Get Your Weekly Digest

Receive the best podcast summaries and insights delivered to your inbox every week.

We respect your privacy. Unsubscribe anytime.

We Study Billionaires - The Investor’s Podcast Network

TECH003: Elon Musk's Tesla Robotaxi, Optimus, and more w/ Cern Basher (Tech Podcast)

with Cern Basher

Published October 1, 2025
Visit Podcast Website

About This Episode

Host Preston Pysh interviews investor and technologist Cern Basher about Elon Musk's ecosystem of companies, focusing on Tesla's pivot away from the Dojo training supercomputer toward custom inference chips, and how this underpins autonomous vehicles and humanoid robots. They explore the economics and deflationary impact of Tesla RoboTaxis and autonomous trucking, the massive potential of the Optimus robot to transform labor and corporate balance sheets, the role of Tesla Energy in enabling abundant power, and how these automation trends connect to Bitcoin as a long-term treasury asset in an AI-driven world.

Topics Covered

Artificial intelligence Electric vehicles Autonomous vehicles Robotics SpaceX and rocketry Renewable energy and decarbonization Automation Bitcoin Labor and workforce issues Investing

Disclaimer: We provide independent summaries of podcasts and are not affiliated with or endorsed in any way by any podcast or creator. All podcast names and content are the property of their respective owners. The views and opinions expressed within the podcasts belong solely to the original hosts and guests and do not reflect the views or positions of Summapod.

Quick Takeaways

  • Tesla has effectively shelved its Dojo training supercomputer and is refocusing the Dojo concept around its own highly optimized inference chips, which Cern Basher believes could be a key long-term advantage.
  • RoboTaxis could be roughly 100x more profitable per unit than selling cars, transforming Tesla from a low-margin automaker into a high-margin mobility and services platform provider.
  • Autonomous electric trucking could cut per‑mile freight costs by around 50%, creating a powerful deflationary force across the entire economy because everything depends on transportation.
  • Cern expects humanoid robots like Tesla's Optimus to eventually outcompete human labor on cost and capability, enabling companies to capitalize labor on their balance sheets for the first time.
  • Musk's broader ecosystem-Tesla, SpaceX, Starlink, the Boring Company, X, xAI, and others-is collectively driving toward a world of abundant energy, transportation, intelligence, and labor.
  • Historically, major technological shifts have destroyed specific jobs but ultimately created more and different kinds of work, and Cern argues this pattern may hold even as AI and robotics advance.
  • Tesla's energy storage business (Megapacks) could double effective U.S. generating capacity without adding new generation by smoothing when existing plants run and storing excess output.
  • Cern believes in a future of "universal high income" enabled by extreme productivity and abundance, and argues that Bitcoin is likely to become the only safe place to store capital as fiat currencies respond to deflation with money printing.
  • He suggests Tesla should steadily build a larger Bitcoin position to prepare for the uncertainty around AGI/ASI and the erosion of fiat purchasing power.
  • Optimus production could scale from a few thousand units in the short term to potentially millions or even hundreds of millions per year over time, with each million robots plausibly supporting around a trillion dollars in market value.

Podcast Notes

Introduction and framing of Elon Musk's ecosystem

Show concept and episode setup

Host describes Infinite Tech and today's focus[0:44]
• The show explores Bitcoin, AI, robotics, longevity, and other exponential tech through a lens of abundance and sound money.
• Preston says this episode will dig into Tesla's expansion from carmaker into AI and robotics powerhouse, the economics of RoboTaxis and Robo-trucking, Optimus humanoid robots, the rise and shutdown of Dojo, and automation-driven deflation tying back to Bitcoin.

Guest introduction and credentials

Who is Cern Basher?[1:27]
• Preston introduces Cern Basher as president and CIO at Brilliant Advice.
• He calls Cern one of the smartest voices on Tesla, AI, robotics, and the future of Bitcoin.
Cern's greeting[1:33]
• Cern thanks Preston for having him and says it's good to be there.

Dojo shutdown and Tesla's AI hardware strategy

Background on Dojo and its original purpose

Preston frames Dojo as Tesla's in-house alternative to NVIDIA[1:53]
• He describes Dojo as Elon's project to build Tesla's own chips and own the full AI stack, from training chips to data collection and model building.
Announcement of Dojo shutdown[2:17]
• Preston notes that in August Elon announced he was shutting down Dojo and abandoning it in its original form.
Is the shutdown a concern for Tesla shareholders?[2:28]
• Preston asks whether this is concerning in the grand scheme and what impact it has for shareholders.

