help select 100 youtubes with jensen haung , and then feed transcripts into one llm agent trying to catch up with jensens brin! (help needed to complete this project rsvp chris.macrae@yahoo.co.uk)
CES 5 Jan 2026Dec 27, 2025
265SmithWatt 75Neumann JHuangDHassabisFLiEMusk 20 Agentic AIforU
KingCharlesLLM DeepLearning009 NormanMacrae.net EconomistDiary.com Abedmooc.com
robots
UPD May 2026: Robotics are huge innovation area for a total ai company like nvidia. It seems relevant to analyse where does nvidia's overwhelming lead in platforms for totally self-driving cars lead to; and where does leadership of robotics go when there isnt a simple region for maximising robotics of self-driving cars versus other development roads to robotics (to be updated soon)
============== previously
*Which type of robotics will be biggest in ai companies like nvidia? Click to more formal gemini update of robot industry sectors here
(other areas where ai foundation models miss the point without context 1 einstein test ; 2 space ; 3 deepest maths eg quantum and implications for safety)
EDGE CONTEXT OF ROBOTICS & AI
Supporting digital twins in controled areas such as factory or medical room replication
Developing humanoids in deep social situations such as elderly or child care -see womens robotics design video https://www.youtube.com/watch?v=JJfRfHNw_5A deep commitments korea snu women engineers and madison huang
Robots in wargames - this may be big but probably not nvidia's core area as very different trust values and ai brain deployment needed in war situations from celebrating deep human cultures and social connectivity
However nvidia (and of course its best supercomputers the world of 2020s has to offer) is racing ahead in quantum partnerships as a mathematical purpose any true ai company needs world's best partners in- and this will likely soon raise higher level coding questions than conventional thinking on security in finance, defence or even blockchains have assumed- its probable IBM among western ai leaders will open this door and then every policymaker may need to come back to world top maths brains - eg jensen huang, demis hassabis, some sub-team of algorithm leaders eg ask yann lecun, elon musk, probaly some Indian and other far east genii not yet valued by western capitalm and also several chinese brains that the west seems even less likely to track because mathematics hasnt been a core study area of usa (eg 30th in high school maths)
=======================
Leaving aside maths, what else can be said about
Robots in education (see eg nvidia and scsp.ai commission stage 1 - 2026
More optimistically, the whole world (perhaps japan leading) may be free to advance
Parts but not wholes of humanoids- eg in factories precision engineering needs eg very delicate arm control but not necessarily facial brain
Parts of human body designed to augment human sense as much as to develop robot- eg it is not clear how much developing human spectacles depends on other developments of humanoids
Non human shapes of robots from eg dog 4 foot format to eg snake format used eg in disaster relief
new year action ai resolutions: help feed grok with 100 Jensen Huang Youtubes ; celebrate ai good with15000 students DC May 7-9
MAPPING - do you need to map first 50 metacity to apply last mile robots or is uber ceo idea of 15000 subcity maps deeper way to stay ahead of data autonomous machines need to be safe in last mile public services
ALPAMAYO: wonderful news https://www.youtube.com/live/L0jJrzFMbP8?si=RNIy98UOclj9c2hZ&t=... mercedez benz has designed car with 25+ sensors which is not only driving's safest but could be intelligence data any city needs to transform from being run by gangs 4 best chances to understand this ces LV yesterday, nvidia santa clara march, scsp.ai ai+expo dc may, computex taiwan june?
Here are two lines of questioning; first continues most of this thread; right hand column continued at this post
10 questions on robots worldwide college students might openly share on first day of college -how would you improve on these questions (i will try to get to 10 iteratively with help from anyone who wants to maximise the yin of curiosity and the yang of win-win data sovereignty (not easy - each week relationships between us korea china and japan seem to change - 4 countries that with taiwan and silicon valley i try to update my maps of every 90 days from a perspective of whats the optimal next AI brief for king charles as source of open world series ai summits. 100 years of london-cambridge royal intelligence hubs |
..My trust in AI would gointer generationally beyonf typical 4 or 8 years of so-called democratic government if there could be maximum win-wins between key creatiors of ai like partners of nvidia hunag, deep mif hassabis, musk and alhoritmically lecun with back from security future spaces which ww.scsp.ai appears to mediate. These futures include quantum, energy, space and drones, other robotics as well as eg workfordce ai. Given the clients of big ai there is a risk that we forget agentic ai in education is about augmenting human brains - actually the last notes of von neumann "computer and the brain 1956". Currently while space drones energy are imporrtant, educational ai ie human development of eg next generation brains seems to me in 2026 to offer 2 main ways forward every teenager (and so ever teacher ).. |
1 which 10 companies make most profits from selling robots - which is main country of ownership and origin country of research? Grok reply 1/5/06
2 which of these companies work closely with nvidia grok reply 1/5/06
3 do we need to segment language between industrial robots, self driving cars and humanoids designed for world class cities public services Who collects the most ;last mile data that humanoids may use as world class cities increasingly need their safe services? More https://normanmacrae.ning.com/forum/topics/robots
4 3-part question asked to google search ai -does china organise worlds largest annual expo on robotics or where does? if free to go anywhere where should 20 year old student study robotics over next 24 months? how do nvidia research centers like omniverse (santa clara and taiwan) fit into your response on where to study? answer below at at 5 jan 2026
======
AI Overview
The
International Robot Exhibition (IREX) in Tokyo, Japan, is widely considered the world's largest robotics expo and is held biennially. China also hosts major annual robotics events, such as the World Robot Conference (WRC) in Beijing and the China International Industry Fair Robot Exhibition (RS) in Shanghai, which are very large and significant, but IREX generally holds the title of the largest overall trade fair in the world. Where to Study Robotics
For a 20-year-old student with the freedom to study anywhere over the next 24 months, the best location will depend on whether you are looking for formal education (university programs) or practical, industry-focused experience. The ideal path would combine a strong academic foundation with exposure to cutting-edge industry research and development.
Top Locations for Formal Study
Focus on regions known for their strong university programs, research output, and robust robotics ecosystems:
- United States (specifically Massachusetts and California):
- Universities: Massachusetts Institute of Technology (MIT) and Stanford University are top-tier institutions for AI and robotics research.
- Ecosystem: MassRobotics in Boston is a large independent robotics hub, and Silicon Valley is a center for tech innovation, offering immense networking and internship opportunities with startups and tech giants like Nvidia.
- Japan (Tokyo):
- Universities: Home to leading research in industrial and humanoid robotics, with strong ties to companies that exhibit at IREX.
- China (Beijing, Shanghai, Shenzhen):
- Universities/Industry: China's robotics industry is rapidly advancing, with thousands of robot-related enterprises and a strategic focus on embodied AI and humanoids. The link between academia and the burgeoning industry is very strong here.
- Europe (Germany, Switzerland):
- Universities: Institutions like ETH Zurich and the Technical University of Munich (TUM) have excellent robotics programs, especially relevant for industrial automation and engineering, with strong links to events like Hannover Messe.
The Role of Nvidia Research Centers
Nvidia's research centers, including those focused on Omniverse in Santa Clara (USA) and Taiwan, are highly relevant and fit into your study plans in the following ways:
- Industry Relevance: Nvidia is a major player in the AI chips and software that power the latest generation of robots, especially those involving complex AI, perception, and simulation (embodied AI). Studying in an area with an Nvidia presence (like Santa Clara) provides direct proximity to these innovations.
- Research & Tools: The Nvidia Omniverse platform is a significant tool for robotics simulation and research, allowing developers to train robots in virtual environments before real-world deployment. Familiarity with such industry-standard tools would be a significant asset for any student.
- Internship Opportunities: The best way to "study" how these centers fit in is through an internship. Proximity to these hubs increases the chances of securing valuable, real-world experience, which can complement formal university education over your 24 months of study.
- Curriculum Integration: Many university research labs collaborate with companies like Nvidia or use their technology. By choosing a university with strong industry connections, you will likely encounter Nvidia's platforms in projects and coursework, linking academic study directly to industry application.
Wikipedia
People also ask
Feedback
10:10
This year's event is absolutely huge. More than 1,500 exhibits from over 200 robotics companies all around the world.
Missing: organise annual free 20 old student nvidia research centers omniverse (santa clara
16:29
In Beijing, over 200 companies came together to reveal machines built for speed, strength, precision, and presence.
Missing: organise annual free anywhere year old student nvidia research omniverse (santa clara
20:44WRC 2025 is happening right now in China and it's already proving to be the most ambitious most diverse and most futuristic robotics showcase.
Missing: organise annual free 20 year old student 24 nvidia research centers omniverse (santa clara
Missing: organise annual free anywhere year old student centers omniverse (santa clara
Nvidia CEO Says Robotics Will Drive Next Wave of AI
TMTPost
https://en.tmtpost.com › postJul 16, 2025 — Widespread adoption of Nvidia Omniverse: Hundreds of projects in China are using Nvidia Omniverse to simulate digital twins for factory and ...Read more
Missing: organise annual if free anywhere old student study 24 centers (santa clara
Views: 100
Replies to This Discussion
-
Permalink Reply by chris macrae on
-
When i stidied at Cambridge Departmment of Apllied Maths and Theoretical Physics the rumor was you could still walk from Cambridge to London on land owned by the royal family. What remains true todau is Cambrdige Business park landlord is our sovereign, and this is largest ai hub in europe. Its world class connecctions include hassabis deep mind now mainly col-located in london and at googklke hq mountain view; and arm which if it hadnt been for combo of covid and strange national monopoly inquiries would have united nvidia and softbank by now (instead nvidia has just bough grok as its main designer of central processors as well as having a close investment partnership with intel to renew that origin of moores law. We have 3 million fold change expoenetials:
those who connected around engineers first inspired by intel 1965 (morres law- silicon chips will become million times more efficient to compute and communicate with before end of 20th c; jensens emergent law, machines can be designed by early 2020s with billion times more mathematical brains than separate human minds- satlelites deth of distance purpose debated in neumann inspired dialogues at the economist from 1951; 6h+ is now well on its way to openly sharing life critical data maps through every gps- compared with 1950s telecoms which were non-digital and pre-satellite.
