TLDR · MISCELLANEOUS

SELF-DRIVING CAR BETS

lesswrong.com40 min read

Summary from the TLDR newsletter

While investors were optimistic enough to bet on self-driving cars back in 2016, most experts and people in tech thought it was pretty unlikely to be ready by 2023. People seem to treat self-driving as if it were an easy AI problem, but perceiving and quickly reacting to the world is something that we evolved to be good at, and driving sometimes requires complex problem-solving. There is a big gap between 'pretty good' and 'actually good enough'.