@Quartr_App: During last week's earnings call, @elonmusk suggested that Tesla could surpass the value of the next top five companies – combined.
That would mean almost $15T or 11x Tesla's current market cap.
☑️ #300 Jan 2, 2024
Elon Musk uses Cybertruck explosion to show Tesla can remotely unlock and monitor vehicles
404media.co: [Excerpt] The fact that the CEO of a car company or someone working on his behalf can—and did—remotely unlock a specific vehicle and has the means of tracking its location as well as what Musk described as the vehicle’s “telemetry” is not surprising given everything we have learned about newer vehicles and Teslas in particular. But it is a stark reminder that while you may be able to drive your car, you increasingly do not own it, that the company that manufactured it can inject themselves into the experience whenever it wants, and that information from your private vehicle can be provided to law enforcement. Though Musk is being thanked directly by law enforcement, it is not clear whether Musk himself is performing these actions or whether he’s directing Tesla employees to do so, but Tesla having and using these powers is concerning regardless of who is doing it.
@elonmusk (1/1/25): We have now confirmed that the explosion was caused by very large fireworks and/or a bomb carried in the bed of the rented Cybertruck and is unrelated to the vehicle itself. All vehicle telemetry was positive at the time of the explosion.
🙂
☑️ #299 Dec 23, 2024
We Set the first Tesla FSD Cannonball Run Record
@AlexRoy144: [Excerpts] The Cannonball Run Is The Nurburgring of Autonomy. So We Set A Record In A Tesla.
For the sake of clarity, I will call cross-country driving records in vehicles with supervised automation surpassing 98% Semi-Autonomous, such as Jay Robert’s impressive Comma AI record. If the vehicle is a production car with native software like Tesla’s FSD, the sub-category is Production Autonomy.
@thejefflutz: FSD V13.2.1, WOW. It felt like I was controlling the car w/my mind. It did everything I wanted it to the way I wanted it to...
I don't know how to describe it other than I think I'm done w/human driving. I was already having FSD do 80-90% of my drives before on V12.5, this feels like it closed the remaining gap. Last year was only about 30% of my drives on V11, I think the progression is important to describe.
With V13.2.1 you feel the reduced latency and improved performance by more human-like acceleration, slowing down, merges, turns - except it's perfect every time and humans are not perfect.
I'm 3 drives in, more testing to do, I'm sure there are edge cases to encounter. But this goes beyond any implicit bias, it's meaningfully better and we know there are more enhancements and features on the way...
Bravo @Tesla_AI team and I can't wait for more to be able to try it themselves.
🙂
☑️ #296 Dec 15, 2024
Tesla Lithium Refinery is cooking
@tesla_na: This week, we fed raw material through our kiln of our Lithium Refinery outside of Corpus Christi, TX for the very first time! Increasing lithium refining capacity is critical to a sustainable energy economy.
+ Related content:
@JoeTegmeyer (9/6/23): Big tanks near the battery cathode plant finally getting moved inside (south end, 2nd floor)! These are material hoppers that will be used for some of the processing chemicals/raw minerals to manufacture the cathode materials in this plant! This is also related to the Lithium Refinery in Corpus Christi I've been covering in my multi-part series!
[UPDATED] @JoeTegmeyer: I've had a lot of requests to find out what the four large "tanks" are being prepared for installation at the battery cathode plant, so I thought I'd try and help answer that.
@GaryStoneSWS: In May 2021, Elon lost $14 BILLION with a single tweet - on purpose. When he announced Tesla no longer accepted the world's # 1 crypto... Stocks plummeted. Everyone called him crazy. Here's why it was actually the most genius move in corporate history.
⚡️
[2/19] Let's rewind to February 2021. Tesla had just announced a groundbreaking $1.5B investment in Bitcoin. They became the first major corp to embrace crypto so openly. Bitcoin soared to record highs of $64,000 by mid-April. But something was brewing beneath the surface...
[3/19] The crypto world worshipped Musk. His tweets about Bitcoin and Dogecoin drove prices sky-high. Millions of investors hung on his every word. Tesla's move was seen as validation of Bitcoin's legitimacy. Then came May 12th…
[4/19] With 1 tweet, Musk wrote: "Tesla has suspended vehicle purchases using Bitcoin. We are concerned about rapidly increasing use of fossil fuels for Bitcoin mining and transactions, especially coal, which has the worst emissions of any fuel." The reaction was swift and brutal…
[5/19] Bitcoin's price nosedived by over 12% in hours. The broader cryptocurrency market saw over $365 billion evaporate within a week. Elon's secret?
