How to turn the AI Economy into wealth, income, and pensions for millions
Six years ago, I floated a business plan for building a new type of company (an open source venture). An open source venture is structured in a way that makes it possible for millions of worker/owners to collaborate on the construction of extremely valuable databases. That initial attempt wasn’t successful, with good reason: it was too early. The databases that could be built weren’t as valuable as they needed to be to support millions of contributors and the system needed to manage decentralized ownership (the blockchain) was still in its infancy.
Things have changed. The rapid rise of AI (more specifically, deep learning/ConvNet), has turned massive, human labelled/curated, databases into some of the most valuable information on earth. Second, the blockchain is now mature enough to support the infrastructure needed to make it possible.
Here’s some more detail. Building AIs requires three things:
- Scalable models of AI construction with high capacity (we have that now deep learning/ConvNet now).
- Lots of processing power (to train AIs). Moore’s law cracked the initial barrier to using these methods. Cloud based training systems filled with GPUs is making this less expensive.
- Big data sets (specifically, human curated data) are used by the AI’s to learn.
You should already see the how well suited open source ventures filled with millions of worker owners fits into this process. For example:
- A recent study by Google showed that the size of the training database correlated to the quality of the AI. The bigger the better. That’s a bit of a problem for the big tech companies. Currently, most of the of the databases currently available were built by very low cost workers for a fraction of a penny a label, using places like Amazon’s Mechanical Turk. That makes them clunky to build and difficult to scale. Further, the labelled, human curated data they do have (on Facebook, Google, etc.), while voluminous, isn’t complete enough to do high quality training. An open source venture fills the void. It could enlist millions of worker/owners to build, maintain and extend the data needed would be very successful here, creating some of the most valuable AIs in the world.
- The processing power needed to train these AIs, as they become more sophisticated, could outstrip current corporate capabilities (of even Google). This suggests that training systems that break apart the computation tasks among tens of millions of participants (cell phone/desktop) could surmount this barrier. In practice, it would be a combination of folding@home and bitcoin mining. Already, deep learning AIs can “run” on cellphones. Configuring them to contribute to the training load and/or actively gather data for database construction isn’t that much of a stretch.
- The people participating (earning equity instead of fee for service for their contributions) could become very, very large. They would be familiar with the platform and become increasingly sophisticated at finding amazing ways to capture new data and new AIs to build. They would also be a built in audience of early adopters for new AIs, pushing them across the chasm into the mainstream and radically improving their operation due to the sophistication of their interaction with them. Over time, a platform like this could become the source of many (if not most) of the best AIs ever built. A source of immense wealth (seen as dividends) for hundreds of millions of owners, earning equity with each contribution.
Worth thinking about.
PS: Given our experience with bitcoin, this isn’t impossible to do. It’s also a much better future than a world that one based on Turking day laborers.