Elon Musk answers all your questions at Tesla's Shareholder Meeting

uh okay so uh when will savage truck production begin and at what rate will the ramp up happen um well um so the this year has been just a a constant struggle with with part supply so just to be clear if if we had like five extra products we would not change our vehicle output at all uh because we were just basically limited by multiple supply chain shortages like so many supply chain of so many types not just chips there were lots of supply chain shortages um so the i so it really wouldn't matter if we had like the semi or have a truck or anything we would not be able to make it um and the semi in particular needs a lot of sales so it needs a lot of sales a lot of chips and so uh that will be uh we gotta have enough otherwise it's pointless so i think most likely what we'll see is cybertruck stock production and next year and then reach volume production in 2023 and hopefully we can also be producing the semi and the the new roadster in 23 as well so we should be through our severe supply chain shortages in 23 i'm optimistic that will be the case let's see will we do a stock split um well we maybe not yet um maybe we'll consider a stock split at some point in the future but i think we'll not quite do that quite yet uh well we see 46.8 battery cell production in texas um i don't think we will see 4680 production in texas this year but we are making 4680s in california at a our pilot plant which is a big plant by normal standards it's like a capable of 10 gigawatt hours a year so um it's just it's just a mile away from our vehicle factory in fremont so it's um that basically that that factory will be able to make more than enough cells for uh gigatexas to to scale production of model y and i do want to emphasize like the from the point at which a factory is able to start making cars to where it reaches high volume production is typically about a year and that's considered very fast so it's uh it it takes longer or at least in tesla land it takes longer to build the factory than it does to reach volume production once the factory is built um so like uh in shanghai we built the factory in 11 months but to get to high volume production took about a year um and so i expect something similar here you know we'll we'll start production uh this year and we'll deliver a comfortable deliver some cars from gigatexas this year model-wise but uh we won't reach high volume production until probably the end of next year um so but then i also expect we will reach high volume production the 4680 sell here in giga texas next year as well more factories man it's hard to build a factory um you know i've said many times that prototypes are easy production is hard uh like you know or whatever it's like one percent inspiration 99 perspiration but i think for high volume production it's like 99 percent perspiration um so uh i think we might start scouting for locations next year but i think we we can do a lot with um berlin and um you know berlin and austin and expanding in china uh you know so and expanding in fremont so um the nice thing is like having at least a factory in europe and a factory in china factory in north america we at least have um factories where for a high volume products where uh they're on the continent where most of our customers are this is great so we do not like one of the biggest challenges we had in q3 was can we get enough ships like there was a huge ship shortage um so that was that was a challenge and that boat that got stuck in the panama i don't mean the suez canal um of course havoc so you know these things you don't expect these things um but they happen um so i know we'll probably get to new factory locations let's investigate them next year maybe make a decision um in 23. does tesla plan to offer dividends uh we don't currently have plans for dividends um i do kind of feel like the time when a company offers dividends is you know it's kind of cresting the hill you know um if you will it's not usually not they've run out of things to invest internally you know we have not run out of things to invest in internally by a long shot can we provide a quarterly update on energy along with automotive um yeah i mean we we need to make more progress on energy um the you know the yeah this year has been limited like i said that the chip shortage the same chips go into our stationary battery packs as in the cars and we prioritize the cars so um you know and we need the inverters for solar and and that kind of thing so um you know this year has not been a good guide for progress and energy i think next year it will be probably a good guide and we can start providing uh updates that i think people would interpret correctly as opposed to you know having to explain well we either had to short cars or um or energy and we picked cars so all right yeah model 2 is not a car there's no no model two three means e um so we're we're model three we try to do get the we're going to call it the model e but then four threatened to sue us and so we said what's called the model 3 you know so it's s3 xy instead of you know anyway uh gigafactory is getting bigger with each iteration um not not all gigafactories will get bigger with each iteration i think uh but uh yeah most of them will they'll get more advanced and more efficient and you're looking at sort of