[6/19] This wasn't a mistake. This wasn't Musk being reckless. This was one of the most calculated corporate moves in history. And it all starts with understanding Tesla's true mission:
⚡️
[7/19] Tesla's mission has always been to accelerate the world's transition to sustainable energy. By highlighting Bitcoin's environmental impact, Musk reinforced Tesla's commitment to this mission. He positioned Tesla as an ESG leader. But the genius goes deeper than that…
[8/19] At the time, Bitcoin faced a massive reputation crisis: Its mining consumed more energy annually than entire countries. Critics lambasted its carbon footprint. Musk didn't just point out the problem. He forced the entire crypto industry to evolve:
[9/19] Within months of his tweet: Major mining operations began transitioning to renewable energy sources. The industry started serious discussions about sustainability. Musk had effectively told the crypto world: "Adapt or die." But there's another layer to this masterpiece:
[10/19] The move resonated with investors focused on sustainability metrics. ESG-focused funds—which manage trillions of dollars—saw Musk as a leader willing to take a stand. While crypto advocates bristled, these massive funds saw the bigger picture. Elon's brilliance becomes clear:
[11/19] In one move, Musk had: • Reinforced Tesla's environmental leadership • Created potential buying opportunities • Attracted ESG-focused investors • Demonstrated unparalleled market influence But the master stroke was yet to come:
[12/19] Musk later clarified Tesla would resume accepting Bitcoin once mining reached a threshold of 50% renewable energy usage. This wasn't just a condition. It was a roadmap for the entire crypto industry. He had shaped the future of both electric vehicles AND cryptocurrency.
@Tesla_Megapack: Now operating - Eland 1, a hybrid solar + Megapack project, owned and operated by Arevon.
This is the first project deployed with Tesla's new site-level controls that seamlessly regulate power output to buffer the intermittency of solar and provide a firm resource to the grid. The hybrid configuration allows more solar power to be installed at the site than the interconnection nameplate, while also avoiding costly solar curtailment.
arevonenergy.com: Eland is a 150 MWac fixed-tilt photovoltaic solar power plant located in California.
arevonenergy.com (12/9/24): [Excerpt] Arevon Announces the Start of Commercial Operations at its Eland 1 Solar-plus-Storage Project in California.
MOJAVE, Calif. and SCOTTSDALE, Ariz.(December 9, 2024) – Arevon Energy, Inc., a leading renewable energy developer, owner, and operator, today announced the start of operations at its Eland 1 Solar-plus-Storage Project in Kern County, California. Eland 1 Solar-plus-Storage, based in the city of Mojave, is a 384 megawatt (MWdc) solar project coupled with 150 MW/600 megawatt hours (MWh) of energy storage.
A second phase of the Eland Solar-plus-Storage Project is currently under construction and is anticipated to be operational in Q1 2025. Collectively, Eland 1 and 2 are one of the nation’s largest solar-plus-storage installations with total capital costs of more than $2 billion. When both phases are completed, the projects’ combined capacity will total 758 MWdc of solar with 300 MW/1,200 MWh of energy storage.
🙂
☑️ #293 Dec 6, 2024
John Murphy (BoA) visited Memphis
@SawyerMerritt: NEWS: Tesla told Bank of America analyst John Murphy during his recent Giga Texas visit that soon Tesla's FSD (Supervised) will require an intervention only once every 10,000 miles. Waymo's robotaxi service operated before getting to one intervention every 17,000 miles, a milestone Tesla said it believes it could achieve shortly with its growing Al compute (50,000 H100 chips were installed and active at the end of October 2024).
⚡️
@rajuvamsi007: FSD: One intervention > 10,000 miles - Any Public Road
Waymo: One intervention > 17,000 miles:
40 Square mile area
Pre-mapped
Remote Operators
5x hardware cost
Can't operate on highways Tesla is way ahead on every metric.
+ Related content:
@herbertong (12/5/24): NEWS: BofA Securities boosts$TSLA price target to $400 (up from $350) while keeping a BUY rating
According to the analyst, the increased presence of robots like Optimus will likely expedite their training and enhance their capabilities. Although Optimus currently utilizes only a single-digit percentage of Tesla's computing power, expectations are that the resources allocated to Optimus will expand as Tesla's robotaxi technology advances.
This progression is anticipated to quicken Optimus's development and contribute to a surge in production starting in 2026, which could assist in reducing costs. With annual revenue of $97.15 billion and a market capitalization of $1.15 trillion, Tesla maintains its position as a dominant player in the automotive industry.
The report also suggested that the "Optimus opportunity," among others, could potentially gain from an infusion of capital through an equity raise. This would be aimed at boosting compute capacity, a move that, according to Bank of America Securities, would likely be received positively by the majority of Tesla investors.
🙂
☑️ #292 Dec 6, 2024
The best-selling electric car for the last 12 months in Harbin
@Tesla_Asia: Model Y is the best-selling electric car for the last 12 months in Harbin, one of the coldest cities in China famous for its icy winters and stunning snow landscapes.
🙂
☑️ #291 Dec 5, 2024
FSD 13.2
@AIDRIVR: Tesla AI Delivered. FSD 13 is MIND BLOWING
bloomberglaw.com (update; 12/4/24): [Excerpt] Judge refuses to reverse landmark decision after investor vote. Sets up appeal to Delaware high court after 6-year legal saga.
After a year of threats, accusations, and novel legal maneuvers, the world's richest man is back where he started 2024: stewing over the $56 billion that got away.
