The corporations of the future
Why do companies even exist? Why isn’t the economy just a vast collection of freelancers buying and selling services from each other in the open market?
The answer, according to economist Ronald Coase, is “transaction costs.” These are the frictions and overhead involved in using the market. Every time you want to hire someone externally, you must spend time and resources searching for the right person, negotiating the price, writing a contract, and ensuring the work is done correctly. A corporation exists because it is often significantly cheaper and faster to organize these activities internally. By hiring employees and establishing internal management structures, a company avoids the constant hassle of market negotiation. The firm thrives as long as its internal coordination is more efficient than dealing with the external marketplace.
Satya Nadella has been doing bit of podcasting rounds and his recent “Cheeky Pint” episode had plenty of interesting nuggets, but his musings on what the future corporations may look like was quite intriguing to me. Nadella connected Coase theorem directly to the profound challenge posed by advanced and ever more capable AI. Traditionally, a company maintained its efficiency advantage (its lower internal transaction costs) through “tacit knowledge” i.e. the unique know-how, processes, and expertise residing within its employees and organizational culture. This specialized internal knowledge is what made the company faster and smarter than the open market.
Nadella’s concern is that powerful, generalized AI models can radically disrupt this balance. If an AI model available to everyone “knows everything,” the cost of acquiring expertise and executing complex tasks in the open market suddenly plummets. The traditional advantage of having specialized knowledge inside the company dissolves if that knowledge is universally accessible via AI. In this scenario, the Coasean justification for the corporation is threatened, as the marketplace becomes just as efficient as the internal organization.
Therefore, Nadella argues that for a company to survive, it must redefine how it maintains its advantage. It must take its unique tacit knowledge and embed it into its own proprietary AI models or specialized layers. This customized “Corporate AI” becomes the new mechanism for lowering transaction costs. By ensuring its internal AI allows it to deploy and compound knowledge faster than the generalized AI available to everyone else, the company preserves its “sovereignty” and its reason for existing.
Here’s the specific back and forth between Satya Nadella and John Collison:
Satya Nadella: “…the ultimate sovereignty question is more of what’s the future of a corporation, right? I mean, if you sort of start to go to the core of the Coase theorem, you say, “Wow, what the heck? If the model is the thing that knows everything, why do I even… I’m supposed to have some tacit knowledge that makes the transactional costs inside my organization lower than just being in the marketplace.” So they’re a mind bender. So in fact, one of the ways I think is, the sovereignty that matters is your company’s sovereignty in an age where there are continual learning increasing returns to a model. So I’m increasingly thinking that hey, the company’s ability to have that intelligence layer that’s a scaffold or even weights embedded in the model. So it’s not somebody else’s foundation model. It’s about do you have sovereignty in your foundation model? So my new concept is the future of a company is that company has its own foundation model that captures essentially the tacit knowledge that makes the transactional costs of how knowledge gets accrued and diffused inside the organization faster. So that’s sort of a long speech on sovereignty.
John Collison: “Well, there’s two versions… That’s very interesting. The idea that AI maybe just changes the nature of companies, and you are saying that if some companies are already collections of IP, right? Disney or we had Dave Ricks from Eli Lilly here, that is an IP company in a big way. And some companies are already collections of IP, but right now that IP is in all the emails and documents and people’s heads most importantly, whereas maybe the IP could be in a single model over time. Where I thought you were going to go with that is just maybe the—people point out a lot that current companies are modeled after manufacturing companies and Alfred Sloan type stuff, despite the fact that we’re doing knowledge work today and not running a little manufacturing line. And do you get more just weird-looking companies? Do you get the famous really tiny billion-dollar company? Do you get more highly distributed internet companies? Do you get some DAOs? I thought that’s where you’re going to go with that.”
Satya Nadella: “I think that those are also possibilities. So the structure itself could change and it’s going to be more possible for whatever the few, the one-person billion dollar company, what have you, maybe could happen or DAOs could happen. But the interesting question, at least for me, is where does tacit knowledge reside? Clearly it resides in people’s heads and it’s the classic know-how that accrues and compounds. I think it’ll also reside and compound as weights in some LoRa layer that is unique to your company. I feel like the new intellectual property at Eli Lilly or at Microsoft or at Stripe at some point can be also, besides all the humans, besides all the other artifacts we have, I think we’ll also say, “Oh, they are in some embedding.”
And so one of the questions for all of us is how do you protect that from essentially leaking over to the base foundation model? Is it just like one capability hop away because it learned how to even do fraud detection? Is it just some other multidimensional, or not? And that I think is the key question to me. I think there are two arguments. One argument is that argument that the models are going to eat the world. You can kind of easily, oh yeah, after all, everything is just a pattern and I’ll learn it all and what have you. But then the thing though is, to your point about Stripe, it can take multiple models, build this unbelievable, sort of, I’ll call it fraud detection layer that is model-forward. And then there is this memory and tools use and action space that’s all unique to Stripe. That to me is the future of a corporation, whether it’s a pharma company, a payments company, or a software company. That I think is the work that we all are doing and will do. And I think that to me that is sovereignty.”
Nadella is basically arguing that modern AI reshapes those transaction costs around knowledge. If a company can encode its tacit know‑how into its own model layer, then the cost of moving, sharing, and applying that knowledge inside the firm drops sharply. Internal coordination gets easier and the firm becomes a faster learning machine. In Coase’s language, the “make vs. buy” frontier shifts inward again because the in‑house path becomes cheaper relative to the market.
To say it differently, Coase explained firms as islands of lower coordination cost. Nadella’s update is that those islands might soon be defined by weights and memories as much as by org charts and processes.
If a corporation’s essential knowledge is distilled into the weights of a proprietary AI model, that model may become the single most valuable asset. Valuation may shift significantly from traditional assets or human capital to the sophistication and uniqueness of the AI itself. As Nadella points out, the critical challenge then becomes preventing this specialized intelligence from “leaking” back into the base foundation models used by competitors, as such leakage could instantly neutralize a company’s advantage. It is perhaps no surprise that Nadella said “they’re a mind bender”!
In addition to “Daily Dose” (yes, DAILY) like this, MBI Deep Dives publishes one Deep Dive on a publicly listed company every month. You can find all the 64 Deep Dives here.
Current Portfolio:
Please note that these are NOT my recommendation to buy/sell these securities, but just disclosure from my end so that you can assess potential biases that I may have because of my own personal portfolio holdings. Always consider my write-up my personal investing journal and never forget my objectives, risk tolerance, and constraints may have no resemblance to yours.
My current portfolio is disclosed below: