Tesla's AI Day Event Did A Great Job Convincing Me They're Wasting Everybody's Time

[HaulingAss]

However as an outsider at this point, I still can’t help thinking there is a reason you are told you have to pay attention to the road. As long as you don’t know what exactly that reason is, any moment you are not paying attention is a crapshoot. Even Munro on the same trip that he was enjoying the view mentioned that his Tesla scared the crap out of him because it tried to take an exit that was not there (or something like that). What I am saying is that as a hypothetical experiment, if everyone did not pay attention at all we would find out that out of every 10,000,000 hours of driving one minute of it could result in death. Obviously I pulled this number out of my …. but you get my drift. The way you drive your Tesla, there is a good chance statistically speaking, nothing will happen to you but if everyone drives their Tesla the way you do, at some point that minute is going to hit someone. So the convenience you are getting is from the added risk that may seem insignificant. The author is saying by flipping the system to be a guardian angel he rather trade the 10,000,000 hours of convenience with the life lost during that one minute of system failure. For most of the folks that don’t experience the death thing, it is a hard trade to make or even believe there is a trade. But what if someone told you it is not 10,000,000 hours and it is 10,000 hours. What if it was 10 hours. That is the thing. We don’t know and those that do only tell you to pay attention. You get many of those benefits by not listening. He is saying the system is pointless if you did really listen.

On one hand you say if you DID have experience using Autopilot you would probably come around to seeing it our way. But then you continue to pretend like it's not all it's cracked up to be. I would suggest it's simply because you don't have enough experience to know what you are talking about. Which is wierd because you seem to agree that you don't know while continuing to argue from a position of not knowing.

Here's a true short story that might help you understand:

In (what seems like) another life, I was a commercial fisherman on a 58' Alaskan salmon seiner. The owner of the vessel was a hardy, life-long fisherman, a high-liner for whom no job was too difficult or too big of a challenge. When the fleet started to upgrade their boats in increasing numbers to auto-pilots, he scoffed at other owners for adopting technology that had such limited functionality. Certainly, he could afford the best autopilots available, but how hard could it be to occasionally look at the compass and bump the steering wheel to maintain course? When towing the net in the middle of a set it was the same story, an occasional bump of the wheel kept our net properly shaped to catch the most fish. Obviously, autopilot was an expensive device of very limited functionality to a skilled, hard-working fisherman who knew how to steer a boat.

After an increasing number of other skippers began to rave about the benefits of AP he finally decided to install one while doing some other upgrades to his boat. He thought he was wasting his money, after all, it was simply not a problem to do what he had been doing his entire life.

But the next season, using the autopilot, he discovered he had almost unlimited energy. He said the auto pilot literally transformed his day, whether he was actively fishing or simply running to new fishing ground, he finished the day in a better mood and without fatigue. He felt like he could fish 24/7.

Although he didn't realize it previously, the simple mental act of maintaining a straight course was wearing him down. It required a lot of mental energy to constantly remember to steer a course even though he had been doing it without giving it a second thought. This was the type of man who would find such an admission embarrassing because that's what fisherman do and he was hardy and physically fit. But the benefits were so great and so obvious he became an outspoken advocate of autopilot (after having previously snickered about other fisherman who had wasted their money on a device of such limited utility). And all the AP on a boat does is maintain a set course. Tesla Autopilot does so much more, including maintaining a consistent following distance effortlessly.

You say there must be a reason why the driver must pay attention at all times. There is. It's because autopilot is not perfect and someone could die. The fact that you have to monitor AP at all times does not mean it doesn't greatly reduce the workload on the driver. Just as the fisherman in the story above discovered, a driver uses a lot of mental energy simply staying between the lines and maintaining a consistent following distance. This is true even though it seems easy. You don't know how much energy you are expending doing these constant chores until you become comfortable using autopilot and can compare experiences. Most people report it makes highway travel much more relaxing and they arrive feeling more refreshed.

 

[Ogre]

For what it’s worth Autopilot is the perfect name for it. Lots of people criticize the name, but it serves a very similar role to autopilot in an aircraft. It eliminates lots of human error doing the mundane parts of flying. Hold heading, hold altitude, let the pilot focus on situational awareness and not the mundane aspects of flight.

