A simulated society; Ted Gioia makes pessimistic near-term predictions; Spencer Schiff makes optimistic near-term predictions; Scott Belsky makes business near-term predictions
I hope Spencer Schiff is right, but he's predicting a lot of improvement in a year. Sora, which he predicts will be far surpassed, is not even available yet.
Belsky seems to be suggesting that AI can reduce production costs as well as transaction friction. 5% to 15% of global GDP is estimated to consist of transaction costs. If AI can actually reduce costs, might it actually reduce GDP? Do economists have criteria for discerning when GDP reduction is a net positive?
This reminds me of programming assignment I had in school where we needed to estimate pi by having a cannon fire a ball at random into a circle inscribed by a square.
Does running a tribe of agents against a problem define what is in the solution space so that it might be easier to focus on what is not in the space?
For example
1) Build the model as you indicate with a bunch of agents
2) Have a collection of human theories about how to solve a problem. All different sorts of answers.
3) Build a Design of Experiments of all possible solution path ways (Go to war, negociate, subsidize, etc - ever bossible action variant)
4) Regression test against the model
5) Some things will have null impact on the known model result
6) Explore why they were null, those are the out of the box things for which the model is missing knowledge.
I hope Spencer Schiff is right, but he's predicting a lot of improvement in a year. Sora, which he predicts will be far surpassed, is not even available yet.
Belsky seems to be suggesting that AI can reduce production costs as well as transaction friction. 5% to 15% of global GDP is estimated to consist of transaction costs. If AI can actually reduce costs, might it actually reduce GDP? Do economists have criteria for discerning when GDP reduction is a net positive?
I'm ok with Gioia's list of AI fakes but that is in no way what artificial in AI refers to. Saying so makes him sound stupid.
This reminds me of programming assignment I had in school where we needed to estimate pi by having a cannon fire a ball at random into a circle inscribed by a square.
Does running a tribe of agents against a problem define what is in the solution space so that it might be easier to focus on what is not in the space?
For example
1) Build the model as you indicate with a bunch of agents
2) Have a collection of human theories about how to solve a problem. All different sorts of answers.
3) Build a Design of Experiments of all possible solution path ways (Go to war, negociate, subsidize, etc - ever bossible action variant)
4) Regression test against the model
5) Some things will have null impact on the known model result
6) Explore why they were null, those are the out of the box things for which the model is missing knowledge.