2025: in balance with our agentic twins

As we enter 2025, one trend stands out above all others: the rise of Agentic AI. While artificial intelligence has been a buzzword for decades, 2025 promises to be the year when AI agents become an integral part of our daily lives, reshaping how we interact with technology and each other. This isn’t a revolution, but rather an evolution – one that’s been long in the making and is now reaching a critical tipping point.

There seems no escape to the idea that we will have a year on agents, an evolution not a revolution, as someone framed it right; we did not invent agents just now, or in the last year, the intelligent agent is as old as the internet almost, or at least from the early days. It is a returning concept that is appearing in different shapes. In the first decade of the internet it became the search agent, finding stuff on the fast growing pool of information that started to appear. And to jump ahead, Google introduced the assistant agent some years ago already to delegate making your appointments with the hairdresser, and toning it down to helpful calendar planning connecting all your channels they master.

And now we are at the Agentic AI, the year of agentic AI. Will it speed up into a new reality very quickly, or do we expect more slowly incremental development? And what form (or format) will the agents get? Will we have a flock of agents we can master as a team of task-performers interacting with each other? I see that happening. Or will we have one general agent in line with the AGI thinking? A personal assistant that arranges everything? Maybe both.

We are still early and still discovering it. Companies will start doing experiments to add AI to their core promises, hoping to stay ahead of the curve. 2025 is the year of experiments, putting the toe in the water to test the temperature. I think it makes a lot of sense to expect that.

At the same time, there is a strong driver in the personal agents, tools for people to build their own. And especially solve their very specific problems. That learns from you and your specific quirks. That will create a world full of agentic twins. But before we can expect that to be a daily reality, we need to have useful tools. Sure, you can do a lot already with current tooling, but it is definitely still far from the ‘word processor’ level. The WP4.2 version will surely be introduced in 2025, and maybe even the Windows 95 level. Are we leapfrogging to an iPhone OS 1?

We can expect that the first sensible agents in mundane products might appear. Will IKEA launch its first Tradfri agent? Or Bosch ‘Like a Bosch’ will not serve you but take over your control. CES is about to happen, and we will get a taste of the agents in our appliances.

We discussed Generative Things last month, TH/NGS 2024, and it makes a lot of sense to think about regenerative things instead. How to prevent a new wave of environmentally heavy things. What Alexandra DS is sketching so nicely is: Can we use intelligence to create products that last longer by being more loveable and repairable? Is this a popular direction for even faster fashion and impulse-driven experiences?

For 2025, I think we will see the first valuable agents and a lot of vaporware. And we will see a dancing on the volcano of AI promises. CJ Trowbridge describes nicely how we have a huge potential with AI to create things that help us to understand what we want, to make stuff that will help us, but at the same time, are the current AI companies looking for a business model that is built on real value, not on over expectations and promises not met. And then I leave out all the unpredictability in the world as a society, finding new balances with unclear outcomes…

That is, for me the core of 2025: a year of finding a healthy balance between potential new supporting fellow technology to shape what relationship we want to build and, on the other side, the fragile delivery of expectations made, both in delivering the real value, embedding it in a solid societal framework that helps us deal with the rapidly rise of uncertainty in our environment. Let’s build a fruitful companionship with the agentic engines towards a more-than-human balance.


Buy Me a Coffee at ko-fi.com