• Akira Toriyama passed away

    Let's all commemorate together his legendary work and his impact here

The Prosumer model stays winning. Goldman Sachs estimates humanoid robot manufacturing cost: $30k to $150k per unit. Let's look at the competition.

ggx2ac

Member
Expert
(They publish Dark Souls)
As usual, if you have not heard of the prosumer model. Linked here is the thread where I talk about it in detail: https://www.installbaseforum.com/fo...e-industry-part-4-rise-of-the-prosumers.2393/

From that thread, you would remember I created an imaginary scenario where you are convincing a bunch of venture capitalists to give you money to produce a pizza shop where the consumer is the worker. In that imaginary scenario I had one of the VCs ask:
"Why wouldn't I just go to a pizza shop where the AI itself can make and cook the pizza for me?"
To which my answer was:
"To do that, you would need a robot and robots are expensive to assemble and maintain, by having the consumer who has intelligence and a body, they can go and do that work themselves, saving us money on capital expenses."

Goldman Sachs, the finance company that is known for producing this report in 2022 on the Metaverse which stated the following:

The Future

Bloomberg estimates the metaverse to be an $800 billion market opportunity with multiple revenue streams across various industries by 2024.5 As the digital economy shifts to the virtual world, the metaverse has the potential to evolve into an $8 trillion market over the next two decades.6

5 Source: Bloomberg and Goldman Sachs Asset Management, SAS. As of December 1, 2021.
6 Source: Goldman Sachs Global Investment Research and Goldman Sachs Asset Management, SAS. As of December 10, 2021.
Source: https://www.gsam.com/content/gsam/u...hts/gsam-connect/2022/into-the-metaverse.html

By the way, if you're wondering what the value of the Metaverse market is currently, this report by Statista from September 2023 which costs $995 to order, states the following:
The Metaverse market covers a wide range of segments in the digital economy. In total the market covers 10 segments, namely, Metaverse Eduacation, Liver Entertainment, Advertising, Digital Media, AR & VR Hardware, Virtual Assets, Health and Fitness, Gaming, eCommerce, and Workplace. The worldwide revenue of US$44.1 billion in 2022 is expected to grow to US$484.8 billion up to 2030.
Source: https://www.statista.com/study/132822/metaverse-market-report/

Yes, if you go to that link right now, you will see the words "Metaverse Eduacation" and "Liver Entertainment" right next to a button that says, "Order Report" and that report costs $995.

Anyway, back to the topic. I didn't even know how much it would cost to manufacture a robot but, Goldman Sachs released a report at the end of February 2024 linked here: https://www.goldmansachs.com/intell...or-robots-could-reach-38-billion-by-2035.html

In that report titled, "The global market for humanoid robots could reach $38 billion by 2035". They estimate the cost of manufacturing a humanoid robot:
There are signs that robot components, from high-precision gears to actuators, could also cost less than previously expected, leading to faster commercialization. The manufacturing cost of humanoid robots has dropped — from a range that ran between an estimated $50,000 (for lower-end models) and $250,000 (for state-of-the art versions) per unit last year, to a range of between $30,000 and $150,000 now. Where our analysts had expected a decline of 15-20% per annum, the cost declined 40%.

The biggest problem is, I have looked at the competitors who are manufacturing robots and of course we, the uninformed majority aren't allowed to know the cost of their products. However, that doesn't stop us from knowing the specs of their robots and the one most recently in the spotlight is the Figure 01 from the company Figure.

Source: https://www.figure.ai/

The Figure 01 is a humanoid robot with a height of 5'6" (167cm), Payload: 20kg (most likely referring to how much weight it can hold), Weight: 60kg, Runtime: 5 hours, Speed: 1.2m/s (Do they mean running?), System: Electric.

Let's go back to the pizza shop from my thread about the prosumer. Prosumer vs robot. In the pizza shop operated by prosumers, there are no workers being paid wages, it becomes easy to undercut the competition with regards to pricing because of the reduction in operating expenses from less wages being paid.

Meanwhile, here is the robot competition (and there are other demonstrations in their channel):


I do have a few thoughts about the part where the robot hands over the apple, is the robot's hands clean? Or are the hands coated with silver? Was it even able to discern a real apple from a fake apple when it was giving someone food? Etc.