A Delaware judge handed out an unprecedented $345 million in legal fees while rejecting Elon Musk's record pay package for the second time Dec. 2, refusing to restore the deal even after Tesla Inc. investors voted for it again. The 103-page decision sets up an appeal after a six-year saga that captivated the state's legal community.
news.bloomberglaw.com (update; 12/3/24): [Excerpt] Musk’s Multibillion-Dollar Tesla Payout Is Gutted by Judge (1). Judge declines to embrace shareholder vote favoring 2018 plan. Tesla to appeal decision; Musk calls it ‘absolute corruption’.
Delaware Chancery Court Judge Kathaleen St. J. McCormickruled Monday that Tesla’s board was improperly influenced by Musk when it adopted the billionaire’s plan in 2018. It was the second time she rejected the pay package as excessive, sticking with her original finding in January even after shareholders backed the plan and Musk asked her to reconsider.
🙂
☑️ #289 Nov 22, 2024
The Tesla Supercharger team announced four improvements
We opened the Supercharger network to be helpful to all EV drivers and car manufacturers going electric. However, different charge port locations on other EVs aren't great when charging on shorter cable Superchargers. Below are our recent efforts to continuously improve the charging experience for all drivers.
Making stall availability more accurate than ever – The latest Tesla software update improves the accuracy of stall availability estimates. We can detect when another EV, with a charge port located somewhere other than the rear left or front right, is plugged into a short-cable Supercharger stall. This update ensures no more overpromising of stall availability, so you can travel with confidence. We will continuously refine this algorithm to be as accurate as possible, including exact site mapping and faster refreshing of stall availability.
Increasing number of long cables - Longer cables mean that V4 posts can serve all port locations. In the next 18 months we will have more long cable than short cable Superchargers.
Modifying our sites to avoid blocked spaces - We have modified over 1,500 sites so that drivers never have to use more than 2 charging spaces to charge, increasing stall availability for all.
Encouraging the best charge port locations - Since opening up the Supercharger network in Europe in 2021, we've encouraged car manufacturers to transition charge port locations to rear left or front right. This provides seamless compatibility with 30k+ short-cable Superchargers available to other EVs globally.
Backend Software Engineer, Tesla Bot & Robotaxi:Tesla AI’s Teleoperation team is charged with providing remote access to our robotaxis and humanoid robots. Our cars and robots operate autonomously in challenging environments. As we iterate on the AI that powers them, we need the ability to access and control them remotely. This requires building highly optimized low latency reliable data streaming over unreliable networks in the real world. At Tesla, we control the entire hardware and software stack, end to end. Our goal is to integrate our hardware, firmware and backend expertise to achieve a cutting-edge system.
C++ Software Engineer, Teleoperation, Tesla Bot and Robotaxi:[additional text] Our remote operators are transported into the device’s world using a state-of-the-art VR rig that allows them to remotely perform complex and intricate tasks. Working with hardware teams, you will drive requirements, make design decisions and implement software integration for this custom teleoperation system.
The "second stage" would be a full self driving solution based on large-scale neural network training
@amasad: Karpathy’s email is prescient in a few ways: calling out scaling as the most important thing for progress in AI and predicting Tesla’s market cap in the next 3 years.
Andrej Karpathy to Elon Musk - Jan 31, 2018 11:54 PM
Working at the cutting edge of Al is unfortunately expensive. For example, DeepMind's operating expenses in 2016 were at around $250M USD (does not include compute). With their growing team today it might be ~0.5B/yr. But then Alphabet in 2016 reported ~20B net income so it's still fairly cheap even if DeepMind had no revenue of its own. In addition to DeepMind, Google also has Google Brain, Research, and Cloud. And TensorFlow, TPUs, and they own about a third of all research (in fact, they hold their own Al conferences).
I also strongly suspect that compute horsepower will be necessary (and possibly even sufficient) to reach AGI. If historical trends are any indication, progress in Al is primarily driven by systems - compute, data, infrastructure. The core algorithms we use today have remained largely unchanged from the ~90s. Not only that, but any algorithmic advances published in a paper somewhere can be almost immediately re-implemented and incorporated. Conversely, algorithmic advances alone are inert without the scale to also make them scary.
It seems to me that OpenAI today is burning cash and that the funding model cannot reach the scale to seriously compete with Google (an 800B company). If you can't seriously compete but continue to do research in open, you might in fact be making things worse and helping them out "for free", because any advances are fairly easy for them to copy and immediately incorporate, at scale.
A for-profit pivot might create a more sustainable revenue stream over time and would, with the current team, likely bring in a lot of investment. However, building out a product from scratch would steal focus from Al research, it would take a long time and it's unclear if a company could "catch up" to Google scale, and the investors might exert too much pressure in the wrong directions.
The most promising option I can think of, as I mentioned earlier, would be for OpenAl to attach to Tesla as its cash cow. I believe attachments to other large suspects (e.g. Apple? Amazon?) would fail due to an incompatible company DNA. Using a rocket analogy, Tesla already built the "first stage" of the rocket with the whole supply chain of Model 3 and its onboard computer and a persistent internet connection. The "second stage" would be a full self driving solution based on large-scale neural network training, which OpenAl expertise could significantly help accelerate. With a functioning full self-driving solution in ~2-3 years we could sell a lot of cars/trucks. If we do this really well, the transportation industry is large enough that we could increase Tesla's market cap to high O(~100K), and use that revenue to fund the Al work at the appropriate scale.