taking a first principles approach to manufacturing you want to look at the volumetric efficiency of a factory and the the average speed of a factory is kind of like the the rocket equation um where delta v is like exhaust velocity times the log of the start to end mass ratios um so i think it's something similar to that for a factory um and you could also think of it like a chip um like do you make chips um get more out of chips by making them bigger or by bringing things closer together and increasing the clock speed and actually i think there's tremendous opportunity in factories to actually bring things closer together and increase the the cycle time um or clock speed um and this actually i think tremendous opportunity to improve the efficiency of factories and tesla's long-term competitive advantage will be manufacturing because all cars will be electric eventually all cars will be autonomous the thing that's the hardest to i think sort of match tesla on or copy is manufacturing and so i think that will be the long-term advantage of tesla uh yeah limiting factory growth is um engineering yeah there's there's really not a like factory stamping out amazing engineers so it's hard to you know work you've got to basically recruit people you got to kind of train them in the right way and this just takes time and uh it like sometimes companies think you could just hire anyone with an engineering degree slam them together and get amazing stuff this is not the case at all um often adding more engineering engineers to a program uh makes it go slower so um you know and and the excellence of an engineer matters tremendously there are the huge differences in engineering talent uh that are all goals in mind but i do wonder today like if nikola tesla applied to tesla would we even give him an interview i'm not sure we would you know it's like it was like you know we should obviously um so i you know that does concern me like i think we could do a better job of um of vetting resumes and really we're just looking for evidence of exceptional ability like not whether somebody graduated from a particular school or whatever um but like just three bullet points evidence of exceptional ability and and do you say wow if you read those three bullet points then great that should be the approach okay um i i'm actually a big fan of the original uh vw sort of bust that was really cool um and uh i haven't really seen the new one but um yeah i mean i think over time tesla will make basically you know all major variants a vehicle um why not you know um one in every significant category i think uh yeah all right uh so so from i guess people can see these slides right so i need to read the question um yeah i think we're actually making uh rapid progress with solar roof um you know um solar so solar in general and energy in general got kind of uh short change for a couple years there because we're desperately trying to make the model 3 and reach volume production so we just moved everyone from anywhere in the company that could possibly work on model 3 was working on model 3. it was all hands on deck as you know so there's just a couple years where we just you know just if we got to save if model 3 fails the whole company's dead so we're just going to move everyone from solo from everywhere to help with model 3. um so we're a couple years behind on that but i think we're now making rapid progress with the solar roof and in particular with new home builders obviously the most efficient way to get a solar roof is to put it on put a roof on once um instead of like doing retrofit is obviously fundamentally more expensive than just than doing solar roof on new homes so um we now have relationships with a number of new home builders and um and we expect that those to to gather significant momentum um and um and then we're gonna have a campaign which is uh kind of like well i'm dating myself here i want my mtv um so i don't want my soul or a roof where is it so um you know i think we'll just get people actively asking in your home developments well does it have a solar roof and a power wall and you know because then you know you generally you you generate your own power and you can help stabilize the rest of the grid using the power wall we're doing this in partnership with utilities like the thing that i think um people need to appreciate is that there's a lot more electricity production that needs to occur as we move to an electric vehicle future for for two homes where both cars are electric um the power electricity usage will approximately double and so if you don't have local power production at the homes with a battery because the battery's got to buffer the power um otherwise you if if you just have so if you just have solar without batteries it causes these massive waves in the grid um uh and and so you have to have the batteries to stabilize the the grid but those batteries can then act collectively to stabilize the whole grid um and that's software that that we've developed that is um i think of course going to go into houston australia um that's going to be i think quite compelling but you really need for a sustainable energy future you have to address electricity at the at the homeowner level this is essential but there will still be very prosperous future for utilities because electric power will approximately double and then if you