The Figure 01 is a humanoid robot powered by OpenAI (The company behind ChatGPT). You can see it's not a simple robot, it's got fingers and apparently according to this article regarding a different robot called Digit from Agility Robotics: https://www.bloomberg.com/news/arti...-provide-glimpse-of-workplace-humanoid-robots

In place of hands, Digit has paddles at the end of its arms, and it can pick up boxes only when it uses both limbs together. Replicating the dexterity and precision of human hands has long vexed engineers. (A breakthrough robotic arm announced by Amazon in 2022 struggled to handle more than a third of the company’s inventory.) Agility’s clamplike approach sidesteps that hurdle for now. For years the robot was headless, but engineers added a blocky head with a pair of digital eyes to offer human co-workers cues about where it was moving.

You can see that replicating a human hand adds to complexity which means higher costs. Considering that Figure 01 is an AI-powered humanoid robot with complex parts, it is reasonable to expect that its manufacturing costs are in the upper end of that $150k estimate by Goldman Sachs.

Now let's put the pizza shops somewhere, I pick... Tokyo, Japan. Why? To show how I would win. The pizza shop running on the Prosumer model would be open 24/7. A customer can come in at any time and make their own pizza. Meanwhile, for the pizza shop that is supposed to be run by expensive AI-powered robots that may cost up to $150k per unit. If we go with the Figure 01's stats, that robot can only run for five hours, that means you'd need a second robot there to take over while the other robot needs to recharge. Not only that, what if the pizza shop gets busy? A customer isn't going to be able to make a robot act faster by telling it to hurry because that robot is restricted by its software and hardware capabilities. Also, I didn't think about this until now. What if there was some guy there heckling the robots that they would be distracted from doing their job? You think the robot is going to kick the person out?

Anyway, now the pizza shop run by robots will need at least an extra two robots or even more to handle peak times when more customers are ordering pizza. Meanwhile, what does the pizza shop running on the prosumer model have to do in peak times? More space for people to assemble their pizzas and more ovens to cook more than one pizza at the same time. The operating costs have gone up for me to handle peak times but it's nowhere comparable to a pizza shop run by 4 or more robots that cost up to $150k each and need to recharge once every 5 hours of runtime.

You the reader might say, "This is an unfair comparison. These robots are being made for industrial purposes; it doesn't make sense to put them in the service sector".

Let's do a deep dive on these companies and their robots then but, before we do that. Why else did I make those pizza shops appear in Japan? Because one of these robot manufacturers mentioned how their robots could replace an aging workforce, especially one that may not be having children, or their children don't want to do those roles.

The Competition

Figure

Let's start with the company Figure. You've already seen the specs of their robot, here is why it is a humanoid:

Why a human form factor?​

We’ve designed our world for the human form. Hands allow us to open doors and use tools; arms and legs allow us to move efficiently, climb stairs, lift boxes, and more.

Figure 01 brings together the dexterity of the human form and cutting edge AI to go beyond single-function robots and lend support across manufacturing, logistics, warehousing, and retail.

From their website at the Master Plan page: https://www.figure.ai/master-plan

ROADMAP TO A POSITIVE FUTURE POWERED BY AI

By: Brett Adcock, Founder & CEO
May 20, 2022

Background: I’m 20 years into building technology companies, previously the Founder of Archer ($2.7B IPO) and Vettery ($100M exit). My sole focus is Figure. My ambition is to build this company with a 30-year view, spending my time and resources on maximizing my utility impact to humanity.

I will summarize key points from their master plan since the page is long:
  • Develop general purpose humanoid robots.
  • They claim there are 10M unsafe/undesirable jobs in the US. Their robots will solve the problem of an aging workforce in the face of declining labor supply.
  • They say, "manual labor compensation is the primary driver of goods and services prices, accounting for ~50% of global GDP (~$42 trillion/yr)"
  • They claim the cost of labor will go down as robots "join the workforce" everywhere from factories to farmland.
  • Robots will build other robots, "the cost of labor will decrease until it becomes equivalent to the price of renting a robot, facilitating a long-term, holistic reduction in costs. Over time, humans could leave the loop altogether as robots become capable of building other robots — driving prices down even more."
  • "Manual labor could become optional and higher production could bring an abundance of affordable goods and services, creating the potential for more wealth for everyone.", more wealth for "everyone", definitely not 99% of the wealth going to the 1% who own the robots.
  • "We will not place humanoids in military or defense applications, nor any roles that require inflicting harm on humans. Our focus is on providing resources for jobs that humans don’t want to perform."

