Welcome to Electrified. It's your host Dylan Loomis. Quick shout out to my newest patrons, D new I and Samuel S. Thank you for choosing to support the channel. On X, Elon said version 12.3.1 addresses several small annoyances and should start rolling out this weekend. Replying to John, Elon said the tendency to drive too slow should be addressed in 12.3.1. Still nothing for the 2024 branch.
欢迎来到《电气化》节目。我是您的主持人迪兰·卢米斯。特别感谢我的最新赞助者D new I和Samuel S。感谢您选择支持本频道。埃隆在X上表示,版本12.3.1解决了几个小问题,应该在本周末开始推出。回复约翰时,埃隆表示版本12.3.1将解决过于缓慢驾驶的问题。对于2024年分支仍没有进展。
We got some interesting comments from a South Korean auto analyst, Kim Jun-sung. The direction I heard at Tesla's informal meeting at the beginning of this year was the FSD and Optimus All In strategies. We internally expected that this strategy would be successful by June to July and believe that volume and profitability would naturally follow based on this. He continued, we asked them to specify what they meant by success and the answer was a driving capability equivalent to that of a Robotaxi.
If we're really successful as per our goal, we believe that securing Robotaxi licenses at major bases will begin in 2025 and FSD license contracts will begin in 2026. This is a very subjective and hopeful opinion. But the point, someone at Tesla, relatively high up, is pretty optimistic about FSD and the potential for these Robotaxi licenses sometime in the next 12-18 months. Kim then clarified his comments to avoid any misunderstandings.
He said the view that FSD will have Robotaxi equivalent capabilities by June or July is the personal opinion of the engineers I've spoken to. I don't believe this is an official goal within Tesla or guidance that can be shared externally. Listen, it could certainly be true that there have been Tesla engineers that thought Tesla was going to solve FSD for each of the past 5 years. However, with 12.3 unexpected significant upgrades set to roll out every 2 weeks, this time feels a little different.
Checking out Kim's LinkedIn page, he's been a research analyst at Merit Securities for over 3 years in South Korea. So, a timeline to keep in mind possible Robotaxi licenses at major cities next year, 2025. Then FSD licensing by other OEMs starts in 2026. Again, maybe not an official internal plan, but it's certainly not nothing either. In MKBHD's recent video of a Cybertruck that he was actually borrowing from somebody else, it had one of the worst driver door panel gaps that we've ever seen.
That's definitely unfortunate, but Elon did say about 15 Cybertrucks in service had an issue where the door striker loosened in the field due to insufficient torque after door fit. This takes 5 minutes to fix in service and has been addressed in production. We got our roughly quarterly update on Tesla's 4680 production progress. The Cybertruck account posted produced over 1000 Cybertrucks worth of 4680 sales at Gigatexis last week. Translation, 1000 car sets per week, a car set is just a battery pack.
Given that 4680s are arguably one of Tesla's 3-5 most important projects right now, it's worth getting into some of the context. Using the size of the Cybertruck battery, 123 kilowatt hours, times 1000 per week, that's 123,000. Converting that to gigawatt hours would be 0.123. That figure times 50 weeks, I like to allow for two weeks of downtime and maintenance, is about a 6.15 gigawatt hour annual run rate. Word of caution, extrapolating with these figures in my opinion is a silly venture.
The growth here is not going to be linear, nor is it going to stay static for long, we're on our way up the S curve. The previous high run rate, which was sometime in quarter 4, was around 4.9 gigawatt hours per year. We'll get to why the growth over the past six or so months was only about 25.5% when last year from quarter 2 to quarter 3 they were growing at about 40%. If we assume, based on data from Troy that the Cybertruck production rate right now is in the neighborhood of 300 per week, that would be an additional 700 car sets that they're making per week, or about 2.3 weeks of inventory that they're building up every week.
Technically, I suppose it'd be what they could start building up from here going forward if they just hit this rate for the first time. On Tesla's Q4 call though, they already said at that time they had a few weeks of car set inventory already set aside. I've seen a lot of speculation out there that given the increased run rate, Tesla might start putting 4680s in the Model Y or the Model 3. Personally, I'm not expecting that just because if they only have a few weeks of extra inventory, if something were to go wrong with these lines or the expansion of these new lines, they don't want to cut it too close.