Cern's interpretation of Dojo's pivot

Initial reaction vs deeper view[2:41]
• Cern says at first blush it feels like a concern because Tesla worked on Dojo for years and investors were excited about it reducing reliance on NVIDIA's high-margin systems.
• He argues the reality is different: the situation highlights how amazing NVIDIA is and how fast their development timeline is.
NVIDIA relationship and capacity[3:09]
• Cern notes Elon is clearly able to get the NVIDIA allocation he needs, and NVIDIA CEO Jensen Huang seems impressed with Elon's ability to stand up data centers quickly.
• He calls Tesla a really nice customer for NVIDIA.

Inference chips vs training supercomputers

Tesla's history with inference chips[3:27]
• Cern explains Tesla has always made its own AI inference chips that go in the car: Hardware 3, now AI 4, soon AI 5, and eventually AI 6.
• Historically TSMC has manufactured these chips for Tesla, and AI 5 is moving to Samsung.
Why inference is critical[3:53]
• Cern says Elon recognizes that building the model is one thing, but running it efficiently on a vehicle or robot is another.
• Ultimately Tesla will need billions of inference chips, since every vehicle or robot needs local compute.
Pivot of Dojo from training to inference[4:13]
• He believes Dojo has shifted from a standalone training supercomputer project to focusing on inference-side hardware.
• Tesla knows exactly what they want from inference chips because they build the software that runs on them, whereas NVIDIA doesn't know Tesla's exact needs.
• Cern is actually excited about the new direction and expects a new Dojo system based on Tesla's own inference chips, potentially stacking many inference chips into a training system.

Explaining inference for listeners

Preston's lay explanation of inference[4:53]
• Preston frames inference as the chip taking inputs and quickly generating useful outputs from a pre-trained model, such as steering and braking decisions in a car based on imagery.
• He suggests the specialized inference step-the quality and speed of outputs-may be the real secret sauce for driverless cars and humanoid robots.
Why inference must be on-device[5:35]
• Cern points out a self-driving car cannot wait for instructions from a central data center; decisions must be made in milliseconds on the vehicle.
• He says the same applies to humanoid robots and any AI system acting in the real world: operation cannot rely on the data center for real-time control.
• He emphasizes energy efficiency and notes Elon's idea of "macro hard" as a competitor to Microsoft: instead of pre-developed software, AI runs software on the fly, which again demands powerful and cheap inference chips.

Elon's execution moats and system-level shift toward abundance

Relationship with NVIDIA and possible deal dynamics

Speculation on chip allocation and Dojo shutdown[7:19]
• Preston wonders how Elon convinced Jensen to allocate so many chips to scale up xAI quickly and whether Elon effectively promised to stop competing in training hardware by shutting down Dojo in exchange for supply.
• Cern guesses the decision evolved over time as Tesla saw they couldn't keep pace with NVIDIA on training systems and team members started leaving, making the project difficult.
• He suggests Elon saw another pathway by focusing on an amazing inference chip that can be scaled into a training supercomputer, without necessarily needing any explicit "we'll do this if you do that" deal with Jensen.

Operational excellence and "the machine that builds the machine"

Preston on production-line execution as Tesla's moat[8:29]
• Preston says a major competitive moat in this space is execution of complex production lines, an area where Elon shines but which markets often undervalue.
• He notes Elon has many production lines in his "stack" and calls the recursive idea of building the machine that builds the machine "insane" and unprecedented in business history.
Cern on Tesla's manufacturing innovation[9:37]
• Cern says Tesla can redesign production lines and build their own machines to manufacture vehicles more efficiently.
• If no external company provides needed equipment, Tesla builds it themselves, and others then try to copy them.