I cannot see anything else that has the potential to reach sustainable
Google-scale capital within a decade.
-Andrej
+ Related content:
@amasad: The Elon vs OpenAI is an interesting read. Here is the full thing. Fun trivia: Elon typically replies with one word-sentence within minutes. Even to long emails such as reviewing the Microsoft deal where he caught a bad clause that they fixed.
Late October Tesla Diner Drone Flyover - The Screen! The Door!
The Tesla Diner site at 7001 Santa Monica Blvd, Hollywood is continuing to take shape. The episode, we see the 45 foot LED screen installed for the 1st time. More concrete work and the full scoop on those Tesla entry doors.
🙂
☑️ #285 Nov 3, 2024
"Tunnel Painting" Tesla Gigafactory Texas 11/2/2024 9:14AM
@BTSflyer: The tunnel under Toll 130 highway is getting a coat of paint on the inside. More glass panels are being installed today. The cooling tower fans are being tested out.
@elon_docs: In 2022, Elon Musk predicted all human code could be replaced by AI in Tesla's self-driving software. This approach materialized two years later with FSD v12.
“The thing that's happened over time with respect to Tesla Autopilot, is that the neural nets have gradually absorbed more and more software.
In the limit, of course, you could simply take the videos as seen by the car and compare those to the steering inputs from the steering wheel and pedals, which are very simple inputs.
In principle, you could train with nothing in between, because that's what humans are doing with a biological neural net, you could train based on video, and what trains the video is the moving of the steering wheel and the pedals, with no other software in between.
We're not there yet, but it's gradually going in that direction.”
@Tesla_AI: As October comes to a close, here's an update on the releases
What we completed:
End-to-end on highway has shipped to ~50k customers with v12.5.6.1
Cybertruck build that improves responsiveness
Successful We, Robot event with 50 autonomous Teslas safely transporting over 2,000 passengers
What's coming next:
Full rollout of end-to-end highway driving to all AI4 users, targeted for early next week, including enhancements in stop smoothness, less annoying bad weather notifications, and other safety improvements
Improved v12.5.x models for AI3 city driving
Actually Smart Summon release to Europe, China and other regions of the world
v13 is a package of following major technology upgrades:
36 Hz, full-resolution AI4 video inputs
Native AI4 inputs and neural network architectures
3x model size scaling
3x model context length scaling
4.2x data scaling
5x training compute scaling (enabled by the Cortex training cluster)
Much improved reward predictions for collision avoidance, following traffic controls, navigation, etc.
Efficient representation of maps and navigation inputs
Audio inputs for better handling of emergency vehicles
Redesigned controller for smoother, more accurate tracking
Integrated unpark, reverse, and park capabilities
Support for destination options including pulling over, parking in a spot, driveway, or garage
Improved camera cleaning and handling of camera occlusions
We have integrated several of these improvements and are already seeing a 4x increase in miles between necessary interventions compared to v12.5.4.
This lays the foundation for the v13 series, and we are targeting to ship v13.0 to internal customers by the end of this week.
Most of the remaining items are independently validated and will be integrated over November in a series of point releases. We are targeting a wide release with v13.3 with most of the above improvements for AI4 vehicles around Thanksgiving!
@nvidia (10/28/24): .@xAI's Colossus in Memphis, the world's largest AI supercomputer with 100,000 NVIDIA Hopper GPUs, achieves new heights with NVIDIA Spectrum-X Ethernet. A testament to NVIDIA's dedication to #AI progress. Read more: https://nvda.ws/4fnuuWG
servethehome.com (10/28/24): [Excerpt] Inside the 100K GPU xAI Colossus Cluster that Supermicro Helped Build for Elon Musk.
🙂
☑️ #150 Oct 25, 2024
🙂
☑️ #281 Oct 27, 2024
Standardizing Automotive Connectivity
tesla.com: [Excerpt] Over the last 20 years, the cost and complexity of electronics across the automotive industry has doubled.* Today a single vehicle typically requires over 200 connections—and the number of electrical connectors and types across new vehicles is only increasing.
To accelerate the world’s transition to sustainable energy, we are simplifying the manufacturing process and electrical connectivity requirements for all our vehicles. This includes the implementation of our Low-Voltage Connector Standard (LVCS), which allows us to reduce the large number of connector types required to just 6. These 6 device connectors are designed to meet the power and signal requirements for over 90% of typical electrical device applications. This standardization unlocks further operational efficiencies, cost reductions and manufacturing automation.
Tesla’s Substack: [Excerpt] Looking at the cost of goods sold (COGS), Tesla is already an efficient automaker but continues to optimize its operations. In Q3, the COGS per vehicle dropped to approximately $35,100, the lowest level ever for the company.
