News
SpaceX CEO Elon Musk talks Starship explosion: “We were too dumb”
Two days after a last-second failure caused Starship SN9 to smash into the ground and explode, SpaceX CEO Elon Musk has returned to Twitter with some harsh preliminary reactions.
Right off the bat, in response to a question about why Starships SN8 and SN9 both attempted their unsuccessful landings with only two of three available Raptor engines, Musk frankly stated that “we were too dumb.” At face value, it’s a decent question, given that there are no obvious showstoppers to explain why Starships couldn’t make the most of the redundancy their three Raptor engines can offer.
After completing an otherwise flawless 6.5 minutes launch, ascent, and belly-flop descent, Starship SN9 began a critical ~120-degree flip maneuver, sequentially igniting two Raptor engines and using that thrust to flip from a belly-down attitude to a tail-first landing configuration. Unfortunately, though the first Raptor did fire up and put in a good effort, the second engine failed to ignite, leaving the building-sized rocket to impact the ground traveling far too quickly.
Ironically, more than three years ago, Musk himself revealed in a Reddit Ask Me Anything thread that he and his engineers had decided to modify Starship’s (then known as BFS) design by adding a third Raptor to its central cluster of two engines.
“Btw, we modified the [Starship] design since IAC [2017] to add a third medium-area-ratio Raptor engine partly for that reason (lose only 1/3 thrust in engine out) and allow landings with higher payload mass for the Earth to Earth transport function.”
Elon Musk – Reddit AMA – October 2017
Primarily meant to enable more efficient landings in Earth’s atmosphere, adding a third engine to that cluster would logically increase the chances of a successful (or at least survivable) landing in the event that one engine fails. Greater thrust and an improved thrust-to-weight ratio both during launch and landing would fundamentally improve the efficiency of Starship, likely making up for most or all of the added weight.
In retrospect, it’s not entirely surprising to learn that a three-engine landing burn is probably the most logical option if three landing-class engines have been included in the design. In SpaceX and Musk’s defense, however, there are also several good reasons to use as few Raptor engines as possible.
It was foolish of us not to start 3 engines & immediately shut down 1, as 2 are needed to land— Elon Musk (@elonmusk) February 4, 2021
Throttling high-performance rocket engines is exceptionally difficult and Raptor is not yet a fully mature engine, meaning that it’s throttle capabilities are likely less than optimal. That’s relevant because the higher a rocket’s thrust-to-weight ratio during landing, the more aggressive its landings have to be. SpaceX is apparently extremely conservative with Starship in this regard, prioritizing slow, gentle landings by only using two of three available engines.
-
Ironically, it’s possible that that attempt at risk reduction resulted in harder landings for both Starship SN8 and SN9, as three-engine landing burns could have potentially slowed them down significantly more before impact.
At the same time, though it may have mitigated the severity of both landing failures, three-engine landing burns would not have resolved the fundamental issues that caused them. In SN8’s case, low fuel header tank pressure doomed the Starship, while SN9 is more ambiguous. Aside from the clear Raptor ignition failure, which a three-engine burn could have resolved by downselecting to two healthier engines, the one Raptor that did ignite appeared to suffer some kind of uncontained failure seconds before landing.
Impressively, despite that apparent combustion chamber or preburner failure, the engine’s landing burn seemed to continued uninterrupted until the moment of impact. As such, it’s hard to say if that lone Raptor was still producing substantial thrust or if it was in the throes of a catastrophic failure. If it could have held on for another 5-10 seconds and the third Raptor (the engine that didn’t reignite) was able to restart and perform without issue, a three-engine landing burn could have easily made SN9’s demise less violent or even have enabled a soft landing.
While a three-engine burn all the way to touchdown appears to be extremely risky or impossible for present-day Starships, Musk implied that there was nothing preventing SpaceX from reigniting all three engines during the initial flip and landing burn and using that time to determine the health of all three engines. If all three were healthy, Starship would shut down one for a soft landing. If one engine failed to restart or lost thrust shortly after ignition, the other two would already be active and able to take over.
Musk says that Starship SN10, already at the launch pad and likely days away from its first tests, will attempt to adopt that approach on an upcoming test flight expected as few as 2-3 weeks from now.