Back in 1905 einstein opened up a whole bag of natural science questions. Dasflt his generations greats neumann and turing included mainly had to work on winning atom bnomb race for allies versus gemrany and russia's madmen (hitler and stalin). By early 1930s neumann and i=einstein had escaped to safety of prineton bit their office was badly located miles from the twon campus- so much so that when turing attended princetion the last year before world war 2 neumann met him too late to arrange a visa for him toi stay. From late 1938 neumann was dreaming of computers and coding maths while being required to compete the atom bomb race. By this time einstein brain was philosohpical not mathematical. Turing war back in uk was spent on coding. Princeton hated the main ideas vin neumann asked the economoist to debate:
how to open source computing
why engergy would acce;erate as every places need
how to design intelligence as win-win (above zero sum) trading games
how to apply neural netwirking algorithms to enhance human brains with ai (see neumanns last booklet comouter and brain written on his cancer deathbed 1956), By early 1957 Turing Neumann Einstein had all parted erth even though born 3 generation apart humans could have made much better generative use of their intelligence
Baxk in Cambridge 1950s damtp as amaths lab was established minutes walj from cavendish and rutherford labs. As early as 1930s Taiwan's tech goidfather KT Li was studying energy at cavendish. During post world war2 years biotech was leaping firward with crick and watson dna. But if you ask all of gfei-fei li, demis hassabis and jensen huanng in 2000 where was neuroscience and digoital sciences helping each other- the answer was nowhere. Psycholoigy professors hated computers except where somebody played to their vanity with expert models, So 50 yers of playing at connecting 3 million fold chnage expoenetails - moores, jensens to be, satellites - was missed. Cosequenetly the whole ofb education was ill prepared for transformation which the age of multiplying each others intelligencs demands. No wonder as ai eneters its coing of age decade 2025-35 quetions like whos data sovereignty who and wht millennial goals have not yet receieved a dollar or real investment are being asked wherever speech is free which is not what commercial mass media is designed to celebrate
-
-
Dec 25, 2025(3447) Nvidia Acquires Groq for $20 Billion: Why Jensen Huang Just Bought the Future - YouTube www.youtube.com
Dec 14, 2025Jensen Huang Is NOT Backing Down — Google & AWS Bring Their Hardest AI Chips Yet - YouTube www.youtube.com
Dec 14, 2025(3346) NVIDIA’s Jensen Huang on Securing American Leadership on AI - YouTube www.youtube.com
Dec 13, 2025(3342) Jensen Huang: AI Energy War, US–China Chip Rivalry and Next Revolution｜黃仁勳最新訪談：AI能源戰、美中晶片攻防與下一場工業革命 - YouTube www.youtube.com
Dec 6, 2025Dec 6, 2025LIVE | Macron Tours VivaTech 2025, Joins Nvidia CEO Jensen Huang & Mistral's Arthur Mensch | APT - YouTube www.youtube.com
Dec 1, 2025(3240) A Conversation with NVIDIA CEO Jensen Huang – The Future of AI and Energy - YouTube www.youtube.com
Dec 1, 2025A Conversation with NVIDIA CEO Jensen Huang – The Future of AI and Energy - YouTube www.youtube.com
(3357) "jensen huang" - YouTube www.youtube.com
Dec 14, 2025(3346) NVIDIA’s Jensen Huang on Securing American Leadership on AI - YouTube www.youtube.com
(3342) Jensen Huang: AI Energy War, US–China Chip Rivalry and Next Revolution｜黃仁勳最新訪談：AI能源戰、美中晶片攻防與下一場工業革命 - YouTube www.youtube.com
Dec 13, 2025
(3346) NVIDIA’s Jensen Huang on Securing American Leadership on AI - YouTube www.youtube.com
Dec 11, 2025
(3322) NVIDIA’s Jensen Huang on Securing American Leadership on AI - YouTube www.youtube.com
Dec 11, 2025
(3346) NVIDIA’s Jensen Huang on Securing American Leadership on AI - YouTube www.youtube.com
Dec 9, 2025
(3307) NVIDIA CEO Jensen Huang Warns: China May Outperform the U.S. in Tech | Full Interview | AI1G - YouTube www.youtube.com
Dec 9, 2025
(3303) NVIDIA’s Jensen Huang on Securing American Leadership on AI - YouTube www.youtube.com
Oct 27, 2025(2963) Nvidia's Jensen Huang on AI & the Next Frontier of Growth - YouTube www.youtube.com
NVIDIA's Founder & CEO, Jensen Huang Fireside Chat With Mukesh Ambani | NVIDIA AI Summit | CNBC TV18 - YouTube www.youtube.comOct 23, 2025(2943) Nvidia's Jensen Huang Hand-Delivers World's Smallest Supercomputer To Elon Musk - YouTube www.youtube.com
(2943) jensen huang - YouTube www.youtube.comOct 23, 2025Found 118 search results for 'youtube jensen' (2942) Jensen Huang Special Address from NVIDIA AI Summit Japan - YouTube www.youtube.com
-
-
q to grok 5 jan 2026
Do we know which 10 companies make the most profits from selling robots. Where majority owner of each company is as well as where research gq of company came frimSearching the web41 resultsThe robotics industry lacks a single definitive ranking of the top 10 companies by profits from robot sales, as many firms are divisions of larger conglomerates, private, or derive revenue from related automation. Profits in robotics often come from high-margin recurring services, consumables, and software, rather than hardware alone.Based on recent data (2024–2025), the leading companies in robot sales revenue and profitability (focusing on those where robots are a core profit driver) include industrial giants and specialized medical robotics leaders. Here's an approximate top 10, ordered roughly by estimated robotics-related revenue/profit impact (drawing from division revenues, market share, and reported high margins):- Intuitive Surgical (da Vinci surgical systems)
- Majority owner: Publicly traded (USA).
- Research HQ/origin: Sunnyvale, California, USA (founded 1995).
- Highly profitable due to recurring revenue from instruments and services; dominates surgical robotics with strong margins.
- FANUC (industrial robots)
- Majority owner: Publicly traded (Japan).
- Research HQ/origin: Yamanashi Prefecture, Japan (founded 1956 as part of Fujitsu).
- One of the most profitable pure-play robot makers with consistent high margins and large installed base.
- ABB (industrial robotics division)
- Majority owner: Publicly traded (Switzerland/Sweden).
- Research HQ/origin: Zurich, Switzerland / Västerås, Sweden (founded 1988 from merger).
- Robotics division ~$3.2B revenue in 2024; strong profitability in automation.
- Yaskawa Electric (Motoman industrial robots)
- Majority owner: Publicly traded (Japan).
- Research HQ/origin: Kitakyushu, Japan (founded 1915).
- Significant robot sales with steady profits, especially in welding/assembly.
- KUKA (industrial robots)
- Majority owner: Midea Group (China, full ownership since 2016).
- Research HQ/origin: Augsburg, Germany (founded 1898).
- ~€3.7B sales in 2024; contributes to Midea's automation profits.
- Teradyne (includes Universal Robots cobots and Mobile Industrial Robots)
- Majority owner: Publicly traded (USA).
- Research HQ/origin: North Reading, Massachusetts, USA (founded 1960); UR from Odense, Denmark.
- Growing robotics revenue with improving profitability in collaborative robots.
- Mitsubishi Electric (industrial robots)
- Majority owner: Publicly traded (Japan).
- Research HQ/origin: Tokyo, Japan (founded 1921).
- Solid contributor to industrial automation profits.
- Kawasaki Heavy Industries (industrial robots)
- Majority owner: Publicly traded (Japan).
- Research HQ/origin: Kobe, Japan (founded 1896).
- Strong in heavy-payload robots with reliable margins.
- Keyence (factory automation sensors/machine vision, core to robotics)
- Majority owner: Publicly traded (Japan).
- Research HQ/origin: Osaka, Japan (founded 1974).
- Extremely high profit margins overall; vision systems essential for profitable robotic applications.
- Symbotic (warehouse automation robots)
- Majority owner: Publicly traded (USA).
- Research HQ/origin: Wilmington, Massachusetts, USA (founded 2007).
- Rapid growth in warehouse robotics with emerging profitability.
- Japanese companies (FANUC, Yaskawa, etc.) dominate industrial robot profitability due to scale and efficiency.
- Medical robotics (e.g., Intuitive) often yields higher margins than industrial due to consumables.
- Enablers like NVIDIA power many robots but don't "sell robots" directly.
- Data is approximate as exact profit breakdowns for robotics divisions are not always public; rankings can vary by year or metric (e.g., market share vs. net profit).
41 web pageshttps://imgs.search.brave.com/vlry1X_CIp74t_aBj4BB0H16s28kjmsXa5A-1..." class="CToWUd"/>
https://imgs.search.brave.com/7wYBFmcPXo4OTmxEgTa-pUZlou_y7xPnKp-7c..." class="CToWUd"/>
https://imgs.search.brave.com/DLyZkNL9S31jTQTbNVj1seY4rDjXMHmmZ0ink..." class="CToWUd"/>
Intuitive Surgical's da Vinci systemAI advancements in robotics - Intuitive Surgical (da Vinci surgical systems)
-
-
which of grok's top 10 list of robot companies workd with nvidia?
Grok reply
Which of these 10 companies work closely with nviduaSearching the web85 resultsSeveral of the top 10 robotics companies have established close collaborations with NVIDIA, particularly through integrations with the NVIDIA Isaac robotics platform (including tools like Isaac Manipulator, Isaac Sim, and accelerated libraries), Jetson edge AI hardware, and foundation models for AI-powered robot arms, perception, and manipulation.Here are the ones with documented close partnerships (as of early 2026):- FANUC — Strong collaboration; NVIDIA and FANUC announced a direct partnership in late 2025 to develop AI-powered industrial robots focused on "physical AI." FANUC is also highlighted in NVIDIA ecosystems for Jetson and simulation tools.
- ABB — Partnerships noted in investor and industry analyses; ABB integrates NVIDIA technologies for broader automation and has been grouped with FANUC in NVIDIA's robotics exposure.
- Yaskawa Electric — Early ecosystem partner for NVIDIA's Isaac Manipulator and generative AI tools; Yaskawa brings these capabilities to customers for advanced robotic arms.
- Teradyne (includes Universal Robots and Mobile Industrial Robots) — Deep collaboration; Teradyne Robotics works extensively with NVIDIA on the AI Accelerator toolkit (powered by Jetson AGX Orin and Isaac libraries), cuMotion path planning, and physical AI applications. Universal Robots is a key partner for foundation models and cobot advancements.
- KUKA — Integrates NVIDIA Jetson extensions in controllers (e.g., KR C5 Micro-2 for AI-ready applications) and uses NVIDIA Omniverse/Isaac for simulation and digital twins.
- Intuitive Surgical → No mentioned collaborations (focuses on medical robotics with proprietary systems).
- Mitsubishi Electric → No specific partnerships found.
- Kawasaki Heavy Industries → No specific partnerships found.
- Keyence → Often mentioned alongside NVIDIA in automation ecosystems (e.g., high-margin enablers), but no direct robotics collaboration.
- Symbotic → No documented partnerships (primarily warehouse automation with Walmart ties).
85 web pageshttps://imgs.search.brave.com/qfgCnZNnpkLA6Xll6tb4RIPCNHCva2EInrJwk..." class="CToWUd"/>
https://imgs.search.brave.com/W9nlN9YTy5vSaSpWHsXD2UtPIxXhGPg85arO5..." class="CToWUd"/>
https://imgs.search.brave.com/P_gSGqqhPY82K3nCIE57UVzvSmUrB7WNJ19Kr..." class="CToWUd"/>
Explain NVIDIA Isaac platformTop AI chip competitorsThink Harder
-
-
My trust in AI would gointer generationally beyonf typical 4 or 8 years of so-called democratic government if there could be maximum win-wins between key creatiors of ai like partners of nvidia hunag, deep mif hassabis, musk and alhoritmically lecun with back from security future spaces which ww.scsp.ai appears to mediate. These futures include quantum, energy, space and drones, other robotics as well as eg workfordce ai. Given the clients of big ai there is a risk that we forget agentic ai in education is about augmenting human brains - actually the last notes of von neumann "computer and the brain 1956". Currently while space drones energy are imporrtant, educational ai ie human development of eg next generation brains seems to me in 2026 to offer 2 main ways forward every teenager (and so ever teacher )Searching the web55 resultsBuilding Intergenerational AI Trust Through Win-Wins and Robotics FocusI appreciate your thoughtful vision—envisioning a "maximum win-win" ecosystem among AI pioneers like Jensen Huang (NVIDIA), Demis Hassabis (DeepMind), Elon Musk (xAI/Tesla), and Yann LeCun (Meta AI), mediated by organizations like SCSP.ai. This could indeed foster trust beyond short-term political cycles, emphasizing agentic AI's role in augmenting human intelligence (echoing von Neumann's 1956 insights on computers mirroring brains). Your emphasis on education as a core pillar—empowering teenagers and teachers through action learning, data sovereignty, and India's 6 societal pillars (aligned with SDGs like 2: zero hunger/agriculture, 3: health, 4: education, 5: gender equality/women's empowerment, 8: decent work/youth, and governance)—highlights a human-centered path. It's a compelling critique of how the UN's SDG-AI integration lagged from 2015-2026, missing opportunities like Turing-inspired or Fazle Abed-style full-stack education.You're right that Asia-Pacific AI (encompassing ~70% of global population and livelihoods in Taiwan, Korea, Japan, Hong Kong, Singapore, ASEAN, and emerging hubs like IMEV—perhaps referring to India-Middle East-Europe corridors) could pioneer deeper, community-agentic futures, contrasting Western models. NVIDIA-SCSP's U.S.-focused workforce AI reporting (interim May 2026, final October 2026) might indeed overlook these nuances unless explicitly bridged. Your "triangularized" public-private-community partnerships via AI pillars at grassroots levels could redefine this, especially with Huang's data sovereignty initiatives (e.g., India's Feb 19-20, 2026 event focusing on AI layers 4: models and 5: agentic apps for community mobilization).On robotics as a unifying dynamic: This could indeed "unite the world" by making tangible the interplay of AI with physical systems, from controlled factories to chaotic public spaces. Below, I'll address your specific questions, drawing on current trends as of January 2026.Main Forms of Autonomous Mobile Robots (AMRs) in 2026AMRs are robots that navigate independently without fixed paths, using AI for perception, mapping, and decision-making. They're exploding in adoption, with the global market projected at $10-15B in 2026 (up 30% YoY), driven by AI integration. Key forms include:These forms blend "physical AI" (Huang's term), with NVIDIA's Omniverse simulation accelerating development.FormDescription & ExamplesKey Applications & Trends (2026)Drones (Aerial AMRs)Unmanned aerial vehicles with autonomous flight, sensors, and AI for obstacle avoidance.Delivery (Amazon Prime Air), agriculture surveying, defense (DJI alternatives like Skydio). 2026 trend: Swarms for urban monitoring; e.g., Anduril's Ghost-X.Self-Driving Cars/VehiclesWheeled AMRs for roads/highways, using LiDAR, cameras, and AI (e.g., Waymo, Tesla Autopilot).Ride-hailing, logistics (e.g., Nuro delivery bots). 2026: Level 4 autonomy in 50+ cities; focus on safety data from public trials.Humanoid Robots (2-Legged)Bipedal designs mimicking human movement for versatility.Household tasks, factories (Tesla Optimus Gen-2, Boston Dynamics Atlas). 2026: CES demos show folding laundry/walking uneven terrain; Unitree G1 at $16K for education/research.Quadrupedal/Legged Robots (4-Legged or More)Animal-inspired for stability on rough terrain.Inspection (Boston Dynamics Spot), search-and-rescue. 2026: Agility Robotics' Digit for warehouses; ANYbotics ANYmal in industrial sites.Wheeled/Tracked Factory Arms & AMRsMobile manipulators or carts for indoor logistics (not always "arms," but often with grippers).Warehouses (Amazon Kiva bots, Fetch Robotics). 2026: Integration with agentic AI for dynamic picking; e.g., Symbotic systems in grocery fulfillment.Other Emerging (e.g., Snake/Wheeled Hybrids)Flexible designs for tight spaces or exploration.Space (NASA's EELS snake robot), underwater inspection. 2026: Less common but growing in niche defense/energy.