Tesla to retail home batteries through Yamada, Japan's top chain
asia.nikkei.com: [Excerpt] As car sales slow, U.S. EV giant expands virtual power plant business
TOKYO -- Tesla will start selling its home battery system through Yamada Denki, Japan's largest electronics store chain, in a partnership with Yamada Holdings, the chain's operator, as part of a strategic push in its virtual power plant (VPP) business, Nikkei has learned.
Tesla is already a major player in the energy storage market in the
U.S. and elsewhere. VPPs can service multiple homes by storing power when less expensive surplus electricity is available and selling that energy during demand spikes, thus generating revenue.
paloaltoonline.com: [Excerpt] City officials discuss using city as testing area for Tesla ‘cybercabs’.
The city recently held a preliminary discussion with Tesla to discuss a potential partnership, according to Meghan Horrigan-Taylor, the city’s spokesperson. If approved, the deal could supplement the existing taxi service that serves the Palo Alto community, including Palo Alto Link, a rideshare program that made its debut last year.
Creating a more sustainable fleet: Could this electric semi truck be a game-changer?
dhl.com: [Excerpt] At DHL, we are committed to reducing emissions in logistics. Improving the sustainability of our fleet is a big part of that. We’re now partnering with electromobility giant Tesla in the United States, working together toward a new era of long-haul trucking – 18 wheels at a time.
Tesla’s Vision for an Autonomous Future - We Robot Event Recap
@MunroLive : Sandy was invited to Tesla's We Robot event on 10/10, where they unveiled the CyberCab and RoboVan, and provided updates on Optimus.
Munro Live is a YouTube channel that features Sandy Munro and other engineers from Munro & Associates. Munro is an engineering consulting firm and a world leader in reverse engineering, costing, and teardown benchmarking.
tesla.com: [Excerpt] We, Robot. The Future Is Autonomous.
To create a sustainable future, we must democratize transportation. We do this by making driving more efficient, affordable and safe. Autonomy makes this future possible, today.
Robotaxi. Source: Tesla, Inc.
Robovan. Source: Tesla, Inc.
Tesla Bot. Source: Tesla, Inc.
🙂
☑️ #274 Oct 14, 2024 🟠 opinion
Fie, Robot
LILEKS (James): I love them. They're amazing. We don't really need them.
Open Source bi-weekly convo w/ Bill Gurley and Brad Gerstner on all things tech, markets, investing & capitalism. This week, Jensen Huang, CEO of NVIDIA, makes a guest appearance. In Bill’s absence, Brad is joined by Clark Tang (Partner at Altimeter) as they discuss with Jensen scaling intelligence towards AGI, the acceleration of machine learning, NVIDIA's competitive advantages, the significance of inference alongside training, future market dynamics in the AI landscape, the impact of AI on various industries, the future of work, inference time reasoning, AI’s potential to enhance productivity, the balance between open source and closed source, Elon’s Memphis Supercluster, X.ai, OpenAI, the safe development of AI, & more. Enjoy another episode of BG2.
(01:12:31)The Balance of Open Source and Closed Source AI
🙂
☑️ #272 Oct 12, 2024 🟠 opinion
Optimus ƨυmiɈqO
@TroyTeslike: It seems some people are still confused about how autonomous the Optimus bots were at the Tesla event. This clip is from another video Tesla released a while ago. It shows how the bots are controlled remotely using a haptic suit, kind of like puppets. Also, during the event, the operators were speaking through a microphone. None of the bots were talking on their own. In short, all the hand movements and speaking were done remotely, but the walking was autonomous.
⚡️
@TroyTeslike: At yesterday's Tesla event, the Optimus bots were remote-controlled and voiced by humans. While the demonstration was interesting, I think Tesla could have been a bit more transparent by doing a few things differently:
It would have been helpful if Tesla showed the people controlling and speaking for the bots, instead of keeping them behind the scenes.
During the presentation, Elon mentioned that the bots would serve drinks, but there’s a difference between them serving autonomously and being controlled to create that impression. It might be helpful to clarify that distinction in the future.
Employees referred to the bots as "human-assisted," as seen in the video below, which implies some level of autonomy. However, it seems all movements, except for walking, were remote-controlled.
A more accurate term might have been "teleoperated" or "telepresence" to include the speaking. Tesla is clearly still in the early stages of developing this technology, and this demo was more about showing what’s possible in the future.
🔹Related content:
@business: Tesla used humans to remotely control some capabilities of its Optimus robot prototypes at a recent event designed to generate investor enthusiasm for forthcoming products.
@hejefflutz: This is not news to anyone at the event who saw the Tesla engineers around the bots with laptops open and anyone in dialogue. And those engineers could be seen in most videos on line too nor did Tesla make any claims otherwise. With all that in mind there is still no other comparable benchmark to what Tesla displayed
MACHINE SOCIETY: Elon Musk's beer-pouring 'robots' are more like puppets. The Optimus robots trotted out at yesterday's Teslafest aren't the social, capable, autonomous AI robots we're led to believe.
🙂
☑️ #271 Oct 11, 2024
Hogner: This was very disappointing for investors.