And look, this situation could definitely be different by quarter 4 this year. I'm just saying I'm not expecting it in quarter 1 or quarter 2. Given Tesla's initial goal for annual Cybertruck production, they would be about 20% of the way there when it comes to this cell production rate. 250,000 trucks divided by 50 weeks is 5,000 per week. They're building 1000 car sets per week, 20%. To that earlier question, why has the growth rate been a bit slower over the past 6 months? We can't forget Tesla was retooling for the Gen 2 Cybertruck during quarter 4.
They were switching one Model Y line over to the Cybertruck cell version. As a refresher, in quarter 4 in Texas, Tesla was running one production line, one assembly line, and they were using two assembly lines for yield and rate improvement trials. The fourth line was in commissioning, and that was all for phase 1, which is going to be four lines. Phase 2 has already been underway now for a few months. That's going to be four more lines being installed starting in quarter 3 of this year. Those lines have been under construction dating back to Q3 of last year.
So once phase 2 is fully up and running, we'll be looking at eight production lines at Gigatexas. This serves as a nice, fun destroyer to what Reuters was spewing out as clearly 4680 production is still ahead of the Cybertruck ramp, and I would argue somewhat comfortably. Plus, given what I said that in quarter 4, Tesla already had a few weeks of car sets already set aside. This way, they can stay ahead of Cybertruck production and their future vehicle programs.
That latter part, future vehicle programs, that's not my speculation. That was aligned directly from the Q4 call. So where we're at now, Gigatexas is the primary 4680 production facility. Kato Road spent much of quarter 4 being retooled for large scale pilot runs of next gen cell designs. Kato's going to serve as the launch pad for new cells, one generation ahead of their mass production facilities. I've also seen some people saying that 4680 production in Berlin was ramping up. As far as I know, that's not accurate. They are producing very initial stage parts of the cell, but all of those are currently being shipped to the United States.
So Berlin is not currently making full 4680 cells and they're not making anything for production in Berlin. There's still the big question mark has Tesla made any progress when it comes to the dry battery electrode for the cathode. There were some reports earlier this year that Tesla was actually importing some wet cathodes from China for their 4680 production. And finally, on the production tax credit, since Tesla is making these batteries in the United States, it's $45 per kilowatt hour for cell production and pack assembly. Multiply that by 123 kilowatt hours per car set times 1000 cyber trucks per week, that's $5.5 million.
In these production tax credits per week. Multiply that by 50 weeks in a year to allow for the two weeks of downtime. That's about $276.8 million per year. That number is most likely with phase one not operating at 100% and we have an entire phase two, which will be doubling phase one set to come online later this year. It's great that Tesla is still making progress. We should also reasonably expect a faster rate of production over the next six to nine months.
But I think it's wise to at least keep in mind at battery day, which was September of 2020, Tesla was expecting to be producing 100 gigawatt hours of 4680 cells in 2022. It's been said that Kato Road has a capacity of 10 gigawatt hours per year, but they're not really producing at that rate. And then the question can be asked what percent of those cells go into production vehicles? The answer might be zero. I'd be the first person to defend Tesla in that building these batteries from scratch with all of the new technology for a car maker is an absurd task.
But I also want to hold them accountable for making projections like this. That's because practically speaking, if Tesla was even at about 50% of this in 2024 today, we'd most likely have model threes that could qualify for the IRA tax credit. There are certainly energy density and charge curve concerns for the 4680s relative to the 2170s, but strictly from a sell availability standpoint that clearly would have given Tesla a lot more flexibility. It is worth pointing out as far as we know, the new model three plus is not using the gigacast.
So what does that mean for a 4680 structural pack, even having the ability to go into a model three vehicle for whatever it's worth. I do still think it's more likely that the model threes requalify for the credits when that Panasonic factory in Kansas comes online, which is supposed to be 2025. It's true. Today, the administration officially put through the strongest ever pollution standards for cars in the United States. But it's also true that these have been peeled back from what they were originally proposing. I'll have the official eight page document linked below if you want to check it out. I just want to highlight what they said. The EPA projects that for model years 2030 to 2032, manufacturers may choose to produce be these for about 30 to 56% of new light duty vehicle sales and about 20 to 32% of new medium duty vehicle sales. I want to make it clear, these rules are not a mandate. It is literally just emissions guidelines that automakers have to hit across their entire fleet. So my point is they can do that in many different ways, selling a combination of full be EVs, hybrids and ice vehicles, and still buying credits. There's a lot of reporting out there that this is requiring 30 to 56% of new cars being sold being EV. But that's just not true. This is literally just a guess or an estimate or a loose suggestion based on what the emissions regulations are. We just talked about it last week. A Toyota executive is already saying that they're going to follow consumer demand and just buy credits to make up any EPA shortfall. These regulations also don't take effect until model year 2027.