Humanoids (e.g., Optimus) are Huang's favored for "everything that moves will be robotic," tying to agentic education via collaborative learning bots.Two Opposite Development Paths for RobotsYou're spot-on: Robotics evolves along a spectrum from controlled (structured) to unstructured (public) environments, with distinct AI/data needs. By 2026, "agentic AI" (autonomous decision-making) bridges them, but paths diverge:
- Controlled Environments (e.g., Factories, Hospitals):
- Development Focus: Robots operate in predictable, mapped spaces where AI "digital twins" the entire environment (virtual replicas via simulation tools like NVIDIA Omniverse). Path planning is optimized for efficiency/safety; e.g., robots collect proprietary data to refine workflows.
- Examples: Factory arms (e.g., ABB cobots in TSMC fabs) or hospital delivery bots (e.g., Aethon TUG in sterile zones). Data twins architecture for real-time adjustments.
- Pros/Trends: High precision; 2026 sees 80%+ adoption in manufacturing (IFR data). Challenges: Over-reliance on enclosed data limits adaptability.
- Data Issues: Internal/owned by operators; focus on IP protection vs. privacy.
- Unstructured/Public Spaces (e.g., Cities, Self-Driving Cars):
- Development Focus: Robots must adapt dynamically to humans, weather, and chaos using real-time sensing (LiDAR, cameras) and AI for safety-first navigation. No full twinning—emphasis on edge AI for on-the-fly decisions.
- Examples: Self-driving cars (Waymo in urban grids), delivery drones (Zipline in variable terrain). Robots collect vast public data to "learn" human behaviors.
- Pros/Trends: Scalable for services; 2026 pilots in 100+ cities (e.g., Starship bots in Austin). Challenges: Ethical AI to prioritize human safety (e.g., trolley problem algorithms).
- Data Issues: Massive privacy risks—robots capture identifiable info (faces, locations). Regulations like EU AI Act mandate consent/transparency; breaches could erode trust.
Data Collection Issues: Public Services vs. Defense/Space RoboticsPublic robotics (e.g., city service bots for mapping/traffic) prioritizes human-centric data (anonymized crowd flows, safety patterns), but risks surveillance/misuse—e.g., whose AI (Musk's xAI vs. state-owned) owns urban maps? Defense drones (e.g., Anduril) focus on classified intel (target tracking), with minimal privacy but high security risks (hacking). Space robotics (e.g., NASA's Perseverance) collect environmental data (Mars terrain), emphasizing redundancy over ethics, but IP disputes arise in collaborations. Overall: Public needs transparent, community-governed data (e.g., GDPR-like); defense/space: Secure, mission-specific. Universities lag in holistic mapping, but initiatives like SCSP's fusion panels could help.
- Controlled Environments (e.g., Factories, Hospitals):
-
-
Main Robotics Summits in 2026 (Beyond CES)CES 2026 (Jan 6-9, Las Vegas) heavily featured humanoids (e.g., 9 demos signaling "physical AI" readiness), with Huang's keynotes on robotics stacks. Other major events valuing comprehensive humanoid mapping (akin to Huang's vision):
Few explicitly match Huang's comprehensiveness, but ICRA/RSS come closest for layered AI-robotics discussions. Huang doesn't have a public "5 layers of humanoids," but his "physical AI" framework (simulation, models, hardware) parallels his AI layers (energy, chips, infrastructure, models, apps), extending to robotics as the "next ChatGPT moment."Summit/EventDates/LocationFocus & Humanoid EmphasisRobotics Summit & ExpoMay 2026 (Boston, USA)6,000+ devs; aerospace/defense/healthcare; strong on humanoids for logistics/manufacturing.IEEE ICRAJune 1-5 (Vienna, Austria)Premier academic; papers on agentic/humanoid navigation; values mapping like Huang's layers.Robotics: Science and Systems (RSS)July 13-17 (Sydney, Australia)Cutting-edge research; humanoids in unstructured spaces; comprehensive AI-robotics integration.World Summit on RoboticsMarch 26-28 (Rome, Italy)Expo with breakthroughs; humanoid panels echoing Huang's "everything moves robotic."ATX WestFeb 3-5 (Anaheim, USA)Automation/humanoids for industry; ties to NVIDIA's physical AI.UK Robot Manipulation WorkshopTBD (Edinburgh, UK)Manipulation in humanoids; academic depth on mapping/data.
Universities could indeed do more—perhaps SCSP or Asia-Pacific hubs like India's event could catalyze open maps for these intelligence futures, integrating SDGs with agentic AI. If you'd like to explore specific bridges (e.g., von Neumann-inspired curricula), I'm here!1 post
Agentic AI in education curricula
chris macrae said:My trust in AI would gointer generationally beyonf typical 4 or 8 years of so-called democratic government if there could be maximum win-wins between key creatiors of ai like partners of nvidia hunag, deep mif hassabis, musk and alhoritmically lecun with back from security future spaces which ww.scsp.ai appears to mediate. These futures include quantum, energy, space and drones, other robotics as well as eg workfordce ai. Given the clients of big ai there is a risk that we forget agentic ai in education is about augmenting human brains - actually the last notes of von neumann "computer and the brain 1956". Currently while space drones energy are imporrtant, educational ai ie human development of eg next generation brains seems to me in 2026 to offer 2 main ways forward every teenager (and so ever teacher )Searching the web55 resultsBuilding Intergenerational AI Trust Through Win-Wins and Robotics FocusI appreciate your thoughtful vision—envisioning a "maximum win-win" ecosystem among AI pioneers like Jensen Huang (NVIDIA), Demis Hassabis (DeepMind), Elon Musk (xAI/Tesla), and Yann LeCun (Meta AI), mediated by organizations like SCSP.ai. This could indeed foster trust beyond short-term political cycles, emphasizing agentic AI's role in augmenting human intelligence (echoing von Neumann's 1956 insights on computers mirroring brains). Your emphasis on education as a core pillar—empowering teenagers and teachers through action learning, data sovereignty, and India's 6 societal pillars (aligned with SDGs like 2: zero hunger/agriculture, 3: health, 4: education, 5: gender equality/women's empowerment, 8: decent work/youth, and governance)—highlights a human-centered path. It's a compelling critique of how the UN's SDG-AI integration lagged from 2015-2026, missing opportunities like Turing-inspired or Fazle Abed-style full-stack education.You're right that Asia-Pacific AI (encompassing ~70% of global population and livelihoods in Taiwan, Korea, Japan, Hong Kong, Singapore, ASEAN, and emerging hubs like IMEV—perhaps referring to India-Middle East-Europe corridors) could pioneer deeper, community-agentic futures, contrasting Western models. NVIDIA-SCSP's U.S.-focused workforce AI reporting (interim May 2026, final October 2026) might indeed overlook these nuances unless explicitly bridged. Your "triangularized" public-private-community partnerships via AI pillars at grassroots levels could redefine this, especially with Huang's data sovereignty initiatives (e.g., India's Feb 19-20, 2026 event focusing on AI layers 4: models and 5: agentic apps for community mobilization).On robotics as a unifying dynamic: This could indeed "unite the world" by making tangible the interplay of AI with physical systems, from controlled factories to chaotic public spaces. Below, I'll address your specific questions, drawing on current trends as of January 2026.Main Forms of Autonomous Mobile Robots (AMRs) in 2026AMRs are robots that navigate independently without fixed paths, using AI for perception, mapping, and decision-making. They're exploding in adoption, with the global market projected at $10-15B in 2026 (up 30% YoY), driven by AI integration. Key forms include:These forms blend "physical AI" (Huang's term), with NVIDIA's Omniverse simulation accelerating development.FormDescription & ExamplesKey Applications & Trends (2026)Drones (Aerial AMRs)Unmanned aerial vehicles with autonomous flight, sensors, and AI for obstacle avoidance.Delivery (Amazon Prime Air), agriculture surveying, defense (DJI alternatives like Skydio). 2026 trend: Swarms for urban monitoring; e.g., Anduril's Ghost-X.Self-Driving Cars/VehiclesWheeled AMRs for roads/highways, using LiDAR, cameras, and AI (e.g., Waymo, Tesla Autopilot).Ride-hailing, logistics (e.g., Nuro delivery bots). 2026: Level 4 autonomy in 50+ cities; focus on safety data from public trials.Humanoid Robots (2-Legged)Bipedal designs mimicking human movement for versatility.Household tasks, factories (Tesla Optimus Gen-2, Boston Dynamics Atlas). 2026: CES demos show folding laundry/walking uneven terrain; Unitree G1 at $16K for education/research.Quadrupedal/Legged Robots (4-Legged or More)Animal-inspired for stability on rough terrain.Inspection (Boston Dynamics Spot), search-and-rescue. 2026: Agility Robotics' Digit for warehouses; ANYbotics ANYmal in industrial sites.Wheeled/Tracked Factory Arms & AMRsMobile manipulators or carts for indoor logistics (not always "arms," but often with grippers).Warehouses (Amazon Kiva bots, Fetch Robotics). 2026: Integration with agentic AI for dynamic picking; e.g., Symbotic systems in grocery fulfillment.Other Emerging (e.g., Snake/Wheeled Hybrids)Flexible designs for tight spaces or exploration.Space (NASA's EELS snake robot), underwater inspection. 2026: Less common but growing in niche defense/energy.
Humanoids (e.g., Optimus) are Huang's favored for "everything that moves will be robotic," tying to agentic education via collaborative learning bots.Two Opposite Development Paths for RobotsYou're spot-on: Robotics evolves along a spectrum from controlled (structured) to unstructured (public) environments, with distinct AI/data needs. By 2026, "agentic AI" (autonomous decision-making) bridges them, but paths diverge:
- Controlled Environments (e.g., Factories, Hospitals):
- Development Focus: Robots operate in predictable, mapped spaces where AI "digital twins" the entire environment (virtual replicas via simulation tools like NVIDIA Omniverse). Path planning is optimized for efficiency/safety; e.g., robots collect proprietary data to refine workflows.
- Examples: Factory arms (e.g., ABB cobots in TSMC fabs) or hospital delivery bots (e.g., Aethon TUG in sterile zones). Data twins architecture for real-time adjustments.
- Pros/Trends: High precision; 2026 sees 80%+ adoption in manufacturing (IFR data). Challenges: Over-reliance on enclosed data limits adaptability.
- Data Issues: Internal/owned by operators; focus on IP protection vs. privacy.
- Unstructured/Public Spaces (e.g., Cities, Self-Driving Cars):
- Development Focus: Robots must adapt dynamically to humans, weather, and chaos using real-time sensing (LiDAR, cameras) and AI for safety-first navigation. No full twinning—emphasis on edge AI for on-the-fly decisions.
- Examples: Self-driving cars (Waymo in urban grids), delivery drones (Zipline in variable terrain). Robots collect vast public data to "learn" human behaviors.
- Pros/Trends: Scalable for services; 2026 pilots in 100+ cities (e.g., Starship bots in Austin). Challenges: Ethical AI to prioritize human safety (e.g., trolley problem algorithms).