@CNBCtelevision: Eirik Hogner, Partner and Deputy Portfolio Manager at Clean Energy Transition, expresses disappointment in the investor reaction to Elon Musk's Cybercab reveal, highlighting unmet expectations for autonomous technology and the vehicle's pricing.
@bradmunchen (10/5/24): As the highly anticipated “Robotaxi Day” on 10/10 draws nearer, the vested interests in Tesla appear to be furiously ramping the stock up via the options market.
The call options at 250 strike that expired today traded 375,539 contracts, which amounted to a whopping $9.4 billion in notional value (see image below).
This is 3x the average traded value seen in recent months. Despite expiring today, the 250 calls were up by 291% at their high this morning. Tesla’s stock price closed up by 3.9% at $250.
I mention this because, when Tesla goes down, the momentum to the downside could be equally vicious and many will get rug-pulled.
And this story below about a retail investor in Vancouver shows how crazy things can get in Tesla’s options market: he went from $88,000 to $415 million by trading Tesla options and then lost everything.
@bogusjack (9/5/24): In the Warner Brothers Studio, where Tesla’s robotaxi event will be held on 10/10:
1. Red box: The part of the studio’s backlot space that actually has real sets and a road, excluding warehouses, etc.
2. Green circle: The large parking space where the main event can be held, expected to be the final arrival point of the robotaxi.
3. Orange line: Possible robotaxi driving route within the red box grounds.
In my opinion, it’s not just one vehicle demonstration, but multiple cars will demonstrate simultaneously according to various scenarios in different places. Also, if the event is set up like a movie set or a live show, it will elicit exclamations of admiration from everyone
🙂
☑️ #269 Oct 9, 2024 🟠 opinion
Tesla's Inconsequential Robotaxi 'Reveal'
Lawrence Fossi’s Substack: Several hugely important things are happening for Musk & Tesla this fall, but the "Cybercab" event is not one of them.
Vision-Based Machine Learning Model for Autonomous Driving with Adjustable Virtual Camera
@seti_park: @Tesla's US20230057509A1 patent introduces an innovative vision-based machine learning model for autonomous driving that uses only cameras, eliminating the need for LiDAR and radar. The system uniquely processes information for vulnerable road users (VRUs) like small animals and pedestrians and non-VRUs like vehicles separately, optimizing detection for each category.
[FIG. 1A: Block diagram illustrating an example autonomous vehicle which includes a multitude of image sensors an example processor system] [FIG. 1B: Block diagram illustrating the example processor system determining object/signal information based on received image information from the example image sensors]
[FIG. 1B: Block diagram illustrating the example processor system determining object/signal information based on received image information from the example image sensors]
Tesla's patent, US20230057509A1, introduces a groundbreaking dual network structure for enhanced perception in self-driving vehicles. This innovative technology likely underpins the recent improvements observed in FSD v12.5, particularly in object detection and tracking. Let's examine the key components of this advanced vision system.
Tesla's new vision system utilizes an array of cameras strategically placed around the vehicle, providing a comprehensive 360-degree view of the surroundings. Read more
🙂
☑️ #267 Oct 7, 2024
Top Tesla Autopilot departures
@TaylorOgan: Tesla’s head of vehicle programs just jumped ship ahead of the robotaxi unveil in a few days. There has been a notable exodus of top-level talent at Tesla in recent days. Our team has compiled top Tesla departures from this team during its glory years (talent-wise)
🔹Related content:
tesla.com: [Excerpt] Autopilot is an advanced driver assistance system that enhances safety and convenience behind the wheel. When used properly, Autopilot reduces your overall workload as a driver. Each new Tesla vehicle is equipped with multiple external cameras and powerful vision processing to provide an additional layer of safety. All vehicles built for the North American market now use our camera-based Tesla Vision to deliver Autopilot features, rather than radar.
🙂
☑️ #266 Oct 4, 2024
FSD Supervised
@Tesla_AI: What's special about FSD Supervised is that it works anywhere in the US & Canada. No high definition maps, no geofence. This means you can even use it in places that no Tesla has never traveled to before.
🔹Related content:
@tesla: What's special about FSD Supervised is that it works anywhere in the US & Canada.
No high definition maps, no geofence.
This means you can even use it in places that no Tesla has never traveled to before.
FSD (Supervised) can drive you almost anywhere, including these common & complex scenarios. FSD (Supervised) requires that the driver pay attention to the road and be prepared to take over.
tesla.com: [Excerpt] With Full Self-Driving (FSD) (Supervised), you will get access to a suite of more advanced driver assistance features, designed to provide more active guidance and assisted driving under your active supervision.
If you have not already purchased FSD (Supervised) and your vehicle has FSD computer 3.0 or above, you can subscribe to FSD (Supervised) from the Tesla app or your vehicle’s touchscreen.
🙂
☑️ #265 Oct 3, 2024 🔴 rumor
Tesla Plans 4 New Batteries In 2026, including for a Robotaxi
@LimitingThe: NEWS: The Information claims that "Tesla Plans 4 New Batteries in 2026, Including For a Robotaxi"
This article contains information on the 4680 ramp as well as 4 new battery cells. Let's take a look at each.