transition heating to electric as well it approximately triples so this is a lot of work both at the local level of electricity distribution and at the utilities utilities better expand and there's got to be solar and batteries at the homeowner level but having a tesla solar with with battery means that you have power no matter what if the utility shuts you off or it is for any reason you still have power um when i was here in the in the austin blizzard it's like i was just staying at a friend's house but you didn't have the solar roof and battery so i'm sitting in the dark in the cold and i was like okay it really really brings home the need for the solar roof and battery um so anyway so i think that's uh i'm gonna see i think solar roof will see massive growth in the years to come uh yeah the 10.2 is going out tomorrow night yeah anyway this is too much inside baseball on fsd but the um we're going to start rolling out the fst beta to all customers with a perfect safety driving record tomorrow night at midnight california time and we'll see how that goes and we want to be very cautious about the rollout um if that's looking good then we'll because there's like a little over a thousand people that have a perfect score then we'll you know start giving it to people with the 99 90 98 97 and so forth um we'll probably have to have some cut off you know it's like 60 i don't know i don't know but it's got to be um well we'll think of people who are extremely conscientious drivers for the beta program so but it's looking really good um i actually drove here from from at the friend's house i'm staying at in austin which is quite far away and has quite a complicated drive um and the car took me all the way from um my friend's house to the gigafactory with no interventions so perfect drive yeah [Applause] all right is that are those all the questions oh i guess we'll take some from yeah audience uh sure or you can just yell or whatever oh when will we not need to do any more mining for batteries um basically it'll be a while because we've got to um extend the fleet like so the fleet of cars out there is gigantic there's about two billion cars and trucks uh in use in the world um and uh so uh that's that's a lot so um uh annual global capacity for vehicles is 100 million a year roughly um which kind of makes sense like cars and trucks tend to last about 20 years before they go to the junkyard so so that's important important statistic to bear in mind like even if all cars were and all vehicles were electric tomorrow it would still take 20 years to um change out the fleet um so i think people don't like you see a lot of noise and sort of stories about electric vehicles but the fleet is what matters um and and still um it's basically rounds down to nothing if you look at total fleet vehicles on earth and electric cars at this point i think are still well under one percent of the fleet so but i don't know uh probably 30 years ish 34 years yeah so bad yeah um yeah and i think um clarify a few things that are sometimes misunderstood out there um lithium is extremely plentiful it is one of the most plentiful things on earth it is not rare at all it's hard to avoid lithium if you said where can we dig a bunch of rock that doesn't have some lithium in it you would have a hard time um it's in sea water in you know it's it's basically let's see let him as a salt so where is there salt a little bit of salt pretty much everywhere so it's lithium everywhere um so that's not an issue also lithium is only like maybe one or two percent of the cell um and so the actual thing that matters is the cathode um and uh i it um most of our well actually our our higher a long-range vehicles use a nickel based cathode and uh you know sometimes people think it's a cobalt-based cathode no cobalt is used in phones and laptops but uh we use nickel because nikolas higher energy density for our long range vehicles but for our standard range vehicles and for stationary storage i think all of that will move to iron-based iron cathodes and iron is also extremely plentiful uh nick that nickel is not rare but uh there's uh i know somewhere maybe 10 to 100 times more in iron than there is nickel so um so so moving to an iron uh based chemistry um which is sort of finally at the point where it's competitive on range when combined with uh inefficient powertrain uh i think that will be the the vast majority of batteries in the future will be iron based so i do not see any shortages um it's just a question of making all the equipment to kind of process that into a cell and then into a pack oh of planet factory i like the way you think so how many years before tesla's first off-planet factory i mean i mean i'd like to see one before i'm dead that would be cool um so i don't know if whatever like 40 years-ish hopefully before dead basically that'd be great what's my safety school oh great question um i don't know actually um yeah i don't know what it is i'll check uh i'll check because i think it's uh mine just got turned on um yeah i'll find out by the way a safety score calculation is obviously imperfect that's why i try to emphasize very much that it is beta if not alpha in in safety score calculation um so it's going to get a lot of changes uh yeah expect it to improve improving its accuracy substantially over time this is really just it's a very early stage