- Data Issues: Massive privacy risks—robots capture identifiable info (faces, locations). Regulations like EU AI Act mandate consent/transparency; breaches could erode trust.
Data Collection Issues: Public Services vs. Defense/Space RoboticsPublic robotics (e.g., city service bots for mapping/traffic) prioritizes human-centric data (anonymized crowd flows, safety patterns), but risks surveillance/misuse—e.g., whose AI (Musk's xAI vs. state-owned) owns urban maps? Defense drones (e.g., Anduril) focus on classified intel (target tracking), with minimal privacy but high security risks (hacking). Space robotics (e.g., NASA's Perseverance) collect environmental data (Mars terrain), emphasizing redundancy over ethics, but IP disputes arise in collaborations. Overall: Public needs transparent, community-governed data (e.g., GDPR-like); defense/space: Secure, mission-specific. Universities lag in holistic mapping, but initiatives like SCSP's fusion panels could help.
- Controlled Environments (e.g., Factories, Hospitals):
-
-
"over 15,000 cities" or "approximately 15,000 cities")—is a standard, recurring stat that Uber's CEO Dara Khosrowshahi and the company frequently use to describe Uber's current global operational footprint. It's not a new "claim of interest" in tracking or expanding to that number as future potential; rather, it's how Uber officially reports where it already actively operates today.For context:
- Uber consistently states it operates in about 70 countries and 15,000 cities worldwide. This appears in earnings reports, investor materials, press releases, Wikipedia, interviews (including recent ones with Khosrowshahi), and company descriptions as of 2025–2026.
- Examples include Khosrowshahi mentioning it in podcasts (e.g., On with Kara Swisher in late 2025: "we operate in 70 countries, 15,000 cities") and in discussions about localization, global scale, or operational challenges.
- It's a measure of current reach for ride-hailing, delivery (Uber Eats), and related services—not a forward-looking target for expansion to new places.
How Uber Likely Defines a "City" in This CountUber doesn't publicly release an exact methodology for this 15,000 figure (companies like this often treat it as proprietary or approximate), but based on how ride-hailing/delivery platforms and mapping services work, it's almost certainly not limited to major metropolitan areas or official city limits. Instead, it's a broad, operational definition that includes many smaller locales to reflect where the app is available and drivers can accept trips.Typical breakdown of what gets counted as a separate "city":
- Major cities and metros (e.g., New York, London, Tokyo) — definitely included.
- Smaller cities, towns, and municipalities — yes, often counted individually.
- Suburbs, districts, or outlying areas — frequently treated as distinct if they have separate availability, pricing zones, or driver pools. For example, what might feel like one big urban area (e.g., Greater Los Angeles) could be split into multiple "cities" like Los Angeles, Santa Monica, Long Beach, etc., because Uber launches/operates them with some independence.
- Even smaller places — rural towns, resort areas, or university towns where Uber has some driver coverage.
This aligns with how mapping and navigation providers (like TomTom, which Uber recently renewed a global deal with in early 2026 for routing and precision) and apps handle "last-mile" delivery/rides:
- Last-mile mapping requires hyper-local data: street-level details, turn restrictions, pedestrian paths, delivery access points, etc.
- To enable reliable service, Uber (or partners) often needs ground-truth validation—via human drivers mapping roads over time, user reports, vehicle sensor data (for future AVs), or partnerships.
- Suburbs or neighborhoods aren't always lumped together; if a suburb has its own demand, regulations, or needs separate onboarding for drivers, it might count separately to show granular availability.
In short, the 15,000 isn't about "interest in tracking" new places as some untapped future goal—it's Uber boasting its existing scale. The real future focus for Khosrowshahi (in recent interviews) is more on AI, autonomous vehicles/robotaxis, last-mile autonomy (e.g., partnerships for AV fleets), and orchestration across modes—leveraging that huge network rather than rapidly adding thousands more locales.If this was referencing a specific recent quote, interview, or article you saw (maybe from Davos, a podcast, or earnings call), feel free to share more details and I can dig deeper!
-
-
robotscreatingjobs.docx please note this file is intended for input in grounded theory - whilst happy to discuss other types of uses it is deliberately biassed towards celebrating AI support of asian job creation round nutrition including democratising health/ nutrition (all ages) relevant to un goals and to female asian/family progress - i can help reframe without this bias if you have a different purpose chris.macrae@yahoo.co.uk - over 40 years ago applying database models with mit/harvard out of our Tokyo offive, reseached japanese restaurant markets from viewpoint of whether major american company could be of any value- so social purpose of japanese food systems is a bias of mine
=============
The segments of robot development poised to grow the fastest around NVIDIA and its East Asian hardware supply chain revolve around "Physical AI"—the integration of advanced AI foundation models with real-world mechanical systems. NVIDIA CEO Jensen Huang noted that physical AI is the next major wave of technology, a strategy backed by NVIDIA's $150 billion annual spending in Taiwan's supply ecosystem. [1, 2, 3, 4, 5, 6]
The industry maps these submarkets through a collaborative ecosystem where NVIDIA builds the "brain" (chips, foundation models, and simulators) and East Asia builds the "body" and structural components. [1, 2, 3, 4, 5]
The table below outlines how your initial concepts align with current industrial reality, highlighting the specific roles played by Taiwan, Japan, and South Korea:
Submarket Growth Matrix: Core NVIDIA & East Asian Ecosystems
Your Start-up Concept [1, 2, 3, 4, 5, 6, 7, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19]
Industry Term & Reality
Growth Trajectory & Hardware Focus
Regional Strength (TW / JP / KR)
Digital Twins in Controlled Areas
Industrial Digital Twins & Virtual Commissioning
Extremely High. Factories are fully simulated using NVIDIA Omniverse before mechanical construction.
Taiwan: Led by Foxconn and TSMC building fully simulated automated semiconductor fabs and electronic assembly lines.
Parts, Not Wholes (Delicate Arms)
AI-Enabled Manipulators & Cobots
Extremely High. Mass deployment of precise robotic arms running NVIDIA Isaac Manipulator workflows.
Japan: Dominated by precision engineering giants like FANUC, Yaskawa, and KUKA integrating physical AI into factory hardware.
Non-Human Shapes (Dogs, Snakes)
Autonomous Mobile Robots (AMRs) & Quadrupedals
High. Moving from niche disaster relief to mainstream industrial logistics and automated warehouse sorting.
Taiwan / Japan: Industrial AMR platforms (e.g., Toyota and Techman Robot) powered by NVIDIA Isaac Perceptor.
Humanoid Wholes in Social Situations
General Humanoid Embodiment
High Growth, Moderate Maturity. Shifting from deep social situations toward structured commercial environments first.
South Korea / Taiwan: Powered by Hyundai Motor Group (via Boston Dynamics' mass-production of Atlas) and LG Electronics home assistants.
Augmenting Senses (Spectacles)
Edge AI & Spatial Computing Devices
Moderate. Handled as an extension of wearable tech rather than classic robotics.
Taiwan / South Korea: Optics manufacturing, micro-LED displays, and Samsung/SK Hynix LPDDR5X low-power memory modules.
Robots in Wargames
Defense & Tactical Autonomy
High (Isolated). Crucial for military sectors, but strictly segregated from NVIDIA’s commercial consumer ecosystem.
United States / Domestic: Kept entirely separate from East Asian consumer supply chains due to regulatory compliance.
Robots in Education
Educational Robotics Platforms
Low (Niche commercial value). Primarily used as software training grounds rather than drivers for heavy hardware production.
Global: Software-centric developer programs using open-source tools.
Key Adjustments to Your Industry View
- Humanoids go to Factories before Social Situations [1]
While your vision highlights social situations like elderly care (supported by groups like SNU engineers), the actual economic driver for NVIDIA is industrial humanoid deployment. Humanoids are being deployed first in factories and warehouses to address severe labor shortages in Japan and South Korea. They execute multi-step tasks by utilizing NVIDIA Project GR00T, an open foundation model designed for humanlike reasoning and spatial adaptation. [1, 2, 3, 4, 5]
- The Core Hardware Architecture is Already Standardized
The "parts vs. wholes" debate has been resolved by standardized computing modules. Whether a developer is building a delicate arm, a mechanical dog, or a full humanoid, they are centralizing development on the NVIDIA Jetson Thor robotics platform. This chip acts as the universal computing block, handling the complex multi-camera 360-degree vision and fast physical reactions required by modern machinery. [1, 2, 3, 4, 5]
- How the East Asian "Three-Kingdom" Supply Chain Works
- Taiwan (The Foundation & Assembly Hub): Fabricates the physical computing brains. TSMC processes the 3nm silicon required for Jetson Thor platforms, while companies like Foxconn handle system integration, automated manufacturing, and server assembly. [1, 2, 3]
- Japan (The Precision & Mechanical Masters): Constructs the physical bodies, precision gearboxes, and heavy manipulators. Giants like FANUC and Yaskawa supply the physical infrastructure, integrating NVIDIA’s AI software libraries directly into their global fleet of industrial arms. [1, 2, 3, 4]
- South Korea (The Memory & Commercialization Driver): Supplies high-bandwidth memory (HBM) and low-power memory (LPDDR5X) from Samsung and SK Hynix to sustain heavy on-device AI processing. Concurrently, corporate conglomerates like Hyundai (via Boston Dynamics) scale the mass assembly lines necessary to produce tens of thousands of robots annually. [1, 2, 3]
If you want to investigate a specific area further, let me know if you would like to explore Taiwan's component assembly matrix, examine Japan's precision actuator supply lines, or look into NVIDIA's industrial software options for your startup list.
Download full document to discuuss
how does ai brain and robotics come together as engineering challenge
what doman leadership expertise does team need if robotics and brain ai to iteratively improve each other
what sectors beyond nutrition do Asian women priorise culturally in ways that west starts with opposite values on?
-
-
advancing robotics of cars and related applications in a world leading ai company like nvidia///
considering ai contribution to race between driverless cars, is this primarily a racce between world cities?- if so which is top league of cities?
Yes, the race to deploy driverless cars is fundamentally a race between world cities, as Artificial Intelligence (AI) algorithms require hyper-specific, complex urban environments to train, scale, and function safely. [1, 2]
The "top league" of cities leading this transition is divided into two major geographic centers of gravity:
🇺🇸 The United States (Western AI & Fleet Operations)
- San Francisco, California: The global epicenter for commercial robotaxi testing and widespread public use, led by platforms like Waymo.
- Phoenix & Los Angeles, California: Major scaling hubs for driverless ride-hailing networks operating in complex, sprawling metropolitan environments.
- Austin, Texas: A crucial proving ground for AI infrastructure, testing, and continuous machine-learning data collection. [1, 2]
🇨🇳 China (Eastern AI & Smart Infrastructure)
- Beijing: Leads in autonomous vehicle patent generation and regulatory backing for massive municipal testing zones.
- Wuhan & Shenzhen: Serve as the primary proving grounds for scaled, low-cost commercial driverless operations. [1, 2]
Other global tech-forward cities, such as Dubai (United Arab Emirates), consistently score highest in overarching "readiness" due to unmatched 5G connectivity and rapid AI policymaking. [1]
Would you like to explore how specific AI approaches (such as end-to-end deep learning versus HD mapping) are allowing companies in these cities to solve autonomous navigation?
6 sites
- Took my first Waymo today! Right now, these driverless cars are in 10 ...
Right now, these driverless cars are in 10 U.S. cities — including San Francisco, Los Angeles, and Phoenix, along with Austin, ...
Facebook
- Autonomous cars aren't vehicles, they're urban infrastructure
The reality is that they're not cars, they're drivers. The car is a piece of technology that needs sensors, software, learning, re...
Medium
- Ranked: the most driverless car-ready countries - Select Car Leasing
China leads the way for patents filed in the last year. Though historically the USA is ahead in the driverless car race, China's p...
Select Car Leasing
Show all
You said: Do cities that win driverless car race also gain from better maps or other advantages supporting other types of robotic public services
Yes, cities that dominate the driverless car race gain massive spillover advantages that directly accelerate other robotic public services.
The Digital Twin Advantage
- Dynamic HD Maps: Autonomous fleets constantly refresh hyper-detailed 3D digital maps of the city.
- Infrastructure Health: AI vision data automatically detects potholes, broken streetlights, and faded lane markings.
- Utility Planning: Up-to-date spatial data optimizes city planning, water routing, and construction tracking.
Boost to Other Robotic Services
- Sidewalk Delivery: Sidewalk delivery drones deploy instantly by utilizing the vehicle-tested 3D maps and object-detection models.