Autopilot in the car is the same. FSD might eventually be something else entirely. But that’s another day… or year or decade maybe.

 

[HaulingAss]

For what it’s worth Autopilot is the perfect name for it. Lots of people criticize the name, but it serves a very similar role to autopilot in an aircraft. It eliminates lots of human error doing the mundane parts of flying. Hold heading, hold altitude, let the pilot focus on situational awareness and not the mundane aspects of flight.

Autopilot in the car is the same. FSD might eventually be something else entirely. But that’s another day… or year or decade maybe.

To elaborate on that, the most common devices in the world called "autopilots" are of the marine variety. Most of them are consumer devices, used by regular folks who have recreational boats, bass boats, etc. There are millions of them in active use by regular consumers with no special training.

No one is fooled by the name autopilot. Everyone using these devices understands all they do is hold a pre-set course. They do not check for cross-traffic or slow down if approaching a slower moving boat. In fact, they won't even steer around it. They do not stop if the water get's too shallow of if they are headed for a rocky shoreline. No one thinks they can take a nap simply because it has the word "auto" in it.

Apparently, it's only a problem when Tesla uses the term.

 

[Crissa]

With Cruise Control, I can set a speed and know the car will hold it. I don't have to look at the dash except when it has an alert on it. I can focus on the road.

With blind spot monitoring, I have a second set of alerts if something moves into that space between when I checked my corner and when I moved.

Cross traffic monitoring can literally see where I cannot.

Traffic-aware cruise control takes the count of how far to the next car out of my head. I can check my corners more often, look further ahead. I don't have to be doing 'one mo-tor-cy-cle, two mo-tor-cy-cle' over and over with each passing stripe.

I don't have alot of use for auto-steer and even automatic transmissions - I still can do that better than they can. But they do a fine job making sure that I can stretch and not hit the limits.

They're all tools, and when they first work isn't the same as when they're better than I am. And that's the same with FSD... Right now, it's just a package that gets you access to more fancy abilities of the car. Anything that's truly a safety advancement gets kicked down to Autopilot, but for the fancy things, like Summon and Park are in the Full Self Driving package.

But then again, I expect to have the Cybertruck for ten years. The odds are for it. I want it to grow as their tech does, and I want to be willing to pay for it. That's where my tech-payout will go, even if it is several magnitudes smaller than Elon's.

-Crissa

 

[CyberGus]

Torchinsky is correct that L2 autonomy is pointless for city driving. Which is why Tesla is not doing that.

As sold, Autopilot is intended for highway use only. It’s insufficient for city streets, and the manual says so. The only valid reason for urban usage is for AI training, which is what the beta-testers are doing.

Tesla will not market FSD for city streets until it is functioning at at least L3 if not L4.

 

[Ogre]

Autopilot is intended for highway use only. It’s insufficient for city streets, and the manual says so.

Hmm, I hadn’t read this.

I like it for certain kinds of city driving. Does need a fair bit of nannying, but it’s less work than manually doing stop and go.

 

[Crissa]

Many US city 'streets' are basically highways. Which is why they're so dangerous.

-Crissa

 

[HaulingAss]

As sold, Autopilot is intended for highway use only. It’s insufficient for city streets, and the manual says so. The only valid reason for urban usage is for AI training, which is what the beta-testers are doing.

A lot of people don't know this but Autopilot isn't sold as an upgrade package, it comes for free on all Teslas these days. While Autopilot is more advanced than other manufacturer's advanced driver aids, it's still just an advanced driver's aid (and always will be).

FSD is a product that is not completed yet but you can pre-order it at today's price to lock-in the current pre-order price, if you want. By doing this you get a few extra bells and whistles to use in the interim. The bells and whistles are not going to be worth the full purchase price for most people so people who buy now are primarily locking in the price. Because when it can actually drive autonomously, without supervision, it's going to be worth a LOT more than $10K (because it turns your car into an autonomous robot).

