insight. How that fits with GDPR, which was designed almost a decade ago, will be one of the biggest governance challenges we face.

Peter:

I 100 percent agree. Another prediction for next year, from the perspective of technology buyers, is that after three years of intense AI hype, there will be growing pushback. Mark and I often talk about how LinkedIn feels polluted by the hype cycle, and that doesn’t help anyone.

It’s not helpful for developers, buyers, or observers of the market. What we’re starting to see now, and what I think will become very significant next year, is buyers pushing back against vague narratives. They’re asking providers for real use cases.

Technology companies can talk endlessly about what AI can do, but buyers want case studies. They want proof. They want providers to say, “This is what we’ve done in this specific industry. This is how we’ve already helped an organization using AI.”

Budgets are tight. The economic and geopolitical environment does not support uncontrolled spending. The money that is spent will be spent carefully. Providers will need to be specific if they want to succeed.

Mark:

You could probably label this phase as pragmatic AI. Since ChatGPT launched three years ago, at the end of 2022, we’ve seen an explosion of interest in generative AI, followed by interest in agentic AI.

Agentic AI is still largely experimental. There are good real-world examples, but most discussions still focus on pilots. Generative AI is more mature, but the problem has been the hype. There’s been a lot of talk about replacing people and eliminating jobs.

Peter and I work a lot in customer experience. For the past couple of years, the focus was, “How do we replace agents? How do we remove people from customer interactions?” What became very clear this year is that a better use of AI is productivity, quality, and service improvement, not replacement.

AI helps people do their jobs better. That’s where the real value is. We’re seeing more pragmatism now. There’s also pushback against pilots that don’t have a clear path to enterprise scale.

Instead of running endless pilots, organizations are asking whether there’s a realistic path to enterprise value. Scaling AI usually requires fundamental changes to how data is organized, and that’s a significant investment. AI projects will continue, but they’ll focus more on delivery than experimentation.

Peter:

From a marketing standpoint, this will resonate. It’s been three years since the ChatGPT moment, and AI is now embedded in many technology tools. It’s part of the inner workings of solutions.

Technology providers need to accept that AI is becoming expected. It’s no longer a differentiator. It reminds me of when Pentium processors first appeared in the 1990s. At first, it mattered. Then it became assumed.

AI is heading the same way. If you don’t have it, that’s the problem. But constantly selling AI as a feature will lose relevance. The focus needs to be on outcomes, not the ingredients.

It’s like ordering sausage at a restaurant. You don’t care how it’s made. You care whether it tastes good. The outcome is what matters.

Mark:

That also raises the question of whether we’re overspending on AI research. Most organizations care about business outcomes. They want to know what they gain and how AI improves what they offer to customers. People don’t care what software runs inside a company. They want faster, better, cheaper service. That creates challenges for pure-play AI vendors.

If you’re charging hundreds of dollars a month for AI tools, how long will customers keep paying? Eventually, AI will be bundled into existing products, just as Microsoft does. People will expect it, not pay extra for it.

We’ll also expect e-commerce platforms to know our preferences automatically. That acceptance could be problematic for companies whose only product is AI itself.

Peter:

AI is everywhere, but cybersecurity remains critical. Data security has never been more important. There’s almost no week without a major breach. From my research with enterprises, security is top of mind in the C-suite and among compliance officers. Protecting organizational and consumer data is a priority.

Mark:

AI cuts across everything. If we’re talking about automation, cybersecurity, agentic, we could just abbreviate it and just say that AI is changing everything. Applying AI to a single system is easy. Scaling it across the organization is not.

Most organizations still operate in silos. Data doesn’t flow freely. Agentic systems require shared data and permissions across departments.

The dream is simple: ask your phone to book a flight within a budget and preferences, and it just works. But for that to happen, organizations need a unified data layer.

That’s going to be one of the biggest blockers in 2026. Executives may want agentic solutions, but CIOs will respond that data foundations must be fixed first. Governance and data architecture come before automation.

One last point is physical technology. Automated vehicles, delivery robots, autonomous taxis are already here. What once felt like science fiction is becoming normal. In a few years, this will be fully normalized. AI in physical systems will be a major trend.

Peter:

I agree. I was recently in San Francisco, saw autonomous cars everywhere, and even tried one. This isn’t going away. It’s on the verge of becoming mainstream. 2026 might be the year when it truly arrives.