Competing with Skynet: Could Distributed Autonomous Corporations Usher In a New Kind of Competition?
If you think competition between companies drives innovation, what might happen when they also have to go up against autonomous pieces of software running distributed across millions of computers through the Internet and around the world? It sounds like something out of a Singularity obsessed science fiction novel, but if you know where to look, the bones of this idea are already beginning to be seen today. The results might look pretty strange still, but there are some fascinating things happening in the “Distributed Autonomous Corporation” (DAC) area.
A DAC is a (so far) hypothetical construction that could perform at least some of the same functions as a corporation, non-profit organization, or other grouping of humans without the centralized legal or physical trappings of those organizations. This could be accomplished by creating a blockchain-type system (similar to that running Bitcoin) in which the code that makes up the DAC runs. DACs are simply algorithms tied to payment accounts that pay for their own computing cycles used, are paid for the services they provide, and can modify their own code.
DACs as an idea have been tossed around the Bitcoin community for a few years, and were somewhat codified in a series of blog posts by Stan Larimer beginning with “Bitcoin and the Three Laws of Robotics”. Larimer posits that the Bitcoin system itself is a DAC, suggesting that much of the network’s value comes from “performing a trustworthy confidential fiduciary service,” much like a Swiss bank would do. Unlike a Swiss bank, however, Bitcoin is open source and thus anyone can look at the code and be relatively assured that the network itself will act as designed and is worthy of trust. Of course, as we’ve seen time and again since the launch of the Bitcoin software, the same cannot be said of the human beings that may provide any related services.
So what if you want to create a service other than a confidential fiduciary? Bitcoin won’t really work since its distributed blockchain was created to do pretty much that one thing (though there have been some interesting efforts to push it beyond those boundaries). What you really need is a distributed blockchain platform for executing arbitrary code. Enter Ethereum. Billed as an update to the Bitcoin idea, Ethereum creates the idea of “contracts”: pieces of code in the network associated with their own balances. These can send and receive currency or information, and must pay for the processing they do.
With this basic infrastructure, the developers of Ethereum posit that any number of DACs are feasible. You can imagine, for instance, a Dropbox-like clone that receives data from users over the network, along with a small fee, and stores them for later retrieval, or a distributed identity service that would somehow store cryptographic authentication information for users and allow others to check against that database to verify identity, again for a small fee. Any profits could be paid back to the human (or other DACs?) that paid to instantiate the contract in the first place. Contracts can also be programmed to modify their own code, perhaps only after a “vote” of the shareholders.
DACs have a couple of attributes that may give them a competitive edge over their older, less distributed cousins. First, all the actions of the DAC are published and must follow the code in the contract. This provides some protections against corruption and fraud. Secondly, while the US provides a fairly fertile ground for new business ideas, DACs would embody the idea of permissionless innovation, if they had physical form. An aspiring entrepreneur with a good idea for a DAC only needs the money required to create a contract in order to get set up.
There’s a lot to be excited in there, but of course there are still more questions than answers. Modern corporate law has been tailored for centuries to help people run businesses. It’s not immediately clear that giving it all up is going to work out in the long run. Just to take one example, what liability protections do DACs have? Will the “directors” of a DAC be anonymous and if so is protecting affirmatively bad corporate actors in that way a good idea for society? There are no answers to those questions yet, but the process of discovering them is sure to be interesting.