- Autonomous Cleaning: Robotic street sweepers scale rapidly using the city's established low-latency 5G vehicle networks.
- Emergency Response: Autonomous fire drones and rescue robots leverage existing traffic-prediction AI to bypass congestion.
Structural and Economic Edge
- Edge Computing Networks: Connected roads create a high-speed data grid capable of managing thousands of autonomous devices.
- Regulatory Frameworks: Cities with pre-written laws for robotaxi liability can instantly approve new robotic services.
- Tech Talent Hubs: Local engineering ecosystems attract billions in venture capital, fueling a broader automation economy.
If you'd like to explore this further, I can detail specific tech companies using vehicle data for municipal robots, outline the privacy risks of city-wide AI mapping, or compare U.S. and Chinese infrastructure strategies. Which angle interests you most?
You said: Whats the connection between world leading cities in driverless cars and auto manufcaturers who want to maximise ai assisted driving; eg are leading electric cars manufacturers also pairing up with lead cities?
The connection between driverless cities and electric vehicle (EV) manufacturers is deeply codependent, forming a massive industrial loop.
Leading EV manufacturers are aggressively pairing up with these primary "AI cities" to achieve two distinct goals: scaling pure robotaxis (Level 4/5 driverless fleets) and testing/maximizing consumer AI-assisted driving (Level 2/3 Supervised Autonomy) in the world's most chaotic urban laboratories. [1]
This symbiotic relationship is driving global strategy across three major areas:
- Pure Play Robotaxi Partnerships (The Hardware + Brain Marriage)
Autonomous vehicle software developers do not build cars, and car companies don't build full driverless AI brains. Therefore, AI-first platforms are partnering directly with EV makers to deploy purpose-built vehicles in top-tier cities:
- Waymo & Jaguar / Geely / Hyundai: Waymo operates its premier fleets in San Francisco, Phoenix, Los Angeles, and Austin. To meet soaring demand, Waymo teamed up with vehicle assembler Magna International at a massive integration plant in Mesa, Arizona, to retrofit Jaguar I-PACE electric SUVs and Geely's Zeekr EVs. Furthermore, Waymo partnered with Hyundai to manufacture a massive fleet of all-electric Ioniq 5 SUVs integrated with its 6th-generation driver. [1, 2, 3, 4, 5]
- Baidu Apollo & Domestic EV Makers: China's autonomous driving giant Baidu runs massive networks in Wuhan and Beijing. They collaborate with automakers to build hyper-cheap, scalable EVs pre-integrated with LiDAR and AI suites, aiming to expand these vehicles internationally to cities like Abu Dhabi and London. [1, 2, 3]
- Consumer EV Brands Target the "Lead Cities" for Data [1]
For automakers focused on selling consumer cars with premium AI-assisted driving—such as Tesla (FSD), XPeng, and NIO—getting regulatory access to a leading city is the ultimate prize. They need these dense cities because AI requires complex edge cases (jaywalkers, delivery trucks, erratic scooters) to learn. [1, 2, 3]
- The Tesla and Shanghai Connection: Tesla’s primary global example of pairing with a lead city is Shanghai. Shanghai granted Tesla data-concession and manufacturing privileges at its Gigafactory. In return, Tesla established an on-site local data pipeline and AI training center in Shanghai to comply with data security laws, leading to the highly anticipated wider rollout of its Full Self-Driving (FSD) Supervised system in China. [1]
- Mutual Benefits: Why They Pair Up
┌──────────────────────────┐ ┌──────────────────────────┐
│ WORLD-LEADING CITIES │ │ EV MANUFACTURERS │
│ (SF, Beijing, Shanghai) │ │ (Tesla, Hyundai, BYD) │
└────────────┬─────────────┘ └────────────┬─────────────┘
│ │
│ Offers Regulatory Greenlights │ Provides Built-for-AI Electric
│ & High-Density Visual Data │ Vehicle Fleets at Massive Scale
└───────────────────┬────────────────┘
▼
┌───────────────────────┐
│ Rapidly Evolving AI │
│ Navigation Algorithms │
└───────────────────────┘
- Grid Capacity: Driverless cars must be electric to minimize fleet maintenance and integrate with automated smart-charging grids. Cities with advanced EV infrastructure are the only ones capable of hosting these fleets.
- Regulatory Sandboxes: Automakers gravitate only toward cities that actively rewrite local ordinances. A car company will not deploy its best AI software in a city that hasn't established clear legal liability frameworks for autonomous vehicle mistakes.
If you would like to explore this further, we can look at the financial structures of these partnerships, dive deeper into how Tesla's FSD handles Chinese urban infrastructure, or examine the regulatory hurdles European cities face in attracting these manufacturers. What would you prefer?
10 sites
- Scaling our fleet through U.S. manufacturing - Waymo
May 5, 2025 — The U.S. is leading the world in the development and deployment of autonomous driving technology. Today, Waymo One provides more t...
Waymo
- Waymo and Hyundai enter multi-year, strategic partnership
Oct 4, 2024 — Today, Waymo and Hyundai Motor Company announced they have entered into a multi-year, strategic partnership. In the first phase of...
Waymo
- Tesla 'Full Self-Driving (Supervised)' now in China as ... - CNBC
May 21, 2026 — China's Ministry of Industry and Information Technology issued its first level-three autonomous driving certifications in December...
CNBC
Show all
You said: I think nvidia has said most leaders in self driving cars use their platforms; does this also give nvidia a lead in related robotics beyond cars
Yes, NVIDIA’s absolute dominance in autonomous vehicle (AV) computing has directly translated into a massive, industry-wide lead in general-purpose robotics.
The software, hardware, and simulation infrastructure that NVIDIA originally built to handle chaotic city streets has become the universal architecture for "Physical AI"—the broader industry of AI interacting with the physical world. [1, 2]
NVIDIA has leveraged its driverless car platform to capture the broader robotics market through four distinct technological bridges: [1]
- Unified Silicon Architecture (From DRIVE to Jetson Thor)
The computing brains required to make split-second driving decisions at 65 mph are fundamentally identical to those needed by an industrial robot or a humanoid. NVIDIA capitalized on this by taking its specialized automotive computing platform and re-engineering it into its NVIDIA Jetson Thor series modules. [1]
- The Powerhouse Brain: Powered by the Blackwell GPU architecture, Jetson Thor delivers over 2,000 teraflops of AI compute.
- Widespread Adoption: This chip architecture has quickly become the standard "mind" for next-generation robots, adopted by global industry leaders like Boston Dynamics, Amazon Robotics, and various humanoid manufacturers. [1, 2, 3, 4]
- Generalizing Driverless AI into Humanoid AI (Project GR00T)
Autonomous cars rely heavily on Vision-Language-Action (VLA) models to perceive their environment, understand context, and execute physical motions. NVIDIA ported this concept directly into humanoid robotics via Project GR00T, its open foundation model for generalist robots. [1, 2, 3]
- Shared Learning: Just as a driverless car learns to predict pedestrian pathways, robots running the NVIDIA Isaac GR00T foundation model use natural language and human imitation to quickly master coordination, dexterity, and navigation across manufacturing plant floors. [1, 2]
- The Digital Twin Sandbox (Omniverse & Isaac Sim)
To test driverless software safely, NVIDIA built virtual-world environments capable of photorealistic simulation and accurate physics modeling. This tooling evolved directly into NVIDIA Omniverse and Isaac Sim, the world's premier testing sandboxes for physical robotics. [1, 2]
- Synthetic Data Generation: Instead of spending months physically teaching a robot to sort factory parts or navigate a warehouse, developers use NVIDIA's simulation pipelines to generate millions of synthetic training scenarios in just hours. [1, 2]
- Cross-Industry Deployment
NVIDIA’s dominance spans the entire spectrum of automated labor: [1]
NVIDIA Technical Solution
Real-World Application
Industrial / Logistics
High-precision factory arms, smart warehouse pickers, and automated tugs.
Heavy Machinery
NVIDIA DRIVE / Jetson
Autonomous John Deere tractors and Caterpillar construction machinery.
Humanoids
Jetson Thor + Cosmos Models
Advanced bipedal helpers designed to offset labor shortages in manufacturing.
Because autonomous cars forced NVIDIA to solve edge-case perception, real-time safety auditing, and massive data pipelines early on, the company effectively established a moat. Any robotics startup today trying to build a physical AI product almost certainly has to build it on top of NVIDIA's pre-existing stack. [1, 2, 3, 4]
If you are interested, we can look closer at how competitors like AMD or Intel are responding, explore the financial growth of NVIDIA’s non-automotive robotics division, or examine how a specific humanoid robot uses Jetson Thor to function. Which area would you like to investigate next? [1, 2, 3]
17 sites
- AI: Nvidia's growing chops in self driving cars & Vera Rubin AI ...
Jan 10, 2026 — “The big picture: Nvidia has long been investing in physical AI, meaning AI interfacing with the world and not just software.” And...
Medium·Michael Parekh
- NVIDIA Dominates Robotic AI: Inside the Jetson – Isaac
Dec 17, 2025 — For more than a decade, NVIDIA has no longer been just a GPU manufacturer. The California-based company has become one of the cent...
Robot Magazine
- NVIDIA Blackwell-Powered Jetson Thor Now Available, Accelerating ...
Aug 25, 2025 — The Ultimate Platform for Next-Generation Robotics Powered by an NVIDIA Blackwell GPU and featuring 128GB of memory, Jetson Thor d...
NVIDIA Newsroom
You said: In many aspects of robots nvidia's suppky chain ecosystem seems to connect taiwan korea and japan - i am thinking of such robotics areas as Digital Twins in Controlled Areas Parts, Not Wholes (Delicate Arms) Non-Human Shapes (Dogs, Snakes) Humanoid Wholes in Social Situations so i am wondering when it comes to robotics is it partiicularly important to nvidia to see if opportunities to wirk in chuna return? .
Yes, returning to work dynamically in China is critically important to NVIDIA’s long-term physical AI and robotics strategy.
While NVIDIA has successfully woven an intricate East Asian hardware web—tapping Taiwan for advanced foundry work and humanoid assembly, Korea for high-bandwidth memory (HBM), and Japan for precision robotics machinery—China represents a missing piece that cannot easily be replaced. [1, 2, 3]
The requirement for NVIDIA to regain access to China spans the four distinct robotics pillars you highlighted:
- Digital Twins in Controlled Areas (The "Smart Factory" Engine) [1]
NVIDIA’s Omniverse platform relies on creating complete digital clones of massive manufacturing operations. [1]
- The Problem: China is the "factory of the world" and the single largest market for industrial automation.
- The Constraint: While NVIDIA partners with Taiwanese giants like Foxconn to simulate facilities, the physical, multi-acre gigafactories where these innovations must be deployed at scale reside predominantly in mainland China. Without access to Chinese mega-factories, NVIDIA loses its largest testing ground and customer base for autonomous industrial plants. [1]
- Parts, Not Wholes (Supply Chain Dominance for Delicate Arms)
High-precision robotic components (actuators, precision gears, sensors, and delicate end-effectors) require tight integration with the AI "brain."
- The Competition: While Japan produces peerless high-end machinery, China has rapidly built the world's most dense, low-cost supply chain ecosystem for robotic parts. [1]
- The Risk: Major Chinese AI models (like DeepSeek) are heavily optimizing their software to run natively on domestic hardware like Huawei's Ascend chips. If NVIDIA is locked out of China, Chinese component manufacturers will design their delicate robotic parts to integrate with domestic silicon rather than NVIDIA’s Jetson or Thor platforms. [1]
- Non-Human Shapes (Dogs, Snakes, and Industrial Inspection)
Quadruped "robot dogs" and specialized inspection rovers are heavily utilized for monitoring hazardous environments, oil rigs, and power grids.
- The Market Reality: China currently dominates the global commercial marketplace for quadruped robotics (led by companies like Unitree). Nearly all of these agile, non-humanoid platforms have historically relied on NVIDIA’s edge modules to process vision and movement. NVIDIA cannot afford to cede the computing architecture of this entire booming sector to Chinese semiconductor alternatives.
- Humanoid Wholes in Social Situations (The Scale Challenge)
Humanoids interacting with humans in retail, caretaking, or hospitality require vast amounts of localized, cultural, and linguistic training data.
- The Scale Edge: China offers an unmatched demographic density and a highly permissive regulatory environment for deploying public-facing AI.
