News for hackers. Stuff that matters.

A comment on Cruise is opening driverless cars to the public in San Francisco. Parent comment

Re:Cruise is opening driverless cars to the public in San Fr... by waffle_maniac on Tuesday February 01, 2022 @04:16PM (#30169424)
> current FSD beta's intervention rate is more like one per 10 miles

Maybe in rural areas? The videos on YouTube are far more than one per 10 miles.

https://www.youtube.com/watch?v=wTybjJj0ptw

[ Parent | on HN ]
Re:Cruise is opening driverless cars to the public in San Fr... by V__ on Tuesday February 01, 2022 @04:24PM (#30169529)
On quick watch the driver intervenes at 4min 45sec and 5min 47sec.
[ Parent | on HN ]
Re:Cruise is opening driverless cars to the public in San Fr... by whimsicalism on Tuesday February 01, 2022 @04:22PM (#30169500)
A helpful link for your perusal: https://en.wikipedia.org/wiki/Selection_bias
[ Parent | on HN ]
Re:Cruise is opening driverless cars to the public in San Fr... by gitfan86 on Tuesday February 01, 2022 @04:41PM (#30169803)
But are you using confirmation bias to find a cognitive bias that fits here.

But, In all seriousness we don't have access to the data across all 60k FSD users to know what the intervention rate is and how it has been changing over time.

[ Parent | on HN ]
Re:Cruise is opening driverless cars to the public in San Fr... by bluGill on Wednesday February 02, 2022 @10:23AM (#30178823)
We do have previous statements that as they get better they are moving to harder situations. Start with empty roads, and once you can do them well start finding harder and harder situations. When you start you avoid construction zones, once you are doing well you start looking for them.
[ Parent | on HN ]
Re:Cruise is opening driverless cars to the public in San Fr... by blahblarblar on Tuesday February 01, 2022 @07:09PM (#30171673)
Dirty Tesla used to track these stats in his testing and gave up because “it’s not changing”
[ Parent | on HN ]
Re:Cruise is opening driverless cars to the public in San Fr... by Retric on Tuesday February 01, 2022 @10:14PM (#30173220)
Which could be a sign some drivers are simply overly cautious. Suppose 1/10 of disconnects prevented a crash, reducing the risk of crashing to 0 only reduces the number of disconnects by 10%.

To actually reduce that number you would need to make drivers feel more confident in the vehicle which is a useful metric, but only indirectly relating to safety.

[ Parent | on HN ]