One thought on “Marc Andreesen”

1. I’m not a luddite, but I’m gravely concerned. My two answers to Andreesen’s “prove that this time it’s different” challenge:

   1) Human wants and needs aren’t infinite, although they’ve been effectively infinite so far. But human brains are finite and the amount of time humans can spend on any activity is also finite. I think we’re pretty close to the point where we might start to bump up against those limits.

   Of course, there are plenty of humans, and they all have different wants and needs, so there are plenty of markets for plenty of goods and services. But when you can produce everything very, very quickly and very, very cheaply, economic activity will eventually shift from a positive sum to a not-quite-zero sum game.

   2) Human competence is finite. The question isn’t whether there’s something new to do that a robot isn’t doing. It’s whether the robot is better suited to do the new thing than x% of the labor pool, where x is a pretty large fraction. I hope Andreesen is right and people have more potential than I give them credit for, but that presupposes unobvious advances in education and/or “cognition technology”. (Perhaps “cognition technology” solves the wants and needs problem, too, but it’s too much of a unicorn for me to count on so far.)

   I think he’s right that a lot of things get really, really cheap, and that’s good. (Although I wonder about housing and real estate…) My prediction, which I understand is hugely unpopular here, is that we’re going to need an extremely robust welfare state to deal with the workers that can’t compete. The good news is that such a welfare state is pretty easy to afford, because many goods will be really, really cheap. NB: I’m not talking about a welfare state to address inequality. I’m talking about one capable of satisfying enough needs that people can live reasonably happy lives.

Comments are closed.