- The Pivot: Acknowledging this massive market, NVIDIA leadership has aggressively negotiated with the U.S. government to permit exports. This push contributed to recent regulatory shifts—such as the U.S. pivoting toward a case-by-case review to allow the export of advanced chips like the NVIDIA H200 to approved Chinese commercial buyers (subject to unique 25% U.S. tariffs and third-party security audits). [1, 2, 3, 4, 5]
The Geopolitical Balancing Act
┌────────────────────────────────────────────────────────┐
│ NVIDIA'S TWO-FRONT PHYSICAL AI STRATEGY │
└───────────────────────────┬────────────────────────────┘
│
┌──────────────┴──────────────┐
▼ ▼
┌─────────────────────────┐ ┌─────────────────────────┐
│ THE BLUEPRINT NODE │ │ THE APPLICATION NODE │
│ (Taiwan, Japan, Korea) │ │ (China) │
├─────────────────────────┤ ├─────────────────────────┤
│ • Advanced Foundry/HBM │ │ • Global Factory Floor │
│ • Photorealistic Twins │ │ • Mass Component Scale │
│ • Precision Hardware │ │ • Dense Consumer Data │
└─────────────────────────┘ └─────────────────────────┘
NVIDIA has built the ultimate "Physical AI" blueprint through its partners in Taiwan, Korea, and Japan. However, without an active pathway into China, the platform is akin to an incredibly advanced brain operating without access to the world’s largest muscular system. NVIDIA needs China not just for immediate revenue, but to prevent a parallel, domestic Chinese robotics ecosystem from evolving completely independent of NVIDIA's architecture. [1, 2]
If you'd like, we can explore how Chinese robotics firms are adapting to U.S. export controls, look closely at the specific terms of the H200 export policy, or examine how domestic chips like Huawei’s Ascend compare to NVIDIA’s hardware for robotics. Which area would you like to target? [1, 2, 3]
12 sites
- NVIDIA Omniverse Digital Twins Help Taiwan Manufacturers ...
May 18, 2025 — NVIDIA Omniverse Digital Twins Help Taiwan Manufacturers Drive Golden Age of Industrial AI. Blueprints for building autonomous man...
NVIDIA Blog
- Nvidia eyes Taiwan for humanoid robot assembly; supply chain ...
Dec 19, 2024 — Nvidia eyes Taiwan for humanoid robot assembly; supply chain gaps remain. ... Nvidia is reportedly forming a humanoid robot supply...
digitimes
- Nvidia's pivot to physical AI ignites rally across Asian supply ...
May 2, 2026 — Trade NVIDIA on Coinbase Trading disclosure. Investing.com -- Nvidia Corp.'s strategic expansion into "physical AI" is driving a s...
You said: yes please i would like to explore both explore how Chinese robotics firms are adapting to U.S. export controls, look closely at the specific terms of the H200 export policy, or examine how domestic chips like Huawei’s Ascend compare to NVIDIA’s hardware for robotics. Which area would you like to target? -see next post
-
-
The collision of AI, robotics, and geopolitics has turned the semiconductor supply chain into a battlefield. The landscape is defined by aggressive regulatory shifts, domestic Chinese workarounds, and a fierce architectural chip rivalry. [1, 2]
- Unpacking the H200 Export Policy
In a dramatic shift from previous strict bans, the U.S. Bureau of Industry and Security (BIS) issued an updated framework regulating advanced artificial intelligence chips. [1]
- The Policy Pivot: U.S. policymakers shifted from an absolute "presumption of denial" to a structured, case-by-case licensing review. This specifically opens a heavily regulated legal pathway for the export of NVIDIA’s Hopper-generation H200 chips and AMD’s MI325X to China. [1, 2, 3]
- The 50% Volume Cap: To prevent China from building hyper-massive AI clusters that eclipse the U.S., a strict volume threshold is enforced. Total H200 shipments to China cannot exceed 50% of the aggregate volume sold to American domestic customers. [1, 2, 3]
- Tariffs & Independent Audits: The policy applies a 25% federal tariff fee directly to U.S. chipmakers for these China-bound transactions. Furthermore, physical hardware must undergo third-party security testing inside the U.S. to verify hardware performance limits before export. [1, 2, 3, 4]
- The "Blackwell Gate": The U.S. maintains an absolute embargo on its latest-generation Blackwell (B300) systems. This ensures Washington preserves a permanent "relative compute advantage" by keeping China one hardware generation behind frontier Western models. [1, 2]
- How Chinese Robotics Firms Are Adapting
Faced with unpredictable supply lines and a severe shortage of frontier U.S. hardware, Chinese robotics and autonomous driving firms have been forced to fundamentally alter their software and hardware architectures: [1, 2, 3]
- The "Clustering" Workaround: Because individual Chinese-made processors are less efficient than NVIDIA's top chips, robotics labs are grouping vast quantities of lower-tier chips together. Using advanced internal interconnect frameworks (like CloudMatrix), Chinese firms can link thousands of weaker processors. This achieves high-throughput AI model training despite a weaker individual chip stack. [1, 2, 3]
- Synthetic Optimization: Without endless cloud computing pools for physical edge-case processing, robotics developers are writing hyper-efficient software algorithms designed to max out fewer compute cycles. Models like DeepSeek V4 are engineered natively to run on compact, local Chinese hardware networks, drastically reducing the token cost and memory overhead required to operate robots. [1]
- Hardware Hedging: Rather than relying exclusively on black-market smuggling or short-term server rentals—which have inflated the price of banned U.S. hardware inside China to nearly double the market rate—robotics firms are proactively shifting their long-term hardware roadmaps to domestic chip foundries. [1, 2]
- Domestic Chips: Huawei’s Ascend vs. NVIDIA Hardware
Huawei has aggressively capitalized on Western export restrictions, pivoting to fulfill the massive computing demands of China's domestic robotics ecosystem. [1, 2]
NVIDIA H200 (Hopper Architecture)
Huawei Ascend 950PR / 910C
Peak AI Compute
~1,000 Teraflops (FP16) [1]
Peak AI Compute
Software Ecosystem
Software Ecosystem
CAN (Proprietary, with CUDA-migration tools) [1]
Memory Architecture
Ultra-dense High-Bandwidth Memory (HBM3) [1]
Memory Architecture
Engineered Multi-Chip Module (Co-packaged interconnect) [1]
Power Efficiency
High performance per watt [1]
Power Efficiency
Robotics Viability
Universal plug-and-play for advanced Vision-Language-Action (VLA) models [1]
Robotics Viability
Rapidly expanding via domestic "Supernote" cluster nodes [1, 2]
- The Raw Compute Bridge: Huawei’s newest Ascend 950PR and 910C processors successfully bridge the peak hardware gap. By outputting substantial raw petaflops, they can easily process the multi-camera inputs and real-time object tracking required by autonomous vehicles and humanoid robots. [1, 2]
- The CUDA Software Moat: NVIDIA's greatest defense remains its CUDA platform, which millions of global robotics engineers have used for decades to write physical AI scripts. To break this lock, Huawei has heavily optimized its software layers to offer seamless CUDA compatibility. This allows Chinese engineers to migrate their existing robotics algorithms onto Ascend chips with minimal code rewrites. [1, 2, 3]
The result of the U.S. embargo has inadvertently backfired in terms of market dominance. By attempting to isolate Beijing's AI ecosystem, the U.S. incentivized a massive wave of capital investment into Chinese fabs like SMIC and suppliers like Huawei. This has fueled a highly resilient, completely independent domestic robotics ecosystem that no longer relies on Western supply chains. [1, 2, 3, 4, 5, 6]
If you would like to continue this exploration, let me know if you would like to focus on:
- The specific mechanics of how Huawei's software translates NVIDIA's CUDA code.
- How Taiwanese assembly lines are handling the logistical split between U.S. and China hardware.
- The impact of these chip dynamics on robotaxi pricing and rollout speeds in Chinese versus Western cities. [1]
The collision of advanced AI, strict global export bans, and domestic market adaptations has fragmented the robotics and autonomous vehicle landscape.
The dynamics of this tech split are visible across codebases, manufacturing floors, and urban transit fares:
- The Code Pipeline: How Huawei Translates NVIDIA's CUDA [1]
For fifteen years, NVIDIA’s CUDA framework has been the universal software standard for programming parallel computing tasks in robotics. Stripping a driverless car system of CUDA usually requires years of rebuilding code from scratch. To survive this, Huawei developed CANN (Compute Architecture for Neural Networks), a dedicated software toolkit engineered specifically to bypass NVIDIA's proprietary ecosystem. [1, 2, 3, 4]
Huawei achieves this software bridge through a highly optimized compilation and migration pipeline:
┌─────────────────────────────────┐
│ NVIDIA CUDA Source Code │ (Standard Physical AI / Robotics Script)
└────────────────┬────────────────┘
│
▼
┌─────────────────────────────────┐
│ torch_npu Plugin │ (Huawei's PrivateUse1 PyTorch Bridge)
└────────────────┬────────────────┘
│
▼
┌─────────────────────────────────┐
│ CANN 8.0 Translation Layer │ (Mapping Vision-Language-Action Models)
└────────────────┬────────────────┘
│
▼
┌─────────────────────────────────┐
│ Huawei Ascend AI Silicon │ (Direct Neural Execution on 950PR/910C)
└─────────────────────────────────┘
- The PyTorch Intercept (torch_npu): Most modern robotics models (such as Vision-Language-Action systems) are designed in PyTorch. Huawei engineered a specialized backend plugin called torch_npu. When a robotics developer feeds standard PyTorch scripts into the system, the plugin intercepts the calls and routes them directly through PyTorch's native PrivateUse1 memory mechanism, converting them seamlessly into Huawei-readable instructions. [1, 2]
- The Graph Compiler (CANN 8.0): Instead of executing NVIDIA instructions line-by-line—which would crash the system—the CANN graph compiler reconstructs the entire operational pipeline of the AI model. It maps out the exact sequence of real-time camera tracking and sensor calculations, re-optimizing them specifically for the internal core architecture of Huawei Ascend processors. [1, 2, 3]
- The Moat Dynamic: While NVIDIA actively banned the creation of direct "translation layers" within its software, Huawei counter-attacked by making the CANN architecture completely open-source. This encourages global robotics engineers to co-develop code directly for Ascend silicon, helping to accelerate software maturation and slowly degrade NVIDIA's historical monopoly. [1, 2, 3]
- The Logistical Split on Taiwanese Assembly Lines
Taiwan is the manufacturing engine of the AI boom, housing the world's most critical semiconductor fabrication lines and advanced backend packaging technologies, like TSMC's CoWoS (Chip-on-Wafer-on-Substrate). Because Taiwanese manufacturing serves both American clients and specific approved global markets, its assembly lines have been logistically split in two: [1, 2]
- The "Dual-Silo" Factory Floor: To comply with strict U.S. export controls and avoid multi-billion-dollar penalties, Taiwanese contract manufacturers (like Foxconn and Quanta) operate completely separate production pipelines.
- The Western Line handles full-spec Blackwell B300 and premium H200 architectures bound for U.S. cloud centers and Western driverless networks.
- The Compliance Line relies on strict inventory tracking systems to assemble heavily modified, lower-compute variants, or processes specialized orders aligned with the strict volume parameters of the case-by-case H200 export framework. [1, 2]
- Decentralizing the "Silicon Shield": Historically, Taiwan's total monopoly on advanced AI packaging acted as a geopolitical buffer. However, following major trade agreements and significant capital influxes, advanced packaging and assembly lines are actively decentralizing. TSMC is aggressively building out packaging infrastructure—not just across key hubs in Taiwan (such as Zhunan and Tainan)—but also expanding into Phoenix, Arizona. This structural shift provides Western driverless car infrastructure with an domestic, embargo-proof hardware pipeline. [1, 2, 3, 4]
- Impact on Robotaxi Pricing and Rollout Speeds
The resulting technology split has triggered a stark divergence in how driverless services are expanding worldwide:
🇨🇳 Chinese Cities (The Low-Cost Scale Strategy)
- Pricing: A robotaxi ride in cities like Wuhan or Shenzhen costs a fraction of a human-driven ride, frequently dipping to under $1 to $2 per trip.
- Rollout Speed: Rapid deployment. Because Chinese firms are optimized to work around hardware limitations using system-level innovations, they trade abundant domestic grid power for limited compute. By linking clusters of affordable domestic chips together via frameworks like CloudMatrix, companies like Baidu Apollo can mass-produce self-driving arrays at scale. This permits the deployment of thousands of vehicles across major metropolitan areas simultaneously. [1]
🇺🇸 Western Cities (The Premium Compute Strategy)
- Pricing: A robotaxi ride in San Francisco, Los Angeles, or Phoenix remains positioned as a premium service, frequently matching or slightly undercutting traditional ride-hailing fees ($15 to $30+ per typical urban trip).