So, people who claim Tesla is using deceptive terminology by calling it "Full Self-Driving" when it can't drive itself are not understanding that just because Tesla allows car buyers to pre-order it to lock-in today's price does not mean the product is ready. It's like the crowd-funding projects where people pay early to fund the development and manufacture of a product that they might not get for a few years but they get a lower price than everyone else who waits for the product to actually be available before buying it.

There are a lot of people in the media who purposefully spread these misconceptions to make it seem like Tesla is doing something unsavory or deceptive. But I've never met a Tesla buyer who thought they were buying a car that could already drive itself. It's just the media that tries to spread these falsehoods and make it seem dangerous and misleading.

 

[DMC-81]

It’s time to stop polishing the turd that is Level 2 driver-assist systems and actually put real effort into developing systems that stop putting humans in the ridiculous, dangerous space of both driving and not driving.

Until we get this solved, just drive your damn car.


SOURCE: JALOPNIK

I started reading and asked myself what is with this guy, and then scrolled to the bottom and sure enough it was written by Jalopnik. Figures.

Here's the thing:
Every time I drive, I'm already surrounded by people who are driving, but not driving, and they are not driving Teslas. They are "driving" in between texts, scrolling, posting, and everything else on their 3 x 5 smartphone screen. Since the advent of these wonderful devices, I saw crashes increase, and more people driving around with a damaged or missing front or rear bumper. The more recent trend from these driving zombies is to stop 15 to 20 feet short of the vehicle in front at a traffic light. Not an ideal environment to drive in.

On the advent of AI:
So, Jalopnik genius, how do we "get this solved" without iterating and the benefit of real world driving data?

smh

 

[TruckElectric]

I started reading and asked myself what is with this guy, and then scrolled to the bottom and sure enough it was written by Jalopnik. Figures.

Here's the thing:
Every time I drive, I'm already surrounded by people who are driving, but not driving, and they are not driving Teslas. They are "driving" in between texts, scrolling, posting, and everything else on their 3 x 5 smartphone screen. Since the advent of these wonderful devices, I saw crashes increase, and more people driving around with a damaged or missing front or rear bumper. The more recent trend from these driving zombies is to stop 15 to 20 feet short of the vehicle in front at a traffic light. Not an ideal environment to drive in.

On the advent of AI:
So, Jalopnik genius, how do we "get this solved" without iterating and the benefit of real world driving data?

smh

FWIW, Jalopnik USED to be a decent site. Around 8 yrs ago I found the site and liked it. I would visit the site occasionally when I would think about it and then I guess I lost interest or forgot about it for a few yrs and when I revisited the site again it had noticeably changed and I didn't like it so I just quit visiting the site.

So recently out of curiosity I did a search and Reddit had a topic titled "What happened to Jalopnik?"
Some people explained that it had changed ownership and went downhill after that.

Oh well.......

 

[HaulingAss]

I started reading and asked myself what is with this guy, and then scrolled to the bottom and sure enough it was written by Jalopnik. Figures.

Here's the thing:
Every time I drive, I'm already surrounded by people who are driving, but not driving, and they are not driving Teslas. They are "driving" in between texts, scrolling, posting, and everything else on their 3 x 5 smartphone screen. Since the advent of these wonderful devices, I saw crashes increase, and more people driving around with a damaged or missing front or rear bumper. The more recent trend from these driving zombies is to stop 15 to 20 feet short of the vehicle in front at a traffic light. Not an ideal environment to drive in.

On the advent of AI:
So, Jalopnik genius, how do we "get this solved" without iterating and the benefit of real world driving data?

smh

Jalopnik is an irrelevant dinosaur. The data shows Tesla's driver aids (including Autopilot) already reduce the chance of an accident by a considerable margin. It can't be rationally denied that the active safety systems and Autopilot already make our roads significantly safer. Preventing the use of something that is already proven beyond any doubt to be saving lives would be a crime against humanity. Really. Because innocent people will die, be paralyzed for life, etc., who otherwise would have been fine and remained healthy.

Either the author is ignorant of the facts or it's just a thinly disguised hit piece destined for the dustbin of history.

