Radical agency
Feeling AI abundance
Almost everyone on my feed is asking what AI does to the economy, management, policy, geopolitics. Asking about who gets fired first, when. Asking about why openAI won’t ghibli studio our children.
Apart from the maddening frustration of being blocked by AI safety, I’ve been thinking more about how AI abundance makes me feel.
I’ve realised that I’m not only doing much more of what I used to do. I’m doing new things. I call this intense and unsettling feeling radical agency.
Most of what I do is turn ideas into words — opinions, research, and policy. I’m in a near-constant group-chat with many colleagues and friends about how AI changes what ideas-producers like us do.
As we write in our upcoming book, Institutional Acceleration, we are in a supertransition — a shift driven by technologies like AI, crypto, quantum computing, low earth orbit satellites. The implication is that we are making digital institutions that help us to coordinate in new ways.
But as I sit here in front of my laptop (see image above), this transition actually feels like more agency. This agency is most tangible with AI, but I suspect it will expand as other technologies combine with it.
What is agency anyway?
By agency I don’t mean that I can do the same things faster. I don’t just mean more (hopefully better) words. I mean I can now produce different things.
I can write an idea — a new business model or democratic system — and then design an infographic, record a podcast, maybe even vibe-code it into a prototype.
I can do this because these tools let me align my goals with the model’s embedded knowledge.
I would never have done all of these things before. Not just because I wouldn’t have paid a designer or a coder, but because the friction was too high.
As I teach my students: the answer is always transaction costs. Transaction costs prevent mutually beneficial trade from happening. Radical agency is emerging because those costs — to produce, test, prototype, share, disseminate — are falling fast.
This gives me more agency.
Agency isn’t free
Ronald Coase taught us why firms exist. The answer was that there are costs to using markets, and so some activities are best done inside firms.
But transaction costs change. And so the boundaries of firms and markets also shift. New types of organisation become possible.
That’s what I feel like AI is doing to me.
Previously I wrote about the problem of what to ship when ideas are free. Fundamentally that’s a problem of agency. But now it goes even further. My life feels more entrepreneurial now. I’ve written plenty about what entrepreneurs do.
The challenge for me today is entrepreneurial: managing the workflows, ideas, and tools, and constant judgement calls, even when I don’t really know what’s going on.
An agentic economy isn’t just more AI robots running (?) around the internet trading. It’s also the deeper problem of navigating a world where I have to decide — constantly — what I should do, because now I have the agency to do much more.
Patterns shifting
The patterns of specialisation and trade that made sense in the pre-AI economy no longer fit. More personally, my own patterns and workflows don’t fit. I’m trying to discover new patterns for how I organise and trade in my own life.
We are in a supertransition. And like we wrote in Unfreeze, following the COVID-19 pandemic, this kind of disruption demands rapid adaptation. I just have to figure out how to do that.