algorithm oh is there anything we do with a grid system here uh yeah um in fact uh tesla is doing um a number of projects in tesla in texas tesla texas um the uh here we've got a a large battery project um i believe in the houston area yeah it's houston scenario um sorry yeah we're doing a hundred megawatts in the houston area um and um we're talking to erica about uh doing uh other major installations um the the tesla um mega pack system is actually really great at dealing with power fluctuations so if you look at uh power usage uh it's it varies dramatically during the day so you get um you know uh particularly on on a summer day or a winter winter day if you've got electric heating um or electric cooling um then you're going to see big big spikes on hot or cold days um and so so just even during the day you see the big big variances uh and then uh uh and then from from when like you know in order for for if a power grid has no no uh sort of energy storage buffer then it's it's as resilient as the worst second of the worst day of the worst year so this is not that's obviously not great uh and you they sort of they'll throw peter plants in there but if the peaker plant is is reliant on natural gas and you and then people start using natural gas for heating which is kind of what happened in in austin people were using the gas to heat then the peaker plants didn't have the gas like uh okay but batteries would work great so um if there'd been tesla mega packs here during the blizzard the power would not have gone out i targeted capacity for the the mega pack factory um no i don't actually know what the target capacity is but it's vague okay 40 gigahertz um yeah um actually that's an example of where we are expanding also where we're expanding manufacturing in california so this is in later california we just opened up a mega pack production facility sorry difficulty here hotel insurance uh yeah so um like the degree in which insurance is uh a regulatory labyrinth is insane um it was like designed to be hard you the feeling you got i mean i don't know but it certainly is very difficult um so there's a zillion applications and you have to wait for a bunch of time and get well it's become long and complicated and a lot of it is state by state most of it is state by state so um so we're just we're processing applications in states across the country and um and then the states also have different regulations so you can't you know actually aren't legally allowed to offer the same insurance in every state so you've got to adjust the software to be different every state it's basically it's very complicated um but i think um we'll be operating it in in texas very soon um and like maybe in a month or something next week great uh phone a friend um so uh yeah so title insurance goes live in texas next week um and um and it was this month but and then uh we have it in california we're going to be upgrading the version in california um because in california we want to have the same kind of real-time insurance that where your insurance costs are based on your actual driving history um which is like the right way to do it um but okay we're currently not allowed to do that in california for some reason um we're so we're trying to get permission from the regulators to be allowed to give accurate scores for insurance it seems like that's the thing you'd want to do um and um and then hopefully i know we'll probably have most of the country next year aspirationally that's our goal we're really getting into the weeds here um so we're definitely going to be making cyber truck here um and so probably the atv too so the atv is an interesting design challenge because atvs are pretty dangerous and so we want to make an atv that is the least dangerous atv so if you're going to atv well you might as well have the least dangerous atv you know so it'll have a really low center of gravity because the battery pack will be down low and i think we can do some things with the suspension it just make it really hard to roll this thing you know so it'll because when atvs roll is when bad things happen so it's gonna be the hvu that won't roll uh so it'd be cool you gotta have one with a cyber track you know okay so i'll take one last question yeah oh what about electric planes uh yeah yeah we have a lot on our plate here um but electric planes um yeah i've been dying to do that for a decade you know honestly but uh we got to we had quite a few fish to fry here um so every one day maybe one day the electric plane um and battery energy density is improving um every year so that's an important uh important metric to get the cell energy density to [Music] um around 450 500 watt hours per kilogram um and have it have a pack efficiency around 400 watts per kilogram that's when uh electric planes start to get to get interesting so i don't know it'd be a fun problem at to work on at some point um but uh we got we got a lot to do over the next few years so we're going to focus on these things get them right and then maybe one day do that all right thanks everyone for tuning in thanks for being here

Channel: CNET Highlights

Views: 31,410

Rating: 4.9572954 out of 5

Keywords: elon musk, questions and answers, investor day, tesla investor day, 2021, tesla shareholder meeting 2021

Id: ErlS9x_EsbQ

Length: 29min 29sec (1769 seconds)

Published: Thu Oct 07 2021