[FSD Beta 8.2] Oakland - Close Calls, Pedestrians, Bicycles!

Maybe I should have linked to [1] as one of the videos from a couple of guys testing it out in Oakland. [1] ▶ this video

Cmon. There's a video of a Tesla FSD driving on the wrong side of the road, cutting off a car it did not see at an intersection, then accelerating straight into a fence, all in the space of five minutes in Oakland. Tesla has by no means solved the vision problem. Their software is laughably bad. ▶ this video

Copy-pasting a glorious Tesla comment from yesterday: Videos of Tesla fails, mostly from 2021: [1] ▶ this video [2] link [3] link [4] link [5] link Other things: [1] link [2] a video which was taken down from youtube link [3] link [4] link . Bonus material: [1] link For saner info regarding self-driving, A.I etc: [1] link [2] link [3] link [4] book to read, suitable even for laypeople >> 'Artificial Intelligence: A Guide for Thinking Humans by Melanie Mitchell Perils of autopilot: Children of the Magenta Line link

Videos of Tesla fails, mostly from 2021: [1] ▶ this video [2] link [3] link [4] link [5] link Other things: [1] link [2] a video which was taken down from youtube link [3] link [4] link . Bonus material: [1] link Edit: For saner info regarding self-driving, A.I etc: [1] link [2] link [3] link [4] book to read, suitable even for laypeople >> 'Artificial Intelligence: A Guide for Thinking Humans by Melanie Mitchell IMO everyone should read that book Edit 2: Children of the Magenta Line link

> I have doubts that Waymo will ever get to Level 5 without 3D mapping the entire world Waymo is not trying to get to L5 at all. They are a strictly L4 technology and their focus is clearly on servicing where the market is, metropolitan cities. L5 is a pipe dream for anyone; it's a moving target no one can ever reach. It essentially means the car can drive anywhere, in all conditions and without the driver needing to take over at all. So it really means Tesla can't declare FSD as "done" until it works for everyone, everywhere and in every condition. How close do you think they are for that to happen? Does that also mean if it encounters a unique situation and gives up, their L5 claims fall flat? Given their current poor performance [1], they have a long way to go to even let drivers take their hands off the wheel, so I'm not holding my breath here (vision stack or not). The real end game for self driving is L4 because that's what you can reasonably promise to your users. That it works in a defined, tested area and when required conditions are met. This is why literally every SDC company is L4-only, they never promise L5. Either Tesla is focusing on the wrong problem instead of providing value to its paying customers or they are blatantly misleading customers by promising "L5 by the end of the year" every year. [1] ▶ this video

Tesla's FSD is a complete fraud. Musk has been saying it's less than a year away since 2018 ▶ this video ▶ this video ▶ this video ▶ this video

It's absurd to me that there haven't already been legal consequences for this. It isn't 'full self driving' and Tesla has been doing it for years. (remember 'autopilot', which I think at least German courts ordered them to remove). And not only is the marketing fraudulent, I also found the manner in which they let 'beta test' software loose on streets completely irresponsible.[1]. I have no idea why that didn't have consequences either. [1] ▶ this video

> Have you seen the FSD beta videos? Looks like they've pretty much solved it for autonomous driving. Have you? Plenty of videos where it fails pathetically (like ▶ this video ). This looks nowhere close to “pretty much solved”. In fact, it’s an active danger to the rider and its surroundings.

In case people are interested, this video from March 16th gives a good idea of the current state of tesla self-driving: ▶ this video I'm not that knowledgeable about Tesla, from what I got they have a "Tesla Full Self Driving Beta" program (which has a terrible name as it isn't fully-autonomous and requires drivers to take over from time to time). Edit: I'm not sharing this video to say that their program is good or bad, I don't have a dog in this fight and personally do not care about the topic that much. I thought people here would be interested to see how their program behaves in a somewhat realistic scenario. The result seems to be mixed.

> The video is worth a watch: ▶ this video For contrast, here's Tesla's promotional video[1] for this feature and its caption: > The person in the driver’s seat is only there for legal reasons. He is not doing anything. The car is driving itself. It's been up on Tesla's site for almost 5 years now. [1] link

The video is worth a watch: ▶ this video I think it's fair to say that there are definitely some sketchy moments in there. The one that surprised me the most was marked "Near Collision (2)" in the video (12:25). A car really close to the Tesla doesn't seem to be picked up at all until it's too late: link You can see in the first frame there that on the screen there's nothing in front of the Tesla, it's just the usual white background. But right out the windshield we can see the car is already very close. At the bottom frame which is a split second later, the car is finally detected and we can see the screen now shows the car. But at that point the driver manually intervened with a sharp swerve right as they were about to hit the car. Going frame by frame on Youtube the green box indicating the car was only there for 4 frames before the driver disengaged. I was aware that self driving is far from ready but I didn't know that even basic detection stuff like this is buggy. Double parked cars, people walking in the street, etc, I would expect to be problematic. But this is not really an edge case, it's just a regular car at close range with no obstructions in between...