I was able to find this which unfortunately I think is going to be important. According to statute, as long as the rules are published, more than 60 legislative days before the end of the presidential term, they cannot be eliminated by a simple majority vote in Congress. The commentary from most of Legacy Auto and the UAW was they were thankful that the EPA actually listened and reduced some of these rules between 2027 and 2030. Here's the thing though, it's almost a guarantee for gone these new regulations are going to be challenged legally. He's definitely not the only one, but Joe Manchin already said he would support Republicans' efforts to overturn the rule. He's already called this reckless and ill-informed and said it'll impose what is effectively an ED mandate without ensuring the security of our supply chains from nations like China and without a realistic transition plan that addresses our domestic infrastructure needs. Here's the table with the actual standards when it comes to the non-methane organic gas and nitric oxide emission levels by year. I just want to make it abundantly clear, if you're looking for what percent of EVs by what year, you're not going to get it, that's not how this regulation works. This is not an EV mandate.
Automakers can sell whatever cars they want, buy whatever credits they want, as long as ultimately across their fleet, they're hitting these emission levels. So just know that anybody out there that you see saying now 56% of all new car sales will be fully electric by 2032, those people don't actually know what they're talking about. We got an update on the Tesla IP thieves, it turns out the technology involved high-speed battery assembly lines that use a proprietary technology owned by Tesla. The two individuals, one that's been arrested, sold products developed with the stolen trade secrets. These two started their own business in China that they expanded elsewhere globally and the business makes the same battery assembly lines that Tesla uses with its proprietary info and it markets itself as an alternative source for the assembly lines. Tesla has also announced it will be raising Model Y prices in China starting on April 1st, but it's only by about $700. Right now it's not official, but the assumption is it will be for all Model Y trims.
Perhaps even more impactful though will be Tesla letting some of these other incentives that they have been running if those also expire at the end of March, then the cost of Tesla vehicles may be going up to the tune of thousands of dollars. I thought it was funny, Xiaomi CEO said I read rumors today that Tesla will increase prices April 1st, Tesla is so cool, truly impressed. He said the current ED market is so competitive that only Tesla dares to raise prices. With the current pure EVs, except for Tesla, many of them are making huge losses. It isn't interesting move by Tesla, you have basically all of the fish in the current swimming one direction in China lowering prices and now Tesla is starting to go the other way. It feels like an end of quarter incentive because Tesla has been forecasting all of these price hikes rather than just doing it which clearly would be a way to get people to buy now.
The word is GM executives think they may finally be crawling out of their production hell when it comes to the Altium production. Their CFO said we've had some challenges scaling it, you don't say, I think most of those are behind us. In 2023, GM sold fewer than 14,000 Altium based EVs when their goal for the year was about 75,000. They did admit part of their problem with scaling the Altium lines was they were implementing these fully automated lines but they didn't do any testing first rather than testing them somewhere else. They chose to install new fully automated battery assembly lines right away. A wise person once told me haste makes waste. Although the truth is when it comes to Tesla building factories, they often do things hastily as well and it's worked out so far. A not so fun fact, the Detroit Fire Department says it's been called nine times to factory zero since August. GM's VP of electrification said he thinks the company has turned the corner at factory zero and they'll continue to ramp up and they're still planning to deploy what they've learned there at a new battery plant in Tennessee scheduled to open later this year.
The employees at Berlin have been voting on the new works council and the majority of Tesla staff in Germany decided against the E-Game at all union and opted for non unionized lists for the new works council. E-Game at all's efforts to gain more control over Gigabrelin have been thwarted once again by the employees. Boston Consulting Group did a study on EV production and they concluded most automakers lose about $6,000 on every EV they sell for $50,000. That's after accounting for customer tax credits. More mainstream folks are finally starting to ask the question whether automakers have the stomach to keep investing until they get to the level of scale and efficiency where they can actually turn a profit. If you've been around, you've probably heard me say at least 10 times that no CEO wants to be at the helm when their company starts burning billions of dollars. For Lucid, it's actually more like 6 figures.
