Edge Case
@edgecase411 • 15,457 subscribers
When AI meets the real world. FSD tester and progress tracker, Builder of @CamCutApp dashcam manager and telemetry visualizer. Tesla fan
Shorts
Videos
The crash analysis below shows AI3 FSD v12.6.4 biggest safety weakness: late reaction. This video uses official dashcam crash footage synchronized with the vehicle’s official telemetry logs obtained from Tesla and visualized in CamCutApp. The telemetry shows FSD waiting about 0.67 seconds after the lead vehicle’s brake lights illuminate before it begins braking. From the moment the lead vehicle starts braking until Automatic Emergency Braking intervenes is about 1.5 seconds. During that time, the Tesla slows by only 4 mph, from 62 mph to 58 mph. By the time AEB takes over, only 132 feet remain before impact. There simply isn’t enough distance left to avoid the collision. Millions of AI3 vehicles are still running FSD v12.x. Situations like this require the driver to intervene before it becomes obvious that FSD isn’t braking quickly and aggressively enough, and that’s not always easy to judge in real time. Tesla today announced AI3 v14 Lite which includes “significantly improved safety.” When I get a chance to evaluate v14 Lite, this is exactly what I’ll be testing first. In my opinion, it’s the most important and critical improvement the system needs.
Edge Case140,091 просмотров • 19 дней назад
Well, I was fully expecting FSD to turn around, but it didn't listen to me when I was yelling at it “don't do it!” over and over again.
Edge Case658,421 просмотров • 3 месяцев назад
FSD v14.3.4 still fails the static object avoidance test. I’ve repeated this exact test on every major FSD release since V11. V13 finally learned to consistently avoid a larger box. But three years later, no version has successfully avoided the smaller box. This time, AJ Agarwal and I expanded the test by using two boxes of the same size but different colors, and we tested them both in direct sunlight and in the shade to see whether lighting or appearance made any difference. 95% of people in the poll believed FSD would avoid at least one of the boxes and I honestly thought V14.3.4 would be the first to do so. I guess V15 it is! Tesla AI Dashcam video edited and visualized with CamCutApp
Edge Case36,920 просмотров • 10 дней назад
Below is official crash footage of my dad being rear-ended while using FSD v12.6.4 on his HW3 Model 3 earlier this year. The telemetry overlay was created using the new crash visualization tools in CamCutApp v1.1. Unfortunately, this experience completely destroyed his trust in FSD, and it’s going to be difficult to earn that trust back. In this case, traffic ahead was slowing down normally. There was no sudden stop or hard braking. For reasons that aren’t entirely clear, FSD delayed braking until Automatic Emergency Braking (AEB) intervened and forced a much more aggressive stop. (Watch the autopilot status switch to “Aborting”) The driver behind failed to react in time and rear-ended my dad’s car. One of my biggest concerns with FSD v12 on HW3 is that its reaction timing could occasionally be inconsistent, making these situations difficult to supervise. Everything can appear normal until the system reacts a second or two too late. FSD v14 on AI4 has been a significant improvement, which is one of the main reasons I upgraded seven months ago. But i think it’s important not to forget that there are many people out there that are still using this outdated software and should be warned about its shortcomings. To my knowledge, CamCut is the first commercially available tool that can combine Tesla crash footage, telemetry logs, and official crash report data into a synchronized visualization of the entire event. The crash analysis tools are currently in beta. If you have Tesla crash footage you’d be willing to share to help improve the feature, I’d greatly appreciate it. Feel free to DM me.
Edge Case100,912 просмотров • 1 месяц назад
FSD V14.3.4 Hand held stop sign test 1 It’s been about 7 months since our last handheld stop sign test on V14, and for me this is probably one of the most fascinating FSD test results. Watching the car respond to tiny changes in body language almost feels like you’re communicating with a living creature. Stand perfectly still with the stop sign up? It won’t budge. Shift your weight slightly? It nudges forward, almost like it’s getting ready. Take a step toward the shoulder? It becomes more willing to move. Then I gave it conflicting signals: a STOP sign in one hand while waving it through with the other. You could just see the neurons firing. If you’ve ever wondered how FSD interprets the world at the smallest scale I think you’ll really enjoy this series with me and AJ Agarwal Telemetry visualized with CamCutApp
Edge Case24,475 просмотров • 9 дней назад
Another user Carnasis submitted this crash footage of his Tesla being T-boned while FSD was engaged. Raw video analyzed using CamCutApp telemetry visualization. The other driver was ultimately found at fault after entering the roadway and colliding with the Tesla while FSD v14.2.2.5 was engaged. What makes this crash interesting is that there appear to be multiple contributing factors: • The driver was pressing the accelerator before the collision, causing the vehicle to accelerate to roughly 40 mph. Had the throttle not been overridden, FSD may have approached the area at a lower speed. • A portion of the striking vehicle was visible between the stopped cars before it fully emerged into the roadway. FSD did not react until the vehicle entered its path, leaving only a few hundred milliseconds to respond. By the time emergency braking began, there simply wasn’t enough distance left to avoid the collision. The other driver caused the crash, but it’s an interesting example of how throttle overrides, speed, and anticipation of partially occluded vehicles can all influence the outcome. Tesla AI
Edge Case62,489 просмотров • 1 месяц назад
Dangerous left turn handled exceptionally well by FSD v12.3 If you live in LA you will know that these unprotected forward facing left turns are the most dangerous part of driving here. And they are all over the city. It is obvious that the Tesla AI team has spent a good amount of effort into training the FSD v12 system to handle these with caution but also confidence. v11 wouldn’t even dare go through that. Elon Musk Tesla AI Tezmeeinc
Edge Case909,700 просмотров • 2 лет назад
Dear Tesla AI team let’s make lane centering non-human again Sincerely, FSD 12.6.1
Edge Case561,767 просмотров • 1 год назад
Do you think FSD took the cue from the high beams to proceed?
Edge Case15,335 просмотров • 10 дней назад
So So So Good! Whoever worked on the lane change neural nets at Tesla deserves a big promotion. Watch how safely FSD Beta 11.4.2 changes lanes while being mindful of cars and trucks from two lanes over merging at the same time, while also keeping flow of traffic, and finally taking the safest opportunity to complete the navigation. Please share this with loved ones who have fear and anxiety from driving. FSD will save many lives. Kaz Meta4 TeslaTrip ValueAnalyst Dirty Tesla Ashok Elluswamy Elon Musk
Edge Case561,372 просмотров • 3 лет назад
This is what a very capable robotaxi looks like. First, FSD 14.3.2 notices the stopped bus blocking the lane, so it signals far in advance and moves over. Then, as we get closer, the bus starts pulling out, and FSD quickly anticipates that it will encroach into our lane, so it safely brakes to let it in. Smooth operator.
Edge Case35,215 просмотров • 2 месяцев назад
