Machine, Platform, Crowd: Harnessing Our Digital Future

At the time, a fierce debate was raging about whether centrally planned economies like that of the Soviet Union—in other words, economies where there was a single core responsible for creating and distributing goods and services—worked better than free market economies where planning and production were done by an undirected, decentralized crowd. M

Wealthfront is a wealth management company that has not only turned away from using human judgment when

everyone’s preferences, and feed all this data into a “single mind”—a giant economic optimization algorithm that would run continuously to “work out the implications”? Because, Hayek explained, that algorithm would never get all the data it actually needed; it could never “secure the best use of resources known to any of the members of society, for

reinforcement learning,†† has turned

Incomplete contracting and residual rights of control will probably always necessitate the existence of companies.

One promising approach here is to allow people to flag fake or offensive content while training machine learning systems to spot it automatically.

the right person or team to help with something. The chances of finding a good fit increase with the number of people who see the request, which explains why platforms for task matching have become so popular.

Markets are therefore called “emergent” systems: prices emerge from all members’ interactions and can’t be observed just by looking at a few.

Linux is without question the biggest and most professional operating system in the world, today found on everything from the servers in data centers bigger than a football field to over 1.5 billion Android phones and tablets.

sequencing technology has spread while declining exponentially in cost,‡ farming, animal husbandry, and other sectors have also been affected.