 

[xodarap1]

Maybe, but by reading the comment section he has quite a few that agree.

To those that agree I would say.. then don't buy it and just drive your car. Simple.

 

[jerhenderson]

Tesla's AI Day Event Did A Great Job Convincing Me They're Wasting Everybody's Time

Level 2-assisted driving, especially in-city driving, is worse than useless. It's stupid.

By
Jason Torchinsky

Today 1:30PM

Tesla’s big AI Day event just happened, and I’ve already told you about the humanoid robot Elon Musk says Tesla will be developing. You’d think that would have been the most eye-roll-inducing thing to come out of the event, but, surprisingly, that’s not the case. The part of the presentation that actually made me the most baffled was near the beginning, a straightforward demonstration of Tesla “Full Self-Driving.” I’ll explain.

The part I’m talking about is a repeating loop of a sped-up daytime drive through a city environment using Tesla’s FSD, a drive that contains a good amount of complex and varied traffic situations, road markings, maneuvering, pedestrians, other cars—all the good stuff.

The Tesla performs the driving pretty flawlessly. Here, watch for yourself:


Now, technically, there’s a lot to be impressed by here— the car is doing an admirable job of navigating the environment. The more I watched it, though, the more I realized one very important point: this is a colossal waste of time.


Well, that’s not entirely fair: it’s a waste of time, talent, energy, and money.

I know that sounds harsh, and it’s not really entirely fair, I know. A lot of this research and development is extremely important for the future of self-driving vehicles, but the current implementation—and, from what I can tell, the plan moving ahead—is still focusing on the wrong things.

Here’s the root of the issue, and it’s not a technical problem. It’s the fundamental flaw of all these Level 2 driver-assist, full-attention required systems: what problem are they actually solving?

That segment of video was kind of maddening to watch because that’s an entirely mundane, unchallenging drive for any remotely decent, sober driver. I watched that car turn the wheel as the person in the driver’s seat had their hand right there, wheel spinning through their loose fingers, feet inches from those pedals, while all of this extremely advanced technology was doing something that the driver was not only fully capable of doing on their own, but was in the exact right position and mental state to actually be doing.


Screenshot: YouTube/Tesla

What’s being solved, here? The demonstration of FSD shown in the video is doing absolutely nothing the human driver couldn’t do, and doesn’t free the human to do anything else. Nothing’s being gained!

It would be like if Tesla designed a humanoid dishwashing robot that worked fundamentally differently than the dishwashing robots many of us have tucked under our kitchen counters.

The Tesla Dishwasher would stand over the sink, like a human, washing dishes with human-like hands, but for safety reasons you would have to stand behind it, your hands lightly holding the robot’s hands, like a pair of young lovers in their first apartment.


Screenshot: YouTube/Tesla

Normally, the robot does the job just fine, but there’s a chance it could get confused and fling a dish at a wall or person, so for safety you need to be watching it, and have your hands on the robot’s at all times.

If you don’t, it beeps a warning, and then stops, mid-wash.

Would you want a dishwasher like that? You’re not really washing the dishes yourself, sure, but you’re also not not washing them, either. That’s what FSD is.

Every time I saw the Tesla in that video make a gentle turn or come to a slow stop, all I could think is, buddy, just fucking drive your car! You’re right there. Just drive!

The effort being expended to make FSD better at doing what it does is fine, but it’s misguided. The place that effort needs to be expended for automated driving is in developing systems and procedures that allow the cars to safely get out of the way, without human intervention, when things go wrong.

Level 2 is a dead end. It’s useless. Well, maybe not entirely—I suppose on some long highway trips or stop-and-go very slow traffic it can be a useful assist, but it would all be better if the weak link, the part that causes problems—demanding that a human be ready to take over at any moment—was eliminated.

Tesla—and everyone else in this space—should be focusing efforts on the two main areas that could actually be made better by these systems: long, boring highway drives, and stop-and-go traffic. The situations where humans are most likely to be bad at paying attention and make foolish mistakes, or be fatigued or distracted.


Screenshot: YouTube/Tesla

The type of driving shown in the FSD video here, daytime short-trip city driving, is likely the least useful application for self-driving.