Key Points on the 4680 Ramp
1) The article claims Tesla is losing 70-80% of the cathodes in test production for the fully dry 4680 cells. For comparison, most battery makers usually see less than 2%.
Thoughts: This is as expected for early production ramp. They just made the first pack a couple months ago. I suggested a couple months ago, and on more than one occasion, that the ramp might be slow for the fully dry 4680.
2) The article claims that most of the 4680 cells going into the Cybertruck are from Fremont.
Thoughts: This doesn't make sense based on what Tesla's told us in the past. Without further explanation, I can only assume this was a miscommunication with the source.
3) The fully dry version of the 4680 is called the 4680D.
Thoughts: Maybe this is what they mean in keypoint #2. It would make sense to manufacture the first 4680D cells in Fremont at their R&D hub.
4) Tesla is pushing ahead with plans to introduce the 4680D in the Cybertruck by the middle of next year.
Thoughts: In the Q2 earnings call, Tesla said that they planned on entering full production with the 4680 in Q4 of this year. Many of us assumed that meant that it was going to start being used in the Cybertruck. However, it may have meant that they're moving from producing the 4680 on pilot scale equipement at Kato road to full scale equipment in Austin. If that's the case, it makes sense that it would take several months to build the ramp enough to start supplying enough cells for the Cybertruck ramp.
5) Tesla has plans to improve the cathode yield to 90% by the end of this year, and they'd like to be mass producing the 4680D by second quarter of 2025. As per my comments above, this makes sense.
6) The cost of production is coming down.
This is to be expected with a production ramp.
7) Next year, Tesla aims to be making 2-3k Cybertrucks per week using 4680Ds, more than twice the production rate of the current production without dry cathodes.
Based on the most recent information, this would mean roughly matching the current production of their lines that produce 4680s that use a wet coated anode.
This is an aggressive goal in my view. But, it's not mission critical if they don't hit that goal. Currently, it looks like they have a glut of 4680 production for the Cybertruck.
If they do hit that production rate, it would be awesome! If they kept the current wet cathode lines going, that could mean Austin is producing more GWh of cell output than Giga Nevada by the second quarter of 2025.
Key Points on the 4 New Battery Cells
The conversation claims that Tesla will introduce 4 new versions of the 4680 by 2026.
This should be expected, but it's great to finally see some rumors about this. Tesla said on Battery Day that they planned on using at least 3 different chemistries in the 4680 form factor.
1) NC05, or New Cell 05 - A workhorse cell that will power the robotaxi, Cybertruck, Semi, and a fourth unkown vehicle.
Odds are highly likely that this is some type of LFP cell. I've been saying for years that they could do an LFP 4680, but that it was at least a couple years off.
Obviously, the Semi and Cybertruck would be shorter range versions of those vehicles if NC05 is an LFP chemistry.
Note: I don't think this necessarily means that the Robotaxi is expected in 2026. They can use more than one pack in that vehicle. I'd expected CATL prismatic LFP cells, but if they can launch an LFP 4680 in time for that Robotaxi, that would be excellent!
2) NC20 - This cell will offer more peak energy, and will be used for SUVs, the Cybertruck and other future vehicles.
This is potentially an LMFP chemistry or layered oxide cathode chemistry that uses something like 66% Nickel and 34% Manganese to reduce cobalt and overall cell cost, but still offers great energy density.
3) NC30 and NC50 - These cells will be ultra-high performance cells that use silicon carbon in the anode. This will increase both energy density and charging speed.
NC30 will be used in the Cybertruck and Sedans, while the NC50 is smaller an more focused on performance and will be used in the Tesla Roadster.
Silicon carbon just means graphite and silicon in the anode. I did a post on this earlier today.
NC30 just sounds like an upgrade to the current version of the Generation 2 4680, or Cybercell.
As I've said many times recently, silicon is the next logical step
There's no way to know what the energy density and charging speed would be, but on the energy density side, I'd expect 8-10%.
NC50 sounds like a specialized low volume cell. If it's not using the 4680 form factor I'd expect it to use the 18650 or 2170 form factor. The 18650 makes the most sense to me. That because it could potentially have better cooling with smaller cells (see Plaid Model S).
Final thoughts
All of this is expected to some degree, but it's always exciting to hear gossip and hype The Information has shown to be mostly reliable with rumors.
I'd be VERY impressed if Tesla manages not just 4 new battery cells, but 4 new battery packs in the next 18-24 months.
Likewise with the production ramp plans for the 4680D.
Overall, it sounds like Tesla is layering increasing expecations on the 4680 ramp, which is a far cry from the rumors earlier this year which said they were going to shut the whole project down.
That's why I try to focus on what Tesla says in the earnings calls rather than rumors. As usual, I'll be doing my quarterly update video a week or two after the earnings call (Late Oct or Early Nov)
⚡️
@seti_park: Here's my guess about @Tesla's four new batteries based on patent analysis: I think NC05 will use manganese spinel CAM (LiMn2O4), and NC30 will be Gen2 4680 with Si/C anode. NC50 will likely be a high-voltage NCM battery. There's less patent information about NC20, but I suspect it might adopt mid-nickel NCM as CAM.
