4 ms·
Somehow EA always comes back to spending money on 'working on AI risk' which somehow never produces anything but blog posts.
by EamonnMR 3y ago
Somehow EA always comes back to spending money on 'working on AI risk' which somehow never produces anything but blog posts.
- jononomo 3y agoFunny how AI risk is seen as this big threat, but self-driving cars apparently can't avoid crashing.
- zaroth 3y agoIn all fairness, the human driven ones crash an awful lot too! Keep in mind that truly reliable and robust AI driving will save 10s of millions of lives and trillions of dollars of avoided economic losses. Creating AI that can drive a car that virtually never crashes will be the pinnacle of software engineering, and a turning point in our civilization. The savings in blood and money can’t come soon enough, but otherwise it’s basically a Very Good Thing it’s going to take a little while to get it really working. We’re not really ready for the implications of software that can reason that well in the real world.
- travisjungroth 3y agoConsider the state of airplanes, rockets and nuclear weapons in 1923 vs 1955.
- PragmaticPulp 3y agoDon’t forget the 80,000 hours organization, which started as a good idea to encourage people to go into altruistic careers but then seemingly only produced volumes of podcasts and inactionable blog posts. I recommended it to a student once who was asking about career guidance. They came back and said they couldn’t figure it out, so I went looking into it deeper. I couldn’t get anything out of it either. It was just endless rambling podcasts at the time