If we’re all collectively serious about wanting automated vehicles, the only sensible next step is to actually make them forgiving of human inattention, because that is the one thing you can guarantee will be a constant factor.

Level 5 drive-everywhere cars are a foolish goal. We don’t need them, and the effort it would take to develop them is vast. What’s needed are systems around Level 4, focusing on long highway trips and painful traffic jam situations, where the intervention of a human is never required.

This isn’t an easy task. The eventual answer may require infrastructure changesor remote human intervention to pull off properly, and hardcore autonomy/AI fetishists find those solutions unsexy. But who gives a shit what they think?

The solution to eliminating the need for immediate driver handoffs and being able to get a disabled or confused AV out of traffic and danger may also require robust car-to-car communication and cooperation between carmakers, which is also a huge challenge. But it needs to happen before any meaningful acceptance of AVs can happen.

Here’s the bottom line: if your AV only really works safely if there is someone in position to be potentially driving the whole time, it’s not solving the real problem.

Now, if you want to argue that Tesla and other L2 systems offer a safety advantage (I’m not convinced they necessarily do, but whatever) then I think there’s a way to leverage all of this impressive R&D and keep the safety benefits of these L2 systems. How? By doing it the opposite way we do it now.

What I mean is that there should be a role-reversal: if safety is the goal, then the human should be the one driving, with the AI watching, always alert, and ready to take over in an emergency.

In this inverse-L2 model, the car is still doing all the complex AI things it would be doing in a system like FSD, but it will only take over in situations where it sees that the human driver is not responding to a potential problem.

This guardian angel-type approach provides all of the safety advantages of what a good L2 system could provide, and, because it’s a computer, will always be attentive and ready to take over if needed.

Driver monitoring systems won’t be necessary, because the car won’t drive unless the human is actually driving. And, if they get distracted or don’t see a person or car, then the AI steps in to help.

All of this development can still be used! We just need to do it backwards, and treat the system as an advanced safety back-up driver system as opposed to a driver-doesn’t-have-to-pay-so-much-attention system.

Andrej Karpathy and Tesla’s AI team are incredibly smart and capable people. They’ve accomplished an incredible amount. Those powerful, pulsating, damp brains need to be directed to solving the problems that actually matter, not making the least-necessary type of automated driving better.

Once the handoff problem is solved, that will eliminate the need for flawed, trick-able driver monitoring systems, which will always be in an arms race with moron drivers who want to pretend they live in a different reality.

It’s time to stop polishing the turd that is Level 2 driver-assist systems and actually put real effort into developing systems that stop putting humans in the ridiculous, dangerous space of both driving and not driving.

Until we get this solved, just drive your damn car.


SOURCE: JALOPNIK

that's because Jalopnik doesn't have the mental capacity to understand what was shown.

 

[rodmacpherson]

Same kinda tech would be essential if you are sending ton of Robots to a new planet were they have to move around in unknown environment with other robots. There will be countless applications we can not think of

This here hits the nail squarely on the head. If SpaceX is going to try to set up a permanent base on Mars, (or NASA on the Moon, or ...) they will benefit greatly from robots that can navigate in that environment without GPS and HD mapping being set up first.
There will be a lot of building to do before it is anywhere near ready for permanent human habitation. Meanwhile back home on Earth we can also benefit from this tech in countless ways. I don't know that Tesla will be first to achieve level 5 autonomous cars, I actually expect Waymo to be able to expand into new cities and keep the Robo taxi crown for a while, but Tesla's solution when they get there will be much more flexible and able to deal with the unexpected in ways I don't think that Waymo will handle very well.

 

[fritter63]

Tesla's AI Day Event Did A Great Job Convincing Me They're Wasting Everybody's Time

Level 2-assisted driving, especially in-city driving, is worse than useless. It's stupid.

By
Jason Torchinsky



What’s being solved, here? The demonstration of FSD shown in the video is doing absolutely nothing the human driver couldn’t do, and doesn’t free the human to do anything else. Nothing’s being gained!

I quit reading right here......