Dutch regulator approves Tesla’s FSD driver-assistance system but won’t release safety-test details

Tesla has pressured officials to withhold safety-testing results from the public

  • bluGill@fedia.io
    link
    fedilink
    arrow-up
    0
    ·
    3 months ago

    Self driving has potential to be better than humans. However lacking independent analysis I have no idea of it is. The only people who are taking have reason to lie with statistics.

    Release the data- all of it- so we can figure out the truth.

    • NaibofTabr@infosec.pub
      link
      fedilink
      English
      arrow-up
      0
      ·
      3 months ago

      So if the self-driving machine learning algorithm learns by collecting data from human drivers, how could it ever be better than the human drivers it learns from? Especially knowing that these models are nondeterministic and often generate unpredictable and wildly incorrect output.

      • vandsjov@feddit.dk
        link
        fedilink
        English
        arrow-up
        1
        ·
        3 months ago

        how could it ever be better than the human drivers it learns from

        Not that I trust these systems too much yet, I do see why they can be better:

        1. Reaction time - can break and/or steer much fast than a human
        2. Awareness - can see “everything” all at once
        3. Never tired, influenced or disturbed
        4. What they lean from humans is not just how to handle different situations but also lean from human mistakes
        5. The fact that there are 2 billion drivers world wide (my own made up number) that needs to learn from the ground up and get better with all the falls humans brings to the table, but the car is just one AI (maybe there are more) that needs to be fine tuned to be the best

        But as you say, the AI is still putting out weird things once in a while, and maybe the next big painting on the side of a truck makes the AI drive over into incoming traffic…