Broader system shift driven by Musk's companies

From manual to hyper-automated manufacturing[10:18]
• Cern characterizes Tesla and SpaceX as moving us from a world of manual manufacturing (including conventional robots) to hyper-automated manufacturing.
• He cites the Starlink terminal factory in Bastrop, Texas, where raw materials go in one side and finished terminals come out the other, highly automated.
• He expects similar manufacturing for humanoid robots, with dozens of factories around the world eventually building them.
SpaceX and multi-planet economy[10:45]
• Cern says SpaceX is taking us from a single-planet economy to a multi-planet economy, though we are in very early stages.
Tesla Energy and decarbonization[10:56]
• He notes Tesla Energy is moving us from fossil fuels to renewables, which he calls very important.
The Boring Company and 3D transportation networks[11:06]
• Cern contrasts today's two-dimensional urban road networks with a future 3D system enabled by tunnels from the Boring Company.
• Preston mentions being told that work has already started on a Boring Company tunnel in Nashville out to the airport, adding to the well-known Vegas project.
• Cern explains tunnel boring machines currently move at slower than a snail's pace and need innovation to become faster and safer, including removing humans from the tunnel during operation.
• He envisions cities becoming more beautiful and quiet if much vehicle traffic moves underground, though full acceleration might be 10-15 years out.

Toward abundance: energy, transportation, intelligence, labor

Costs trending toward zero[12:53]
• Cern argues these technologies are driving the cost of energy, transportation, intelligence (via AI), and labor (via humanoid robots) down, trending toward zero over time.
What abundant energy enables[13:23]
• With unlimited energy, he says we could power all data centers, power the Bitcoin network without energy-usage worries, and desalinate unlimited water to eliminate water scarcity.
• He suggests we could even grow forests in deserts and make food more abundant.
Elon as a driver of system transition[13:56]
• Cern says Elon, through his various companies, is pushing the forefront in many areas and effectively taking us from one system to another.

Labor displacement, historical analogies, and geopolitical race

Concerns about job loss and UBI

Preston voices common fear of being displaced by robots[14:36]
• Preston says naysayers worry that humans won't be able to compete with Optimus-like robots, raising questions about how they add value and get paid.
• He notes this leads to universal basic income discussions and points his wife as someone who consistently raises this concern when he gets excited about the tech.

Historical perspective on technological disruption

From farm labor to tractors[15:10]
• Cern compares current fears to farmers facing the advent of tractors, noting they couldn't foresee new jobs that would emerge as farm work was automated.
Switchboard operators and other displaced roles[15:34]
• He cites the example of three quarters of a million women who were switchboard operators being put out of work by the automatic switchboard, yet they found jobs in other areas.
Unpredictability of new job categories[16:04]
• Cern emphasizes that historically, each massive technological imbalance has created more jobs than it displaced, but future roles are hard to predict in advance.
• He notes that when the internet was built, few would have predicted YouTube shows like this podcast as a job category.

AI surpassing humans and distribution of benefits

What if AI is better than humans at everything?[17:02]
• Cern entertains a darker scenario where AI and robots can do all labor and knowledge work better than humans, including jobs like his in financial advice.
• He says in that world, humanity is better off in aggregate from cheaper, more efficient services, but the question becomes how individuals survive and how benefits are distributed.
Universal high income concept[16:52]
• Cern notes Elon has talked about a future of "universal high income" rather than just universal basic income, though Elon has not provided much detail on the transition.
• He argues that if costs of energy, labor, and transportation drop, we should be able to produce everything in massive abundance, supporting this idea.
Analogy to telecom pricing[17:48]
• He compares this to long-distance telephone calls shifting from per-minute pricing to flat monthly fees because costs became "too cheap to meter."
• He imagines food following a similar pattern: if food becomes abundant and cheap, people might just pay a flat monthly fee to a grocery store and take whatever they want.

Optimism vs "this time is different"

Preston on contrarian view that AI is unprecedented[18:22]
• Preston notes contrarians argue this time is different because we are building intelligence smarter than humans, unlike past technologies.
• He self-describes as an optimist and suggests it's historically a "fool's errand" to assume new opportunities won't emerge, given repeated past patterns.

Geopolitical competition and inevitability of pursuing tech

Pandora's box and US-China technology race[18:51]
• Cern says AI and these technologies are like Pandora's box-now open and impossible to close.
• He argues that if the U.S. doesn't pursue technologies like Bitcoin, AI, rockets, and digital money, China will; he asks rhetorically whether we want China to have superior AI, rockets, or digital security.
• He notes Elon has warned that if the U.S. loses superiority in drones or energy production, it becomes vulnerable and uncompetitive.

