4 ms·

I'm using the same data Tesla used when they made their claims.

• greglindahl 10y ago
  And that adds to a well-reasoned and statistics-informed discussion on HN, because... ?
  • studentrob 10y ago
    None of Tesla's recent press releases indicate their data is error-prone [1] [2] Rather, they mention a "wealth of internal data demonstrating safer, more predictable vehicle control performance when the system is properly used." I'm drawing a different conclusion from the same data. I'm not claiming it's gospel. I'm just an internet commenter. Forthcoming investigations from the NHTSA and NTSB may shed more light. [1] https://www.teslamotors.com/blog/tragic-loss https://www.teslamotors.com/blog/tragic-loss [2] https://www.teslamotors.com/blog/misfortune https://www.teslamotors.com/blog/misfortune
    • greglindahl 10y ago
      I was trying to engage you in a discussion about your error, not Tesla's. Hopefully you're an internet commenter trying to make Hacker News a better place?
      • studentrob 10y ago
        > Hopefully you're an internet commenter trying to make Hacker News a better place? For sure. Through this discussion about assisted-driving cars, I also try to contribute to a safer driving community, more transparency from companies offering driver-assistance systems, thoughts about the best ways to advance AI, and better capital allocation. > I was trying to engage you in a discussion about your error, not Tesla's. I don't see what error of mine that you are pointing out here. It sounds like you're talking about Tesla's data: >> Given the large error-bar on the Tesla data, you have no idea if Tesla is less safe or more safe than the average American driver.