AI tends to accentuate many of the types that Foster describes, and the trends in Merton's rubric above. AI millenarians (Musk, EA, MIRI) combine certainty tha

AI tends to accentuate many of the types that Foster describes, and the trends in Merton's rubric above. AI millenarians (Musk, EA, MIRI) combine certainty that superintelligence will come about with concern that it will be bad. They are in Foster's "shouldn't" class. AI boosters, which are most of the people working on AI in big tech and startups, are in the "could" bucket. They still see the rainbows, short term and long term. AI pragmatists (Hinton, LeCun, Ng) minimize what AI can do, focus on near term gains, and on the difficulties in achieving them. These people are also mostly in big tech and startups, but usually closer to the code and the data. They are the "shoulds" and sometimes the "mights". But there are a couple other dimensions to AI that I'm not sure I have the right words for. There's a certain kind of person who is more of an adventurer. They just want to be close to the action, almost regardless of the outcome. They are building AI, sometimes very powerful AI, but the ultimate consequences of that work are less important than the fact that they, as individuals, are at the fulcrum of this historical moment. And there is a large faith-based element to AI, which I suspect applies to other futurist topics to a lesser extent. And that faith-based framework creates tribal arguments about unfalsifiable claims, like the advent of superintelligence. Instead of "should" or "should", they fit into the "will" class, and they provoke skepticism that is just as strong and irrational. Pragmatists sometimes tip over into this group.