RoboTaxi economics and autonomous mobility platform

Survey of Musk's projects and systems thinking

List of Elon's ventures[23:21]
• Preston lists Tesla Auto, Tesla autonomy, Robo-trucking and cars, Tesla Energy, Optimus robot, Dojo, SpaceX, Starship, Starlink, xAI, Neuralink, X (with data and payments), the Boring Company, and Hyperloop.
Systems-of-systems value vs siloed valuation[24:36]
• Preston says Wall Street looks at each business individually and misses the emergent value that comes from how all the pieces fit together.
• He credits Cern with being able to piece together the technologies and see the systems-level value.

RoboTaxi as the next major S-curve

Cern's answer: RoboTaxi[25:10]
• Cern says the short answer to which piece takes off next is RoboTaxi, though he expects incredible progress across all Musk projects.

Comparing today's auto business to RoboTaxi model

Current Tesla auto economics[25:23]
• Cern says today Tesla sells a $40,000 car and might make about $4,000 net profit, roughly a 10% margin, possibly including tax credits.
• He assumes an average consumer buys a new car every five years, so Tesla earns that profit once per five-year period per customer.
RoboTaxi unit economics[25:10]
• He says RoboTaxi vehicles will have about half the battery capacity of a Model Y, making them cheap to manufacture.
• Cern estimates that over five years, a RoboTaxi could generate around $200,000 in profit (total over the period, not per year).
• Adjusting for the smaller battery, he equates that to roughly $400,000 over five years per full-battery-equivalent, versus $4,000 from a sold car-about 100x more profitable.
From consumer products to industrial "cybercabs"[26:24]
• He notes today Tesla sells to picky consumers whose tastes change, but in a RoboTaxi world they're making utilitarian cybercabs with advantages from standardization.

Total addressable market: beyond Uber

Disrupting all personal vehicle ownership[27:25]
• Cern says RoboTaxis are not just about disrupting Uber and rideshare; they could disrupt all transportation, including personal vehicle ownership.
• He expects households with multiple cars to first ditch a third car, then the second, and eventually even the first car as RoboTaxi costs drop and availability rises.
• He argues economics will win: if people can save about $5,000 a year on transport costs, they will adopt RoboTaxis and drop insurance, parking, and payment hassles.

Per‑mile cost comparison and consumer savings

Uber vs car ownership vs RoboTaxi[28:05]
• Preston says current Uber rides cost about $2-$3 per mile.
• He estimates owning a $40,000 car for five years and driving an average American amount works out to about $0.65-$0.80 per mile.
• He says Elon thinks RoboTaxis can get down to about $0.25-$0.30 per mile-roughly half the cost of car ownership in his example.

RoboTaxi as a services and advertising platform

Layering additional services on the fleet[28:59]
• Cern emphasizes that once a ubiquitous RoboTaxi network exists, it can deliver food and packages and host many other layered services.
Potential for free rides via advertising[29:15]
• He suggests rides could eventually be free if passengers accept advertising, because operating costs will be so low that advertiser payments could cover them.
• He notes restaurants might pay Tesla to advertise or even pay to have RoboTaxis recommend and drive riders to specific locations.
Integration with xAI/Grok and personalization[30:13]
• Preston points out that xAI and X will give Tesla deep data on user preferences, letting it tailor ads and experiences in the car.
• Cern adds that passengers will be able to talk to the cars, using AI like Grok as a conversational interface.
• He imagines flying to a city like Chicago and telling the car to take you to the best pizza place, with Grok asking questions about preferences and acting as a tour guide en route.
• Preston notes early user feedback online for Grok in cars has been very favorable, with people saying it felt like conversing with a real person who could suggest new local experiences.

Autonomous trucking and deflationary impact on logistics

Tesla Semi production ramp

Factory capacity and timeline[33:34]
• Cern says Tesla is building a Nevada factory with capacity for 50,000 semi-trucks per year.
• He expects the factory to be finished around end of this year or early next year, with ramp-up leading to 50,000 units per year by about 2027.