Tesla briefly offered one Robotaxi part in its Parts Catalog for your personal car, only to remove it just a short time after it was first noticed.
Tesla’s Robotaxi camera washer apparatus was briefly available for purchase on the company’s Online Parts Catalog. The camera washer was first noticed on Model Y Robotaxi vehicles about six months ago in Austin.
🚨 Tesla looks to have installed Camera Washers on the side repeater cameras on Robotaxis in Austin
pic.twitter.com/xemRtDtlRR— TESLARATI (@Teslarati) January 23, 2026
First noticed by Not a Tesla App, the Camera Washer entries appeared for the new “Juniper” Model Y under a category called “Halo,” which has also now disappeared. Interestingly, Halo probably is related to Tesla’s internal “Project Halo,” which was a project that aimed to retrofit customer-owned Model Ys into functional Robotaxis.
This hardware addition would likely be required for the vehicle to operate as a Robotaxi, as the Camera Washer seems to be a non-negotiable part of the vision-based system Tesla utilizes for self-driving efforts.
However, this part has since been removed and is no longer visible on the EPC.
Now the true question lingers: Why would Tesla add this Camera Washer to the Model Y parts catalog? Is it planning to make it available for owners to utilize on their own cars for personal use, or will it become a prerequisite for Robotaxi operation in customer-owned cars?
While discussing the upgrade options for Hardware 3 vehicles during the Q1 Earnings Call, Tesla CEO Elon Musk had said that the company could establish small, satellite shops that would upgrade cameras and self-driving computers. Perhaps this same strategy could be utilized for vehicles that want to be included in Robotaxi but do not have the correct hardware.
AI4 is currently represented as capable of unsupervised self-driving, and the same was said about HW3 at one point, only for Tesla to admit last quarter that it would, unfortunately, not be possible. Perhaps AI4 vehicles might need this camera washer as a prerequisite, just as HW3 cars will need that camera and computer upgrade.
This could be the first hint of that’s where we are headed.
Tesla has started rolling out a broader change to the ride experience for its Robotaxi fleet by silencing turn signals, but the change is certainly polarizing.
Tesla has generally made it clear that its purpose-built ride-hailing platform, Robotaxi, will cater to the rider in nearly every way possible. This includes having climate preferences, music, and other personal settings loaded up in the car as the rider enters.
But Tesla is taking it a step further by muting turn signal chimes altogether, a change that appears to be a way to make the ride more peaceful:
Tesla has finally turned off the blinker/turn signal noises in their Model Y Robotaxis.
This is a small update, but will definitely be noticeable for riders. It didn’t really make much sense to leave the turn signal sound on when it’s Unsupervised. pic.twitter.com/ITwx60PAMO
— Sawyer Merritt (@SawyerMerritt) July 27, 2026
However, there are a handful of people who are not thrilled about this change. Turn signals are a conditioned part of the human mind for those who ride in a car regularly.
Taking a turn without one feels strange and odd, and not hearing it click while activated could set off some alarms for riders, who might use the noise as confirmation that other drivers know of their intention to turn.
I kind of like it being on, threw me off when riding with Zoox who has had it disabled
— Dan Burkland (@DBurkland) July 27, 2026
Turn signal noises are still audible in customer cars, so if you use FSD in your personal vehicle, you will still hear the turn signal.
The move is certainly one that is unique, but not one that separates it from other ride-sharing services. In a normal car, the clicking sound confirms to the driver that the blinker is active. In a fully driverless Robotaxi, that feedback serves no purpose for passengers, other than peace of mind.
Tesla released Full Self-Driving version 14.3.6 last week, and after what was potentially one of the best FSD releases in v14.3.5, there has been a bit of a regression. While there are some bright spots, the changes made to v14.3.6 seem to have backtracked some behaviors.
Overall, it is hard to really complain about FSD in any sense; it has revolutionized how I travel literally anywhere. According to my self-driving app, the last time I went a day without using it was 59 days ago.
However, I think it’s also important to recognize when things are just plain bad with FSD. There are times it does truly mind-boggling things, and I’ll dive into those here. Additionally, I only had these issues on local roads, not on highways. Highway operation, generally, is always incredible other than the occasional complaint about speed or left lane camping.