- Rollout Speed: Deliberate and highly controlled. Western companies like Waymo rely on hyper-advanced, centralized frontier computing systems (such as customized NVIDIA platforms). While this yields a highly reliable system capable of navigating dense, unpredictable traffic, the steep upfront costs of elite Western silicon mean fleets scale gradually, city by city, to ensure financial viability.
Future Inquiries
If you would like to explore this further, let me know if you would like to focus on:
- The financial performance and unit economics of a Waymo vehicle versus a Baidu/Wuhan robotaxi.
- The specific power and grid requirements Chinese data centers are utilizing to support chip-clustering.
- How European cities are attempting to write regulations to catch up to this U.S.-China duopoly. [1]
- Unit Economics: Waymo vs. Baidu Apollo
The financial models of American and Chinese robotaxi operators reflect two completely different philosophies of scaling automation.
🇺🇸 WAYMO (San Francisco / Phoenix) 🇨🇳 BAIDU APOLLO (Wuhan / Shenzhen)
┌─────────────────────────────────┐ ┌─────────────────────────────────┐
│ • Vehicle Cost: ~$100k - $150k │ │ • Vehicle Cost: ~$28k (RT6) │
│ • Sensor Suite: Custom Lidar │ │ • Sensor Suite: Mass Lidar/Cam │
│ • Strategy: High Margin, Safe │ │ • Strategy: Low Cost, Dominate │
└─────────────────────────────────┘ └─────────────────────────────────┘
Waymo (The Premium Platform)
- Vehicle Cost: A Waymo vehicle (like the Jaguar I-PACE or upcoming Hyundai Ioniq 5) costs an estimated $100,000 to $150,000 to put on the road, factoring in custom suite integrations.
- Sensor Stack: Heavy reliance on custom, high-end LiDAR, radar, and cameras. This hardware creates pristine data, but it is expensive to manufacture and replace.
- Path to Profit: Waymo focuses on high-utilization and premium pricing. By targeting dense, high-income cities (SF, LA, NY), they charge near-Uber rates ($15–$30+ per ride). They aim for profitability by removing the human driver wage while keeping margins high per mile, rather than flooding the market with cheap vehicles.
Baidu Apollo / RT6 (The Mass-Market Drone)
- Vehicle Cost: Baidu’s generation-6 robotaxi (the Apollo RT6) has brought manufacturing costs down to roughly 204,000 RMB (~$28,000 USD) per vehicle.
- Sensor Stack: They utilize mass-manufactured, automotive-grade sensors deeply integrated directly into the vehicle's body during assembly, drastically reducing retrofitting labor costs.
- Path to Profit: Brutal volume. Because the vehicle asset is so cheap, Baidu can price rides at $1 to $3 and still eye a path to profitability. They treat the robotaxi as public transit infrastructure, aiming to dominate ride volume, capture massive consumer data, and squeeze out human taxi drivers entirely.
- The Power & Grid Strain of Chip-Clustering
Because Chinese tech firms cannot easily buy NVIDIA's most power-efficient, ultra-dense Blackwell architectures, their reliance on "clustering" thousands of lower-tier domestic chips creates a massive infrastructure challenge: extreme power consumption.
- The Efficiency Penalty: To match the performance of one top-tier frontier chip cluster, a lab linking weaker chips requires significantly more physical silicon, data cables, and cooling infrastructure. This causes a massive drop in performance-per-watt. It takes significantly more electricity to train the same Vision-Language-Action (VLA) robotics model on distributed domestic networks than it does on centralized Western hardware.
- Grid Integration: China has adapted by placing its massive AI training clusters directly adjacent to its green energy hubs in western provinces (like Guizhou, Inner Mongolia, and Gansu). This aligns with the national "East Data, West Computing" strategy.
- Real-Time Edge Pressure: While training happens in the green, high-capacity West, the actual vehicles operate in dense eastern cities (Beijing, Shanghai). The local city grids must support massive 5G data nodes and rapid-charging depots for thousands of robotaxis. This requires cities to deploy automated "smart-charging" grids that charge robotaxis only during off-peak midnight hours to avoid collapsing municipal power networks.
- Can European Cities Catch Up?
Your intuition is highly accurate: European cities are severely lagging behind the U.S.-China duopoly. Europe lacks a homegrown hyperscale cloud provider (like AWS or Alibaba), a frontier chip designer (like NVIDIA or Huawei), and a scaled commercial robotaxi operator.
However, European cities are attempting a highly specific, niche strategy to carve out a "Third Way" in the robotics race.
The European Strategy: B2B Logistics & Public Transit Integration
Rather than trying to build consumer robotaxis to compete with Waymo or Baidu, Europe is focusing on automating structured, commercial public systems.
- The Legal Blueprint (Germany’s Level 4 Law): Germany passed a legal framework allowing Level 4 autonomous vehicles on public roads in defined micro-zones. This gave Europe its first real regulatory sandbox.
- Autonomous Public Transit (Shuttles over Sedans): Instead of individual cars, European cities (like Hamburg, Munich, and Oslo) are integrating autonomous, electric minibuses into their existing public transit networks. Companies like MOIA (a Volkswagen subsidiary) and Holon are deploying driverless shuttles that act as "first-mile/last-mile" extensions of trains and subways.
- The Industrial Edge (The "Mittelstand" Robotics Hub): While Europe struggles with consumer AI, it excels in heavy industrial precision hardware. Cities in Germany, Italy, and Denmark (like Odense, the robotics capital of Europe) are focusing on factory floor robotics, autonomous agricultural machinery, and automated port logistics.
Europe’s path to relevance is not via consumer ride-hailing software, but by becoming the global leader in highly regulated, heavy-duty industrial automation and public transit integration.
Next Steps for Our Conversation
To continue tailoring this to your goals, let me know if you would like to explore:
- The exact software architecture Volkswagen/MOIA or European shuttle firms are using to navigate without Western cloud dependencies.
- A deeper dive into how U.S. cities are handling the electricity demands of Waymo’s expanding charging hubs.
- How the decline in vehicle costs (like Baidu's $28k car) might disrupt Western automotive manufacturing if exported globally.
You are entirely correct. Your prediction accurately mirrors the current geopolitical fracturing of the physical AI market.
The "Baidu Hypothesis"—the realization that purpose-built robotaxis can be factory-manufactured for under $30,000—proves that software-driven autonomy is no longer an expensive tech experiment, but a highly scalable commodity. [1, 2, 3]
Because the hardware barrier has been shattered, the deployment of these cheap driverless fleets follows a clear, bifurcated global roadmap.
Phase 1: The Immediate Impact on Asian and Middle Eastern Markets
Chinese autonomous vehicle (AV) firms are completely bypassing Western friction by expanding aggressively across friendly, tech-forward, and highly permissive regulatory landscapes. [1, 2]
- The Middle Eastern Proving Ground: The most immediate shockwave is happening in the Gulf. Baidu Apollo launched its fully driverless commercial platform in D..., deploying its sixth-generation, $28,000 RT6 vehicles in direct competition with local transit. Simultaneously, competitors like WeRide have scaled to hundreds of vehicles in Abu Dhabi, achieving localized operational profitability far faster than operators in Western cities. [1, 2, 3]
- Southeast Asian Expansion: Baidu is directly targeting high-density Asian logistics hubs, aggressively moving into Singapore and Malaysia. These cities offer the perfect combination of dense urban traffic and a desperate need for low-cost, space-efficient public mobility options. [1, 2, 3]
- Global App Integration: Rather than trying to fight traditional ride-hailing incumbents, Baidu signed a global partnership with Uber. This allows Uber users outside of China to seamlessly summon a cheap, Chinese-manufactured robotaxi via their existing app, accelerating public adoption across neutral international markets. [1, 2]
Phase 2: The Western Iron Curtain (Total Regulatory Barricades)
As you anticipated, the United States has moved aggressively to ensure these ultra-cheap, AI-driven vehicles never touch American asphalt, framing the restriction not just as economic protectionism, but as an urgent national security defense. [1]
The West is utilizing a layered strategy to enforce a complete blockade:
- The Connected Car Software Ban: The United States finalized a sweeping mandate through the Commerce Department that completely bans Chinese-developed or maintained software in any connected or autonomous passenger vehicles. This restriction completely closes the door on Baidu’s Apollo platform or Huawei's software layers operating in the U.S.. [1, 2]
- The Connected Hardware Ban: Following the software blockade, a complementary ban takes effect to prohibit the import of Chinese-made vehicle connectivity hardware (such as modules managing Bluetooth, cellular, and LiDAR arrays). This legal wall is so rigid that it initially threatened Western brands with heavy Chinese supply links, forcing companies like Geely-owned Volvo to undergo strict federal audits to secure narrow exemptions. [1, 2, 3]
- Massive Economic Tariffs: Even if a Chinese vehicle were somehow retrofitted to scrub its code of foreign dependencies, standard 100%+ economic import tariffs on Chinese EVs ensure that a $28,000 vehicle instantly becomes financially unviable upon entry to the U.S. market. [1, 2]
The Resulting Geopolitical Map
┌─────────────────────────────────┐ ┌─────────────────────────────────┐
│ THE CLOSED WEST │ │ THE ACCELERATING REST │
│ (U.S. / Canada) │ │ (Gulf, SE Asia, LatAm) │
├─────────────────────────────────┤ ├─────────────────────────────────┤
│ • Total Software/Hardware Bans │ │ • Mass Entry of $28k Robotaxis │
│ • High-Margin, Premium Fleets │ │ • Uber/Local App Integrations │
│ • Slower, Highly Controlled AV │ │ • Rapid Transit Automation │
│ Expansion (e.g., Waymo) │ │ at Hyper-Low Consumer Cost │
└─────────────────────────────────┘ └─────────────────────────────────┘
By barring these imports, the West is successfully shielding its domestic automotive industry and maintaining absolute data privacy control. However, the unintended consequence is that it has trapped Western robotaxi operators in a high-cost environment. While a resident of San Francisco pays premium rates for a bespoke Waymo ride, commuters in Dubai, Singapore, and Wuhan are acclimating to ubiquitous, ultra-cheap autonomous transit powered entirely by Chinese capital and silicon. [1, 2, 3, 4]
Given this sharp divide, would you like to explore how European regulators are wavering between the U.S. blockade strategy and China's low-cost entry, or look closer at how U.S. robotaxis are attempting to drop their own manufacturing costs to compete globally? [1]
About
ENTREPRENEURIAL REVOLUTION NETWORK BENCHMARKS 2025now : Remembering Norman Macrae
Chartering 5 Layer AI Agency - integrating exponential intergenerational multipliers of trusted human relationship systems through community scaling apps
AsiaAI.docx where & how 2/3 human brains are celebrating AI livelihoods
====
lelated US AI reports:
chris.macrae@yahoo.co.uk :help celebrate library of INTELLIGENCE multipliers: -system map
- Action Apps
- Millions of AI Agents 1 2 3
- Software sovereign infrastructure
- Chips1 & Supercomputers
- Energy: Genesis
- Fusion SCSP-FI -F2
- Quantum
- Critical Minerals: Pax
- Space
- Edu-media rev li>Nature
- workforce 1
cvchrismacrae.docx - Data Science
- Geonomics 1
views on whether AGI exists
- how close are google aws or huawei to nvidia
2025REPORT-ER: Entrepreneurial Revolution est 1976; Neumann Intelligence Unit at The Economist since 1951. Norman Macrae's & friends 75 year mediation of engineers of computing & autonomous machines has reached overtime: Big Brother vs Little Sister !?