Economics of autonomous electric trucking

Impact on freight costs and inflation[34:27]
• Cern notes line-haul trucking currently costs about $1.60-$2.00 per mile all-in, with driver costs around $0.70-$0.90 per mile, plus fuel and maintenance.
• He says Tesla's autonomous electric trucks could bring this down to about $0.60-$0.80 per mile, roughly a 50% reduction.
• He calls even a 10% reduction in trucking costs huge; 50% is "absolutely enormous" and would be a massive deflationary force in the economy because trucking affects the price of everything.
Commercial vs consumer adoption dynamics[34:50]
• Cern points out the trucking market is entirely commercial, unlike personal vehicles where individuals can already operate cars cheaply, making displacement dynamics different.
• He says as long as Tesla can undercut the roughly $2-$2.30 per-mile revenue that carriers earn, it will gain market share, and continued reductions will drive many existing operators out of business.

Competitive landscape in autonomy and current FSD performance

Competitors in RoboTaxis

Chinese players and domestic protection[38:54]
• Cern expects China to be hesitant about letting Tesla control the RoboTaxi market and to support homegrown players that already show impressive results.
Waymo's cost disadvantage[39:15]
• He contrasts Waymo, which adds expensive sensors and hardware to existing cars, with Tesla's approach of cameras plus a computer in vehicles designed for autonomy from the factory.
• He argues Waymo cannot compete with Tesla on cost and may never be able to because it lacks its own manufacturing plants.

Cameras vs LiDAR debate

Elon's stance on vision-only[40:03]
• Preston summarizes Elon's view that if humans safely drive using visible-spectrum vision, cars should be able to do the same without LiDAR.
Sensor disagreement risk[40:26]
• Cern notes Elon has raised a risk that if two sensors (e.g., camera and LiDAR) disagree-such as one seeing a wall and the other a plastic bag-the system must resolve it, and a wrong resolution could cause problems.
• Preston is skeptical that extra sensors necessarily degrade safety, arguing that AI typically benefits from more information if trained properly.

Safety, weather, and current FSD rollout

Handling of adverse conditions[41:55]
• Cern says autonomous vehicles can and do operate in rain up to a point, but will stop if conditions are unsafe, whereas humans often keep driving.
• He argues this will make travel safer overall, even if cars shut down in some conditions.
Current RoboTaxi pilots and cautious rollout[42:25]
• Cern says Tesla is operating pilot RoboTaxi services in Austin and Silicon Valley, and testing in many other places.
• He stresses Tesla must be extremely careful not to rush deployment; being six or twelve months later than expectations is acceptable compared to the risk of problems.
FSD v14 and "sentient" comments[43:23]
• Cern says Elon has talked about FSD version 14 increasing model parameters by 10x and described the car as seeming "sentient" in internal testing.
• He notes Elon sees the AI effectively nested into a body (the car), and the community is eagerly awaiting the public release.

Evidence and anecdotes of FSD safety

Crash rate vs human drivers[43:59]
• Cern says based on data Tesla shares, the cars show lower crash rates per mile than human drivers, and by a significant margin, though external observers lack full underlying data.
Cybertruck highway incident[44:15]
• Cern recounts driving his Cybertruck on a Tennessee highway at about 75 mph when the truck suddenly accelerated.
• A car going roughly 140 mph approached from behind in the right lane and cut left; his Cybertruck anticipated a possible rear-end collision and accelerated to lessen potential impact before slowing back down after the car passed.
• He says he hadn't even seen the speeding car until it passed, illustrating machine perception and reaction beyond typical human capability.
Other anecdotal safety examples[45:53]
• Cern shares reports of FSD cars slowing for children emerging from behind vans-children the human driver did not see-because the AI saw the child approaching the road.
• He acknowledges there are also instances of odd behavior in certain edge cases, so the system is not yet perfect.

Tesla Energy and capital allocation in an AI-driven world

Megapacks and grid-scale storage

Design and role of Megapacks[46:10]
• Cern describes Tesla's Megapacks as large battery units roughly the size of a box that fits on a flatbed trailer, intentionally big due to weight and logistics.
• They allow storage of energy produced by intermittent sources like wind and solar for later use.
Doubling effective U.S. generation without new plants[46:58]
• He cites Elon's statement that with enough battery storage, the U.S. could double its generating capacity without adding new generation, by running existing plants at higher capacity and storing unused output.
• He says adding storage to existing infrastructure is an "easy win" to accelerate energy availability.