Sony Honda Mobility Adopts North American Charging Standard (NACS) for AFEELA EVs in the U.S. and Japan – First non-Tesla vehicle for Japan market
shm-afeela.com: [Excerpt] Sony Honda Mobility Inc. (hereinafter referred to as SHM) has announced an agreement with Tesla Inc. to adopt the North American Charging Standard (NACS) for its “AFEELA” EVs, which will start sales in 2025, first from United States and later in Japan. This also makes SHM the first company in Japan, to adopt the NACS standard, and have access to Tesla V3 and V4 Superchargers.
@tesla: Helping Replace Hawaii’s Last Coal Plant | Tesla Megapack.
Megapacks at KES are supporting the replacement of Hawaii’s last coal plant and proving the grid-forming technology that is critical to stabilizing the grid.
pluspower.com > kapoleienergystorage.com: [Excerpt] About the Project. Kapolei Energy Storage (KES) is ideally located on roughly eight acres of land in Kapolei on the island of Oahu, where it interconnects at a critical Hawaiian Electric substation. The 185 MW / 565 MWh battery storage project provides load shifting and fast-frequency response services to Hawaiian Electric, enhancing grid reliability and accelerating the integration of readily available renewable energy. KES received approval from the Hawai’i Public Utilities Commission in May 2021. The Kapolei Energy Storage facility is now online.
The KES project helps replace the AES coal-fired plant that closed on September 1, 2022 and supports the state’s goal of shifting from fossil fuels to 100 percent renewable energy generation by 2045. The KES project received unanimous support from the local Neighborhood Board and approval of its Conditional Use Permit-Minor from the City and County of Honolulu. Situated in I-2 (Industrial) zoning outside the Tsunami Evacuation Zone, the KES project site is an optimal location for new energy infrastructure.
Source: Plus Power, LLC [Kapolei Energy Storage (KES) - 185 MW / 565 MWh - alancing Oahu’s electric system for HECO
🙂
☑️ #261 Sep 18, 2024
GM EV Drivers Will Have Access to Even More Public Chargers with the Addition of Over 17,800 Tesla Superchargers
news.gm.com: [Excerpt] DETROIT – Beginning today, General Motors (NYSE: GM) will open up access to more than 17,800 Tesla Superchargers for its customers, with the use of a GM approved NACS DC adapter, a move that will help accelerate fast and convenient charging options for current and future EV drivers.
With the addition of the Tesla Supercharger Network, GM customers will have access to more than 231,800 public Level 2 and DC fast chargers in North America – a number which will grow as GM continues to help accelerate the deployment of infrastructure in communities and heavily-traveled corridors through its various charging collaborations. GM approved NACS DC adapters will first be made available to customers in the United States, followed by availability for Canadian customers later this year.
@NoPriorsPodcast: No Priors Ep. 80 | With Andrej Karpathy from OpenAI and Tesla.
Andrej Karpathy joins Sarah and Elad in this week of No Priors. Andrej, who was a founding team member of OpenAI and the former Tesla Autopilot leader, needs no introduction. In this episode, Andrej discusses the evolution of self-driving cars, comparing Tesla's and Waymo’s approaches, and the technical challenges ahead. They also cover Tesla’s Optimus humanoid robot, the bottlenecks of AI development today, and how AI capabilities could be further integrated with human cognition. Andrej shares more about his new mission Eureka Labs and his insights into AI-driven education and what young people should study to prepare for the reality ahead.
[Translated text on Mac] The so-called "Giga-Train" is the first all-electric, battery-powered train in Berlin and Brandenburg, as it is called by both the US car manufacturer and the Niederbarnimer Eisenbahngesellschaft (NEB) as operator. The cars offer 120 seats each. Since two are attached to each other for Tesla, 500 people can be transported per trip including standing places. In addition, the trains have bicycle compartments, ground-level access and an information system for passengers.
Theresa Eggler, project manager at Tesla, told rbb on Wednesday: "We are particularly pleased that the Tesla train shuttle is now battery-electric, because it is simply in line for us to comply with our corporate mission: to accelerate the transition to renewable energies." According to Eggler, the new railcars now save 50 tons of CO2 per week compared to the old train type.
![[FIG. 1A: Block diagram illustrating an example autonomous vehicle which includes a multitude of image sensors an example processor system] [FIG. 1B: Block diagram illustrating the example processor system determining object/signal information based on received image information from the example image sensors]](https://substackcdn.com/image/fetch/$s_!Cg_W!,w_720,c_limit,f_webp,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F2774183d-34ef-4ce6-b236-cdaec44b63a9_3300x2550.jpeg)
![[FIG. 1A: Block diagram illustrating an example autonomous vehicle which includes a multitude of image sensors an example processor system] [FIG. 1B: Block diagram illustrating the example processor system determining object/signal information based on received image information from the example image sensors]](https://substackcdn.com/image/fetch/$s_!xOjA!,w_720,c_limit,f_webp,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2Fb53b95f9-05d1-4941-ac33-88f377f69ce9_3300x2550.jpeg)
