5 ms·

For the most part I agree with you, but >I don’t care for arguments about safety statistics is insane. If it's safer it's safer, and the only way to know is s

For the most part I agree with you, but >I don’t care for arguments about safety statistics is insane. If it's safer it's safer, and the only way to know is statistics. The fact that (seemingly) no government department is collecting and publishing safety statistics on every self driving car manufacturer defies belief.

Waterluvian 3y ago
They’re my kids so I just don’t care if using them as a test subject results in longer term safer roads for everyone. But I think what you’re talking about is if autonomy makes my kids safer in the immediate, and not as an end result of public testing. Yes, I support that. I don’t think this is that yet. The word beta is right there.
theptip 3y ago
Strongly agree with the sentiment. Note NHSTA in the US recently started publishing incident counts for self-driving vehicles, but its total counts by company, not per-mile, and therefore mostly meaningless
6D794163636F756 3y ago
They specifically said test track. Once it's proven then put it on the roads. But in it's current state it should be severely restricted. They may need training data but then they can pay for it or pay to construct suitable tests
- sebzim4500 3y ago
  It's already on the road driving millions of miles. Someone should just count how many accidents it has had, and compare to how many are expected given the number of miles driven under each driving condition. If FSD has more accidents then ban it.
  - Waterluvian 3y ago
    I love this “logic.” You can do something wrong but if it works out, hey, it’s fine right? This also creates a wonderfully sloppy veil. Hey it’s proven safe. So we’re going to go ahead and fuck with the settings more.”
    - sebzim4500 3y ago
      If it improves safety then it isn't wrong
      - Tempest1981 3y ago
        I think the point is that if it's "proven" safer today, the manufacturer can then make changes and test publicly, and be absolved of responsibility if there are regressions.
  - TheCleric 3y ago
    Sure this will go fine since it's not like we have a track record of people dying in Teslas, with Tesla taking zero accountability, and withholding the crash data from open scrutiny. This experiment is sure to be unbiased.
    - sebzim4500 3y ago
      That's why the data should be collected by a government department rather than Tesla
      - TheCleric 3y ago
        I agree it should, but good luck getting Elon to cooperate to gather the necessary data points (e.g., was FSD enabled?).
      - Tempest1981 3y ago
        How quickly can the governments detect a regression? Weeks or months later? The claim here is that an overnight minor update from v11.4.4 to v11.4.7 caused a significant regression.
        jacquesm 3y ago
        That doesn't matter. What matters is that Tesla doesn't get to fudge the record, wipe evidence or misrepresent things.

Waterluvian 3y ago

They’re my kids so I just don’t care if using them as a test subject results in longer term safer roads for everyone. But I think what you’re talking about is if autonomy makes my kids safer in the immediate, and not as an end result of public testing. Yes, I support that. I don’t think this is that yet. The word beta is right there.

theptip 3y ago

Strongly agree with the sentiment. Note NHSTA in the US recently started publishing incident counts for self-driving vehicles, but its total counts by company, not per-mile, and therefore mostly meaningless

6D794163636F756 3y ago

They specifically said test track. Once it's proven then put it on the roads. But in it's current state it should be severely restricted. They may need training data but then they can pay for it or pay to construct suitable tests

sebzim4500 3y ago
It's already on the road driving millions of miles. Someone should just count how many accidents it has had, and compare to how many are expected given the number of miles driven under each driving condition. If FSD has more accidents then ban it.
- Waterluvian 3y ago
  I love this “logic.” You can do something wrong but if it works out, hey, it’s fine right? This also creates a wonderfully sloppy veil. Hey it’s proven safe. So we’re going to go ahead and fuck with the settings more.”
  - sebzim4500 3y ago
    If it improves safety then it isn't wrong
    - Tempest1981 3y ago
      I think the point is that if it's "proven" safer today, the manufacturer can then make changes and test publicly, and be absolved of responsibility if there are regressions.
- TheCleric 3y ago
  Sure this will go fine since it's not like we have a track record of people dying in Teslas, with Tesla taking zero accountability, and withholding the crash data from open scrutiny. This experiment is sure to be unbiased.
  - sebzim4500 3y ago
    That's why the data should be collected by a government department rather than Tesla
    - TheCleric 3y ago
      I agree it should, but good luck getting Elon to cooperate to gather the necessary data points (e.g., was FSD enabled?).
    - Tempest1981 3y ago
      How quickly can the governments detect a regression? Weeks or months later? The claim here is that an overnight minor update from v11.4.4 to v11.4.7 caused a significant regression.
      - jacquesm 3y ago
        That doesn't matter. What matters is that Tesla doesn't get to fudge the record, wipe evidence or misrepresent things.