Tesla, Bitcoin, and treasury strategy

Tesla's prior Bitcoin purchase and Elon's conditions[48:43]
• Cern recalls Tesla buying Bitcoin around February 2021 when its market value was near $1 trillion.
• He notes Tesla later sold some and Elon remarked he wanted to know when more than 50% of Bitcoin's energy mix was renewable.
Advocacy for more Bitcoin on Tesla's balance sheet[49:45]
• Cern says he has been agitating for Tesla to acquire more Bitcoin-not necessarily to become a full Bitcoin treasury company, but to build a larger position.
• He argues that approaching a world of AGI and ASI will be highly uncertain, and companies should go into it with as much capital as possible.
• He believes technological deflation from AI and robots will push fiat currencies toward heavy money-printing responses, making Bitcoin the only safe place to store capital.
Tesla's current cash and Bitcoin position[50:58]
• Cern notes Tesla has about $37 billion in cash and around $1 billion in Bitcoin.
• He supports Tesla reinvesting heavily in its business instead of buying back stock, but wants to see Bitcoin holdings grow over time.
Critique of NVIDIA's capital allocation[51:52]
• He criticizes NVIDIA for aggressively buying back stock while the capital needs for AI infrastructure are skyrocketing every quarter.
• He thinks giving away capital now may cause NVIDIA to miss future opportunities, though he doesn't view it as existentially risky for the company.

Optimus humanoid robot and the capitalization of labor

Engineering difficulty and development timeline

Relative difficulty vs Starship[52:01]
• Cern says Elon has stated that designing a humanoid robot hand is insanely difficult and that Optimus is the second-hardest problem he's working on, after Starship.
Near-term production expectations[53:04]
• Preston says Tesla originally targeted building 5,000 Optimus units this year and assembling 1,000 by June, but redesign work (like on the hand) has delayed the ramp.
• He cites updated ranges: 2,000-5,000 units by end of this year, 20,000-50,000 by end of 2026, over 100,000 units by 2027, and about 1 million per year by 2030.
Cern's view on first factory deployments[54:37]
• Cern says bots already do useful work like picking and moving boxes, relevant to many factory and warehouse roles.
• He notes Tesla had planned 5,000-10,000 robots deployed in Tesla factories and suppliers but pushed back after hand redesign, and he now expects more earnest production starting around 2026.
• He mentions Elon's rule of thumb that it often takes three versions before a robot is truly commercial-ready, likening it to iterative iPhone improvements.

Economic superiority of humanoid labor

Robots outcompeting humans on cost and performance[56:01]
• Cern believes for the first time in history we are developing something-humanoid robots-that can eventually outcompete human labor at all levels.
• He notes robots can be "craftsmen" all the time with no sloppy work, and can work 22-23 hours a day, so even at one-third human speed they can outperform humans over 24 hours.
• He anticipates robots will become more capable than humans over time, not just equal.
Robot-as-a-service vs direct sales[57:18]
• Cern estimates an Optimus could generate roughly $150,000 per year in revenue per bot in early deployments.
• He advocates for "robot as a service" pricing so Tesla can share upside as bots become more capable, rather than selling hardware outright and missing future value.
• He muses that Tesla could alternatively charge for software upgrades that raise the hourly wage-equivalent work the bot can perform.

Capitalizing labor and impact on corporate value

From un-capitalized employees to capitalized robots[58:32]
• Cern explains that human employees, although essential, do not appear as capitalized assets on company balance sheets, unlike players in some sports clubs.
• With robots, he says, companies can capitalize labor for the first time because robots are assets that can be valued and financed.
Back-of-the-envelope valuation math[58:36]
• He gives an example: if a robot earns $25,000 per year in profit and the market applies a P/E of 40, that robot supports about $1 million of market value.
• Thus, every million robots could correspond to about $1 trillion in new market cap for whoever owns or controls them.
• He notes Elon has talked about making billions of robots, suggesting enormous long-term capitalization potential.
Optimus as majority of Tesla's future value[59:30]
• Cern has previously argued Optimus could become about 80% of Tesla's market value by 2040 or even earlier; he says Elon has recently echoed that view.
• He notes the global labor market is roughly 50% of world GDP, and shifting a large portion of that to robotic labor could "create market value out of thin air."

