Elon Musk UNLEASHES AI Powerhouse: The Dojo Supercomputer will SHOCK the Industry

Elon Musk reiterated at the recent Vivatech conference that he thinks self-driving cars will soon be solved and that the value of Tesla is primarily driven on this basis of autonomy. Tesla has multiple irons in the fire, each with exponential growth curves that could converge next year in 2024. Their electric vehicle fleet will double over the next year and a half, more people are using FSD which is causing user miles to skyrocket, Tesla's AI is advancing rapidly, and Tesa's Dojo Supercomputer will achieve 100 exaflops of compute, more than 40 times where Tesla is today with their NVidia graphics car based approach. Therefore with all of these major windfalls coming to fruition, Elon Musk may finally be right about solving Full Self Driving. #tesla #ElonMusk #AI

Disclaimer: We are not financial advisers, and nothing on this channel is meant to be financial advice. The ideas expressed on this channel are purely opinions and should not be regarded as objective information. Nothing on this channel is a recommendation to buy or sell securities. Do not assume that facts and numbers in any video are accurate. Always do your own due diligence.

[TMIOTesla]

Do you think Elon Musk will finally be right about Full Self Driving in 2024, given that Tesla's AI, Dojo, and EV exponential curves are all coming together next year? Be sure to watch:

[keylime]

Don’t agree with everything Elon does, but he is a genius with good intentions. I’m amazed that he has the ability to discus vision, roadmap, roadblocks and technical details across so many different industries.

[Dogman]

Mr Dojo is going to be huge revue stream in the up coming years! Great planning by Elon and the Tesla team.

[michaelleffler]

History has demonstrated that Elon has been correct time and time again on his outrageous predictions. He has completely changed both the paradigm and narrative in every industry he has attempted to change (for the better). At what point and time do we, as a community, resonate with Elon and his endeavors which appear to benefit everyone, even those who bet against him.

[halohat]

Tesla and Space X have some of the brightest minds, I wouldn't bet against them. 😊

[theknifedude]

Elon may run a little late on his predictions but IMHO if he says it it’s going to happen, it will happen.

[jimbert]

Exa flops can be misleading when applied to Dojo. Since the neural networks do not always need the high precision of the standard IEEE floating point standard, Dojo supports multiple floating point formats with reduced precision to speed up those operations.

[SpaceXploits]

At this point, it should be apparent that Tesla is the only company that can solve autonomy, here are a few facts to back my observations,
1. an exponentially growing fleet
2. a powerful dojo computer
3. Super Talented AI team

With all these boxes checked and no other companies having even a fraction of the capabilities of Tesla, it is only a matter of time.

[healingbyqurannow]

We should call this age the age of Elon Musk.
May God protect him, guide him, abd use him for the defense of humanity from the evil politicians.

[hardyvonwinterstein]

Very nice video essay. Logical and controllable. I would personally ditch the UNLEASHES and SHOCK from the title, but hey you can't have it all.

[alexanderbanman]

Another great video — I'm addicted to your channel lol!

[brunosmith]

The fact that Tesla FSD (both in hardware and software) will result in autonomous TaaS, is now certain. As is highlighted here, the "march of nines" is the determinant as to just when such vehicles will be viable. To me, the timeline is not really relevant, because it will happen - it's inevitable.
When it does happen, the disruption will be nuclear. Not only will the need to buy a car - or even own a car - evaporate for an exponentially growing number of people, resulting in a swift demise of non-FSD vehicles (EV or not), but industry sectors allied to the traditional automotive industry will collapse too.
I live in one of England's most remote market towns, with a history going back to Roman times and further. In the late 1800's, this tiny town had four farriers, several livery stables and even a manufacturer of equestrian accessories (saddles, harnesses, bridles). As is the case with modern automotive aftermarket activities, the equestrian industry was also served by suppliers to the trade (horse-feeds, vetinarian, carriage repair shops, clothing...) By 1920 all these industries had shut down when the petrol-engined car became ubiquitous. It is estimated that over 20% of the town's economically active residents lost their livelihoods.
My example can be extrapolated across every community on the planet where cars usurped the horse. This disruption took around 30 years, but it was a disruption nevertheless.
If we examine the number of commercial and industrial sectors that are wholly (or even partly) reliant on the traditional automotive industry, what the car did to the horse a century ago will be chicken-feed in comparison.

[kennethausten]

Another thought. One day an autonomous car will drive from one end of UK to the other. That would be a worth while exercise.

[jonesingforfood]

Good summary of what is going on with Tesla FSD progress. I will be sharing this video with others. FSD Alpha is coming in under a year if they ramp Dojo as planned. So exciting!

[bigpicture]

It is an interesting layered set of challenges. There is the unchanging (things that don't move) background, and the vehicles "now" position in relation to that background. Then there are the (things that move) "now" situational information, and the probability / predictability of all the next movements of the things that can move. Then there is the actual physical road layout in that area, and the applicable road rules that may apply in that jurisdiction. Things that don't change moment to moment can be pre-programed, but the dynamically changing things need to be accurately predicted (instantly calculated) and reacted to. Even the challenge of realistically how much compute power is required, and how much of that can be on board and what information is it realistic to dynamically bi-transmit to a super computer, given required quantities of data and reaction times..

[ChickensAndGardening]

FSD as it works today is probably significantly better than a human drunk driver.

[jesgo]

Sell everything in your portfo & go all in on TSLA ⚡Tesla will only gain going forward ⬆

[nickwhite-zzvf]

Energy decision is way way undervalued also. Every solar project going back 2-3 years is mandated to have 4 hour back up here. Tesla can easily play on the synergy coming out of Jeff Dunns lab to absolutely crush cost curve on stationary products

[TheKerryswan]

do you think Energy generation and storage revenue could be 1/10th vehicle revenue end of 2026with about 26 billion in revenue for the year with 260 billion revenue in car with 6 million production and average selling price of about 43 k

[rhiantaylor]

Tesla - making Moore's law look glacial by comparison...