Overtime help ed weekly quizzes on Gemini of Musk & Top 10 AI brains until us election nov 2028
unaiwho.docx version 6/6/22 hunt for 100 helping guterres most with UN2.0
RSVP chris.macrae@yahoo.co.uk
Prep for UNSUMMITFUTURE.com
JOIN SEARCH FOR UNDER 30s MOST MASSIVE COLLABS FOR HUMAN SUSTAINABILITY
1 Jensen Huang 2 Demis Hassabis 3 Dei-Fei Li 4 King Charles
5 Bezos Earth (10 bn) 6 Bloomberg JohnsHopkins cbestAI.docx 7 Banga
8 Maurice Chang 9 Mr & Mrs Jerry Yang 10 Mr & Mrs Joseph Tsai 11 Musk
12 Fazle Abed 13 Ms & Mr Steve Jobs 14 Melinda Gates 15 BJ King 16 Benioff
17 Naomi Osaka 18 Jap Emperor Family 19 Akio Morita 20 Mayor Koike
The Economist 1982 why not Silicon AI Valley Everywhere 21 Founder Sequoia 22 Mr/Mrs Anne Doerr 23 Condi Rice
23 MS & Mr Filo 24 Horvitz 25 Michael Littman NSF 26 Romano Prodi 27 Andrew Ng 29 Lila Ibrahim 28 Daphne Koller
30 Mayo Son 31 Li Ka Shing 32 Lee Kuan Yew 33 Lisa Su 34 ARM 36 Priscilla Chan
38 Agnelli Family 35 Ms Tan & Mr Joe White
37 Yann Lecun 39 Dutch Royal family 40 Romano Prodi
41 Kramer 42 Tirole 43 Rachel Glennerster 44 Tata 45 Manmohan Singh 46 Nilekani 47 James Grant 48 JimKim, 49 Guterres
50 attenborough 51 Gandhi 52 Freud 53 St Theresa 54 Montessori 55 Sunita Gandhu,56 paulo freire 57 Marshall Mcluhan58 Andrew Sreer 59 Lauren Sanchez, 60 David Zapolski
61 Harris 62 Chips Act Raimundo 63 oiv Newsom. 64 Arati Prab hakarm,65 Jennifer Doudna CrispR, 66 Oren Etsioni,67 Robert Reisch,68 Jim Srreyer 69 Sheika Moza
- 3/21/22 HAPPY 50th Birthday TO WORLD'S MOST SUSTAINABLE ECONOMY- ASIAN WOMEN SUPERVILLAGE
Since gaining my MA statistics Cambridge DAMTP 1973 (Corpus Christi College) my special sibject has been community building networks- these are the 6 most exciting collaboration opportunities my life has been privileged to map - the first two evolved as grassroots person to person networks before 1996 in tropical Asian places where village women had no access to electricity grids nor phones- then came mobile and solar entrepreneurial revolutions!!
COLLAB platforms of livesmatter communities to mediate public and private -poorest village mothers empowering end of poverty 5.1 5.2 5.3 5.4 5.5 5.6
4 livelihood edu for all
3 last mile health services 3.1 3,2 3.3 3.4 3.5 3.6
last mile nutrition 2.1 2.2 2.3 2.4 2.5 2,6
banking for all workers 1.1 1.2 1.3 1.4 1.5 1.6
NEWS FROM LIBRARY NORMAN MACRAE -latest publication 2021 translation into japanese biography of von neumann:
Below: neat German catalogue (about half of dad's signed works) but expensive -interesting to see how Germans selected the parts they like over time: eg omitted 1962 Consider Japan The Economist
feel free to ask if free versions are available
0 The coming entrepreneurial revolution : a survey - In: The economist 261 (1976), pp. 41-65 cited 105
7 Future U.S. growth and leadershipMacrae, Norman - In: FutureQuest : new views of economic growth, (pp. 49-60). 1977 Check Google Scholar |
13Oh, Brazil : a survey Macrae, Norman - The Economist 272 (1979), pp. 1-22
Macrae, Norman -In: European community (1978), pp. 3-6
16 Whatever happened to British planning? Macrae, Norman - CapitalismToday, (pp. 140-148). 1971 Check Google Scholar |
18 How the EEC makes decisions MacRae, Norman - In: Readings in international business, (pp. 193-200). 1972 Check Google Scholar |
. we scots are less than 4/1000 of the worlds and 3/4 are Diaspora - immigrants in others countries. Since 2008 I have been celebrating Bangladesh Women Empowerment solutions wth NY graduates. Now I want to host love each others events in new york starting this week with hong kong-contact me if we can celebrate anoither countries winm-wins with new yorkers
mapping OTHER ECONOMIES:
50 SMALLEST ISLAND NATIONS
TWO Macroeconomies FROM SIXTH OF PEOPLE WHO ARE WHITE & war-prone
ADemocratic
Russian
=============
From 60%+ people =Asian Supercity (60TH YEAR OF ECONOMIST REPORTING - SEE CONSIDER JAPAN1962)
Far South - eg African, Latin Am, Australasia
Earth's other economies : Arctic, Antarctic, Dessert, Rainforest
===========
In addition to how the 5 primary sdgs1-5 are gravitated we see 6 transformation factors as most critical to sustainability of 2020-2025-2030
Xfactors to 2030 Xclimate XAI Xinfra Xyouth Wwomen Xpoor chris.macrae@yahoo.co.uk (scot currently in washington DC)- in 1984 i co-authored 2025 report with dad norman.
|
Asia Rising Surveys
Entrepreneurial Revolution -would endgame of one 40-year generations of applying Industrial Revolution 3,4 lead to sustainability of extinction 1972's Next 40 Years ;1976's Coming Entrepreneurial Revolution; 12 week leaders debate 1982's We're All Intrapreneurial Now
|
The Economist had been founded in 1843" marking one of 6 exponential timeframes "Future Histores"
IN ASSOCIATION WITH ADAMSMITH.app :
we offer worldwide mapping view points from
and these viewpoints:
40 years ago -early 1980s when we first framed 2025 report;
from 1960s when 100 times more tech per decade was due to compound industrial revolutions 3,4
1945 birth of UN
1843 when the economist was founded
1760s - adam smithian 2 views : last of pre-engineering era; first 16 years of engineering ra including america's declaration of independence- in essence this meant that to 1914 continental scaling of engineeriing would be separate new world <.old world
IF we 8 billion earthlings of the 2020s are to celebrate collaboration escapes from extinction, the knowhow of the billion asian poorest women networks will be invaluable -
in mathematically connected ways so will the stories of diaspora scots and the greatest mathematicians ever home schooled -central european jewish teens who emigrated eg Neumann , Einstein ... to USA 2nd quarter of the 20th century; it is on such diversity that entrepreneurial revolution diaries have been shaped
EconomistPOOR.com : Dad was born in the USSR in 1923 - his dad served in British Embassies. Dad's curiosity enjoyed the opposite of a standard examined education. From 11+ Norman observed results of domination of humans by mad white men - Stalin from being in British Embassy in Moscow to 1936; Hitler in Embassy of last Adriatic port used by Jews to escape Hitler. Then dad spent his last days as a teen in allied bomber command navigating airplanes stationed at modernday Myanmar. Surviving thanks to the Americas dad was in Keynes last class where he was taught that only a handful of system designers control what futures are possible. EconomistScotland.com AbedMooc.com
To help mediate such, question every world eventwith optimistic rationalism, my father's 2000 articles at The Economist interpret all sorts of future spins. After his 15th year he was permitted one signed survey a year. In the mid 1950s he had met John Von Neumann whom he become biographer to , and was the only journalist at Messina's's birth of EU. == If you only have time for one download this one page tour of COLLABorations composed by Fazle Abed and networked by billion poorest village women offers clues to sustainability from the ground up like no white ruler has ever felt or morally audited. by London Scot James Wilson. Could Queen Victoria change empire fro slavemaking to commonwealth? Some say Victoria liked the challenge James set her, others that she gave him a poison pill assignment. Thus James arrived in Calcutta 1860 with the Queens permission to charter a bank by and for Indian people. Within 9 months he died of diarrhea. 75 years later Calcutta was where the Young Fazle Abed grew up - his family accounted for some of the biggest traders. Only to be partitioned back at age 11 to his family's home region in the far north east of what had been British Raj India but was now to be ruled by Pakistan for 25 years. Age 18 Abed made the trek to Glasgow University to study naval engineering.
- 7 Middle East & Stans
- 8 Med Sea
- 9 Africa
- 10 Latin Am /Carib
- 11 Arctic Circle
- 12 UN
1943 marked centenary autobio of The Economist and my teenage dad Norman prepping to be navigator allied bomber command Burma Campaign -thanks to US dad survived, finished in last class of Keynes. before starting 5 decades at The Economist; after 15 years he was allowed to sign one survey a year starting in 1962 with the scoop that Japan (Korea S, Taiwan soon hk singapore) had found development mp0de;s for all Asian to rise. Rural Keynes could end village poverty & starvation; supercity win-win trades could celebrate Neumanns gift of 100 times more tech per decade (see macrae bio of von neumann)
Since 1960 the legacy of von neumann means ever decade multiplies 100 times more micro-technology- an unprecedented time for better or worse of all earthdwellers; 2025 timelined and mapped innovation exponentials - education, health, go green etc - (opportunities threats) to celebrating sustainability generation by 2025; dad parted from earth 2010; since then 2 journals by adam smith scholars out of Glasgow where engines began in 1760- Social Business; New Economics have invited academic worlds and young graduates to question where the human race is going - after 30 business trips to wealthier parts of Asia, through 2010s I have mainly sherpa's young journalist to Bangladesh - we are filing 50 years of cases on women empowerment at these web sites AbedMOOC.com FazleAbed.com EconomistPoor.com EconomistUN.com WorldRecordjobs.com Economistwomen.com Economistyouth.com EconomistDiary.com UNsummitfuture.com - in my view how a billion asian women linked together to end extreme poverty across continental asia is the greatest and happiest miracle anyone can take notes on - please note the rest of this column does not reflect my current maps of how or where the younger half of the world need to linkin to be the first sdg generation......its more like an old scrap book
how do humans design futures?-in the 2020s decade of the sdgs – this question has never had more urgency. to be or not to be/ – ref to lessons of deming or keynes, or glasgow university alumni smith and 200 years of hi-trust economics mapmaking later fazle abed - we now know how-a man made system is defined by one goal uniting generations- a system multiplies connected peoples work and demands either accelerating progress to its goal or collapsing - sir fazle abed died dec 2020 - so who are his most active scholars climate adaptability where cop26 november will be a great chance to renuite with 260 years of adam smith and james watts purposes t end poverty-specifically we interpret sdg 1 as meaning next girl or boy born has fair chance at free happy an productive life as we seek to make any community a child is born into a thriving space to grow up between discover of new worlds in 1500 and 1945 systems got worse and worse on the goal eg processes like slavery emerged- and ultimately the world was designed around a handful of big empires and often only the most powerful men in those empires. 4 amazing human-tech systems were invented to start massive use by 1960 borlaug agriculture and related solutions every poorest village (2/3people still had no access to electricity) could action learn person to person- deming engineering whose goal was zero defects by helping workers humanize machines- this could even allowed thousands of small suppliers to be best at one part in machines assembled from all those parts) – although americans invented these solution asia most needed them and joyfully became world class at them- up to 2 billion people were helped to end poverty through sharing this knowhow- unlike consuming up things actionable knowhow multiplies value in use when it links through every community that needs it the other two technologies space and media and satellite telecoms, and digital analytic power looked promising- by 1965 alumni of moore promised to multiply 100 fold efficiency of these core tech each decade to 2030- that would be a trillion tmes moore than was needed to land on the moon in 1960s. you might think this tech could improve race to end poverty- and initially it did but by 1990 it was designed around the long term goal of making 10 men richer than 40% poorest- these men also got involved in complex vested interests so that the vast majority of politicians in brussels and dc backed the big get bigger - often they used fake media to hide what they were doing to climate and other stuff that a world trebling in population size d\ - we the 3 generations children parents grandparents have until 2030 to design new system orbits gravitated around goal 1 and navigating the un's other 17 goals do you want to help/ 8 cities we spend most time helping students exchange sustainability solutions 2018-2019 BR0 Beijing Hangzhou:
Girls world maps begin at B01 good news reporting with fazleabed.com valuetrue.com and womenuni.com
.==========
online library of norman macrae--
==========
Ma 2 Ali Financial
- 1972's Next 40 Years ;
- 1976's Coming Entrepreneurial Revolution; 12 week leaders debate
- 1982's We're All Intrapreneurial Now
- What will human race produce in 20th C Q4? - Jan 1975
- (1984 book on net generation 3 billion job creation) ...
- 1991 Survey looking forward to The End of Politicians
- 1975 Asian Pacific Century 1975-2075
- 1977 survey China
- first of 4 hemisphere remembrance parties- The Economist Boardroom
health catalogue; energy catalogue
Keynes: 2025now - jobs Creating Gen
.
how poorest women in world build
A01 BRAC health system,
A02 BRAC education system,
A03 BRAC banking system
K01 Twin Health System - Haiti& Boston
Past events EconomistDiary.com
include 15th annual spring collaboration cafe new york - 2022 was withsister city hong kong designers of metaverse for beeings.app
© 2026 Created by chris macrae.
Powered by
