I think this bit was my favorite part of the article, because it points towards the sort of munchkinry that machine intelligence makes easier:
I can’t wait to see more combinations of the practical and eccentric. A few years ago, a company like Orbital Insight would have seemed farfetched—wait, you’re going to use satellites and computer vision algorithms to tell me what the construction growth rate is in China!?—and now it feels familiar.
I had first heard of this sort of project… three years ago, maybe? One sample use case then was counting cars in Wal-Mart parking lots to estimate sales figures to trade ahead of Wal-Mart earnings calls. It calls to mind the early commercial buyers of telescopes—merchants who would use them to see what ships are coming into harbor (and possibly price info displayed by those ships with flags), and then trade on the early knowledge.
I think this bit was my favorite part of the article, because it points towards the sort of munchkinry that machine intelligence makes easier:
I had first heard of this sort of project… three years ago, maybe? One sample use case then was counting cars in Wal-Mart parking lots to estimate sales figures to trade ahead of Wal-Mart earnings calls. It calls to mind the early commercial buyers of telescopes—merchants who would use them to see what ships are coming into harbor (and possibly price info displayed by those ships with flags), and then trade on the early knowledge.