Apparently the "Barrier" [1] bug is only in AP2 per latest A/B tests done by cpddan (who did the very first crash reproduction scenario in Chicago [2]) [link] Tesla Model X with AP2.5, SW 2018.6.1 [link] MS 75, AP1, SW 2018.6.1 1. [link] "works for 6 months with zero issues. Then one Friday night you get an update. Everything works that weekend, and on your way to work on Monday. Then, 18 minutes into your commute home, it drives straight at a barrier at 60 MPH." 2. [this video]
First crash reproduction scenario in Chicago by cpddan
This is gonna be interesting as apparently the "Barrier" [1] bug is only in newer AP2 models per latest A/B tests done by cpddan (who did the very first crash reproduction scenario in Chicago [2]) [link] Tesla Model X with AP2.5, SW 2018.6.1 [link] MS 75, AP1, SW 2018.6.1 1. [link] . "works for 6 months with zero issues. Then one Friday night you get an update. Everything works that weekend, and on your way to work on Monday. Then, 18 minutes into your commute home, it drives straight at a barrier at 60 MPH." 2. [this video]
First crash reproduction scenario in Chicago by cpddan
This isn't great news for Tesla. This 30 second video also recently surfaced where someone reproduced the fatal accident involving autopilot from this week: [this video] This could also be bad for Tesla if this is what occurred. Interestingly in this video the chevron lines exist, something people have been claiming would have averted the accident, also interesting that no warning alarm sounded at any stage.
The NHTSA faulted Tesla for the last fatal crash and its likely they will get faulted for the current. I am just shocked that after this one that they haven't suspended its use. The level of chutzpah this company operates at is very dangerous to its future but the public as well. The production issues are a whole different can of worms but the autonomous driving accidents are what the public sees. People need to understand that Tesla is not the only company that can deliver self driving hardware nor are they the only manufacturer of EVs. They are currently the maker of the more desired EV but with some of the old school automakers joining the fray they will see their high end model profit vanish. finally, I am still not convinced they ever plan to actually sell a base model. downvote edit Please see an owner recreation using the same version AP with the same condition z [this video] What is the rational thought it permitting that on the road.
[this video] that is what happened in the recent fatality, tell me how that is even remotely acceptable regardless of disclaimers. I seriously doubt adaptive cruise control would do the same, the majority of them work by following other vehicles.
If your point is "Autopilot only works on perfect roads" then they might have well pull the plug right now. Dirt, snow, rain, fog, dark, and other common conditions could obscure even correctly painted lines, and it just isn't realistic to expect perfect roads throughout such a large network anyway (in particular as human drivers can infer the line is meant to be there from the neighboring lines). At its core, a system like Autopilot is meant to stand in for humans. Plus in this video where they reproduced the accident there was a full chevron pattern so the argument is flawed/wrong anyway: [this video]
Who cares if he had 5 seconds to see the barrier if he only had 0.5 seconds to realize that the car went into casual murder mode right as it veers into the barrier. This video is quite relevant: [this video] A fully alert driver trying (and succeeding) in reproducing this behaviour. Observe how long the barrier is visible, how long it takes him to react, and how close the car was to hitting the barrier.
> The car is always in control, and its primitive object avoidance routines won't let it crash into a barrier.The car is always in control, and its primitive object avoidance routines won't let it crash into a barrier. Of course, how inconceivable that a self driving car could crash into a static barrier! [this video]
> Would the Volvo system have picked up and appropriately responded to the gore point involved in the recent fatal Tesla crash? I'd rather ask, would the Volvo system steer towards the barrier like Tesla's AP may have done [1] [2]? "IIHS research shows that AEB systems meeting the commitment would reduce rear-end crashes by 40 percent." [3] AEB on its own may save lives. Whether autosteer systems do or not is an open question. [1] [link] [2] [this video] [3] [link]
It still feels like you are driving but the path of least resistance is always pulling you towards the center of the lane. ...except when it doesn't, and you're forced to suddenly comprehend the unusual situation and react before its too late: [this video]
Wasn't the recent reproduction of the merge lane issue [1] showing that lane following is also not competent? I'm personally not sure any of the individual features of autopilot are really all that reliable. [1] [this video]
I don't know of any. But for some cases that Teslas seem to have trouble handling, like plowing into stopped white vehicles or concrete dividers[1], I think a 2D LIDAR like this one[2] would still be useful enough to have saved lives. [1]: a la [this video] or [link] [2]: [link]
Video list and Hacker News comment labels from HackerTube (prompted by @karpathy), released under ODC-By 1.0, built on the open-index/hacker-news archive. Hacker News data is provided by Y Combinator; comments remain the work of their authors. Video titles, channel names and thumbnails belong to their creators and are shown via YouTube. HackerTube Watch is an unofficial edition. Data export of 2026-10-31.
Plays via YouTube's privacy-enhanced embed
12 Hacker News comments
From the HN archive; up to 20 kept per video, favoring different commenters. Recommendations first.
Read more
Read more