With those things being said, my personal experience may not represent others’ experiences. A handful of people have said they have had a similar experience on v14.3.6, while others have said it is more than normal.
Turning Hesitancy, Inaccuracy
I’ve noticed more inaccuracy turning into multi-lane stretches of road than in any version I can remember. I’ve had at least three instances of FSD turning into a stretch of roadway that has two or more lanes, and not selecting a lane confidently as it has in past versions.
I think Tesla FSD v14.3.6 is a regression from previous .3 branches
This version in particular has been incredibly hesitant, jerky, and indecisive at times. There’s actually been two drives that I have decided to not used FSD for the remainder of the trip. These were local trips… pic.twitter.com/YobHDOKRS6
— TESLARATI (@Teslarati) July 26, 2026
Instead, the car will drive over one of the dashed road lines, and the steering wheel will jerk back and forth before picking the lane. It should be said that it has always picked the correct lane when choosing based on the navigation, but it is still very indecisive. The steering wheel jerking is reminiscent of some of the later versions of v13.
I admit I really hate to see the steering wheel jerking come back. However, I think when Tesla releases v14.3.7, it won’t be present. When there are occurrences of it in FSD versions, it is usually resolved by the following release.
FSD Disregards Manual Turn Signals
This is my biggest bone to pick with FSD other than Navigation issues, but this one seems like it would be such an easy fix.
If Tesla is going to put the word “Supervised” on the end of “Full Self-Driving,” then when I tell the car to do something, it should do it. If I input an increase in speed by pressing the accelerator, the car will immediately respond. It does not disregard my input because it feels it is traveling at the right speed.
FSD should never disobey and turn off turn signals that the driver inputs. Trying to direct the car into the correct lane, I had initiated the left turn signal not once, not twice, but three times, with the car turning it off all three times and continuing in a lane that would end in just one block. The only solution at this point would be to zipper merge.
This goes back to the fact that self-driving’s biggest bottleneck might be rider preference. A zipper merge might have been more than reasonable, might have saved me time that I spent sitting through an additional light cycle, and might be something many drivers would do. I was in no hurry, I traditionally do not try to zipper merge because it feels inconsiderate, and lastly, the car should have just followed my input.
This caused me to disengage and drive manually the rest of the way home. Sometimes I just do not need FSD to try to pass every car it can at intersections.
Bird Braking is a Thing of the Past
The big complaint with recent versions of Full Self-Driving has been what we’ve coined as “bird braking,” which is when the car will brake suddenly as a bird flies past.
There have been zero issues with this so far in v14.3.6, which is an excellent improvement.
FSD Might Already Be Taking Note of Driver Preferences
Another thing I have noticed over the past few days is that v14.3.6 seems to already be taking my preferences with navigation into account.
This is something that is supposed to be rolling out with the Summer Update, but I have a hunch it’s already present and might have been included in this v14.3.6 build. On Friday, FSD pulled into an entrance to a local convenience store that it had never attempted to go into before.
Typically, I manually pull into this entrance because it avoids heavy cross traffic at the main entrance. FSD has always chosen that congested main entrance.
I don’t want to get anyone excited but my Tesla running FSD v14.3.6 just:
✅ Pulled into an entrance at my local Sheetz that I routinely pull into manually but FSD never has entered
✅ Pulled into my assigned parking space at home despite many other spots being vacant https://t.co/ZUShvknTWO pic.twitter.com/83O5a2S0eh
— TESLARATI (@Teslarati) July 24, 2026
Additionally, FSD has pulled into my assigned parking spot at my townhouse community on multiple occasions with this release. This is something that used to happen ocassionally, but not consistently.
It also navigated back to the same convenience store last night, drove through crazy cars scrambling to gas pumps, navigated out of the parking lot correctly, drove me home, and, once again, parked in my assigned spot.
As previously stated, this release just seems to have a few things that need to be brought to Tesla’s attention, and also to make others who use FSD aware of some things that I’ve experienced. I look forward to the next release that will remedy these issues, just as Tesla has always done in the past.
Tesla briefly offered this Robotaxi part for your personal car
Tesla Robotaxi gets sweeping but polarizing change