Production scaling and early vs long-term use cases

Scaling relative to car production[59:21]
• Cern says an Optimus weighs about 3% of a Model 3; if Tesla can produce 10 robots per car, then 1 million robots is like making 100,000 cars-relatively easy for Tesla.
• If they can produce 20 robots per car, then 1 million robots would be equivalent to 50,000 cars; he notes Tesla made that many Cybertrucks in the first year despite complexity.
Initial deployments and natural attrition of human roles[59:17]
• Cern expects early deployments in factories and large retailers like Walmart, doing dirty, dangerous, or boring jobs with high turnover, such as stocking, unloading, and moving goods.
• He points out Walmart's turnover is about 40% per year, so many roles can be replaced through attrition without firing people.
Long-term consumer and space applications[1:00:00]
• He anticipates strong eventual demand for home robots as caregivers for the elderly and helpers or companions for kids.
• He also highlights off-planet uses: in his view, space is like a new continent, and we will send bots, not people, for tasks like lunar work or asteroid mining.

Closing remarks and where to follow Cern

Wrap-up and future appearances

Preston's appreciation and suggestion to reconvene[1:00:13]
• Preston says he thoroughly enjoyed the conversation, praises Cern's breadth of knowledge, and suggests doing this quarterly in the future.

Cern's contact information

Where to find Cern online[59:51]
• Cern tells listeners to find him on X at @cernbasher and clarifies that "Cern Basher" is his real name.
• Preston says they will link to Cern's X account and his firm Brilliant Advice.

Lessons Learned

Actionable insights and wisdom you can apply to your business, career, and personal life.

1

Owning and optimizing the critical infrastructure layer-like Tesla's custom inference chips-can create a durable competitive advantage because it tightly couples hardware with the software that runs on it and lets you scale independently of external suppliers.

Reflection Questions:

  • • Where in my business or work do I rely heavily on a third party for a core capability that I could eventually internalize or better integrate?
  • • How could tighter integration between the 'hardware' and 'software' parts of my operations (people, tools, processes) unlock step-change efficiency or new products?
  • • What is one dependency I should start mapping and de-risking over the next 6-12 months to strengthen my strategic position?
2

Major technological shifts usually destroy specific job categories but create entirely new ones that are hard to foresee, so resilience comes from adaptability and a willingness to move toward where value is emerging rather than defending old roles.

Reflection Questions:

  • • Which parts of my current skill set are most vulnerable to automation, and which are hardest to replicate with AI or robots?
  • • How can I start experimenting with adjacent roles or technologies now so I'm positioned to benefit from, rather than be displaced by, these shifts?
  • • What learning habit (course, project, or partnership) can I commit to this quarter to keep myself on the frontier of change in my field?
3

When a platform can drive the marginal cost of a service toward zero-like RoboTaxis or autonomous trucking-it doesn't just undercut incumbents; it enables entirely new business models, such as ad-subsidized transport or bundled logistics services.

Reflection Questions:

  • • In my industry, what core cost driver might be slashed by an emerging technology, and what new offerings would that make possible?
  • • How could I redesign my product or service if usage were nearly free at the margin and revenue came from complementary layers (ads, data, subscriptions, ancillary services)?
  • • What small-scale test could I run this year to explore a 'platform plus services' model instead of a one-time product sale?
4

In a world where automation and AI create powerful deflationary forces, capital allocation becomes even more critical, and holding scarce, non-debasable assets like Bitcoin may be a strategic hedge against aggressive money printing.

Reflection Questions:

  • • How exposed is my wealth or business to currency debasement or policy responses to technological deflation?
  • • What mix of productive reinvestment, cash reserves, and harder assets would give me both flexibility and protection in a volatile AI-driven economy?
  • • What is one concrete step I can take this year to audit and, if needed, rebalance my long-term stores of value?
5

Thinking of labor as a capitalizable asset-via robots or software-invites a different mindset: design work so that repeatable, high-volume tasks are done by scalable systems, while humans focus on higher-leverage, uniquely human contributions.

Reflection Questions:

  • • Which recurring tasks in my organization are so standardized that a robot or software agent could eventually perform them?
  • • How might my role change if those tasks were fully automated and I was freed to focus only on creative, relational, or strategic work?
  • • What process could I start documenting and systematizing now so that it's ready to be automated or delegated in the next few years?

Episode Summary - Notes by Kendall

TECH003: Elon Musk's Tesla Robotaxi, Optimus, and more w/ Cern Basher (Tech Podcast)
0:00 0:00