When it comes to artificial intelligence, I am an optimist. In fairness, this may be something of an understatement: Over the past year and a half, I have argued that AI will improve education, boost productivity, enrich the internet, make markets more stable and land more valuable, accelerate the pace of scientific progress and increase social trust.

I remain optimistic. Nonetheless, I do have worries about AI — or, to be more precise, about how we will respond to it. There is one big thing, if I may indulge again my inclination for understatement: AI will overturn the existing world order, and we are not very well prepared for it.