Stellantis has just invested an undisclosed amount in a LiDAR company called Steerlight. This company is touting a lower volume production cost than currently available LiDAR. There's very little public information about this Steerlight company and given they did not disclose the amount that usually means it's small. Looking at global data for the month of January this year for only full BEVs, that category grew 48% compared to January 2023. Globally, full BEVs took a 10% market share for January. Then things go slightly south as we have plug-in hybrids that were up 91% year over year in January and China represented over 60% of all EV registrations in January.
On that point, Martin Vieja said unfortunately people use plug-in hybrids mainly as gas cars which means their CO2 emissions are far worse than official EPA or WLTP ratings suggest. Just like officially rated energy consumption of EVs has been getting closer and closer to reality, same should be done for plug-in hybrids. He quote posted this that said plug-in hybrids perform shockingly bad, emitting an average of 3.5 times their official and publicly reported CO2 emissions. Jaguar Land Rovers performed the worst with a 147 grams of CO2 gap for every kilometer driven. The green portion of each bar shows you the gap or how much more CO2 per gram per kilometer these vehicles are actually emitting in the real world. Talk about false advertising. I'll use this opportunity to make a point. I think for us as we try to promote EVs, talking about climate change may not be the right thing for most of the general public. It's now one of those politicized polarized terms. I think we just need to focus on breathing cleaner air if you want to touch on the green aspect of EVs. I don't know many rational human beings that wouldn't want cleaner air for themselves and their future generations. What happens with an ICE vehicle in a closed garage when it's turned on is always a good example.
VW and Mobileye are deepening their ties. As they said today, they're going to bring Mobileye tech into Audi, Bentley, Lambo and Porsche vehicles. They'll also use Mobileye's self-driving systems in commercial vehicles, including a fleet of robo taxis based on the VW ID Buzz EV. We'll see. VW did not comment on a specific time frame to reach serious production with the systems. Naturally, VW tried to save face here by saying in the long term, the group will rely on its own complete in-house stack for automated driving across all brands. Mobileye's supervision is already on some roads in China with Gili. For this partnership, VW has been touting Level 3 automated driving, but they did say that a VW subsidiary is testing a driverless fleet in Austin.
As of today, Ford is now allowing dealers to advertise its EVs below, MSRP, this according to a bulletin sent to dealers. While dealers have been able to sell Ford EVs for any price, the rules prevented them from advertising their best prices. Given that Ford recently cut the Mach E price by up to $8,000, this probably is not a great sign. On Weibo, Polestar has announced that in China, they now have access to the Tesla Supercharger network. The last we heard, Polestar vehicles were supposed to have access to the Supercharger network in North America sometime this spring.
截止今日,福特现在允许经销商以低于建议零售价(MSRP)的价格进行电动车的广告宣传,这是根据一份发送给经销商的公告。尽管经销商一直可以任意出售福特电动车,但规定禁止他们宣传最优惠价格。考虑到福特最近将 Mach E 的价格下调了最多$8,000,这可能并不是一个好迹象。在微博上,Polestar 宣布他们在中国现在可以使用特斯拉超级充电站网络了。据我们了解,Polestar 车辆应该在今年春季能够访问北美的超级充电网络。
City officials in San Francisco are assessing the feasibility of a curbside EV charging program. They're saying the hope is to implement a pilot by next year, but city agencies still need to determine the project's cost. It was a green day across the board for our auto and AI sector, thanks to the Fed, Tesla stock closed the day at $175.66, up 2.53%, while the Nasdaq was up 1.25%. It was another quiet volume day for Tesla trading about 15 million shares below the average 30-day volume. As expected, the Fed left rates unchanged and based on the summary of economic projections or the dot plot, the expectation is still three rate cuts for this year, but that's definitely not a guarantee.
Pal did say they're prepared to retain the current rate for longer, and they also hinted that they may begin slowing the pace of quantitative tightening fairly soon. Based on the dot plot, we probably won't be going back to 0% rates unless something goes terribly wrong. And longer term for the normal rate going forward, that was boosted up from 2.52 to 2.6%. Hope you guys have a wonderful day, you can find me on X linked below. Please like the video if you did, and a huge thank you to all of my Patreon supporters.