Don't give me that. When they run out of robots to sell that are supposed to do the undesirable jobs, what are they going to do next? Sell robots to do other types of jobs they said they weren't going to do. It's either that or planned obsolescence, make the robots last two years and then you have to spend another $150k per unit to replace them. Having robots that build other robots would make their own company obsolete.

Highlighting more points:
  • "We believe humanoids will revolutionize a variety of industries, from corporate labor roles (3+ billion humans), to assisting individuals in the home (2+ billion), to caring for the elderly (~1 billion), and to building new worlds on other planets. However, our first applications will be in industries such as manufacturing, shipping and logistics, warehousing, and retail, where labor shortages are the most severe. "
  • "Unit Cost: We’re aiming to reduce individual humanoid unit costs through high-rate volume manufacturing, working towards a sustainable economy of scale. We are measuring our costs through the fully burdened operating cost/hour. At high rates of volume manufacturing, I am optimistic unit cost will come down to affordable levels."
They make it pretty clear here what other industries they will disrupt; it's not going to stop at jobs that are undesirable. They will move from industry-level robots to consumer robots. Don't even waste time looking at their idea of using robots to build infrastructure on other planets.

Remember the IMF said that 40% of jobs globally are exposed (at risk) to AI, this doesn't account for the jobs that AI-powered robots will replace. Let's pretend that 90% of jobs globally will be at risk from AI and AI-powered robots, who are the remaining 10% that are safe? Obviously, it's the jobs that belong to the ruling class. The ones that have power will hold onto their jobs unlike you workers. What are their jobs? They are the owners, they are the rentier, they own capital, they own assets, they will continue to accrue wealth while you have nothing because their job is to own stuff.

At this point you're thinking, "Hey, wait a minute! If I can get a robot that can build other robots, then those robots will make me money!"

You've forgotten something very critical, who owns the robots? You can build more robots but, if I was the person who is the IP owner of those robots, then you should remember how often I talk about economic rent. Those AI-powered robots are running on software, if they don't have that software then they're effectively useless. So, what am I going to do to make money off of you? That's right, you should know this one by now, the subscription model. You will pay a subscription per robot that you have, meaning that I get to continue making money in my sleep while you struggle to compete with millions of other "entrepreneurs" that are finding work to make money.

Agility Robotics

Let's move onto the next company, Agility Robotics: https://agilityrobotics.com/

Their vision is pretty similar to Figure: https://agilityrobotics.com/about/our-vision

People are the most valuable resources of our economy, yet we are faced with so many roles that exert the limits of our physical limitations and don’t leverage our unique strengths as humans. This, combined with the aging population and the decrease in birth rates in most countries has exerted tremendous pressure on our industries, which is set to only get worse in years to come.

We believe in a future where people and robots can co-exist, where robots can take over the dull, dangerous, dirty, and repetitive tasks demanded by the industry, while people do what people do best; elevate our society with creativity, problem-solving, collaboration, strategic thinking, and human compassion.

Here's their Digit robot which was mentioned in the thread earlier:


They don't provide many specs: https://agilityrobotics.com/products/digit
  • 5'9" height
  • Reach: Floor to 5'6"
  • Carrying Capacity 35lbs

Notice the pattern of dangerous jobs mentioned by all these companies? Why is it that these workers get injured or killed in their jobs? Why is it that they're not becoming safer after the workers tell their bosses to fix the issues? Hmm?

From that Bloomberg article I linked before, look at this quote: https://www.bloomberg.com/news/arti...-provide-glimpse-of-workplace-humanoid-robots
For Amazon, which is under fire from regulators for a high injury rate in its warehouses, humanoid bots could take the place of people in some of the most dangerous chokepoints. Washington state’s workplace safety regulator, which has imposed fines on the company for violations at a handful of warehouses, say moving boxes of inventory to or from conveyors—Digit’s wheelhouse—was among the most injury-prone jobs in its facilities. Amazon, which is appealing the fines, says it’s working to improve safety.

Hey, why does Amazon have such a high injury rate in its warehouses? Surely it makes more sense to figure out why and fix it than to skip that step and throw Agility Robotics' Digit robots at the solution.

1X Technologies

Moving onto the next company, 1X Technologies: https://www.1x.tech/

Here's one of their androids in action called EVE:


Yes, they did have "that" cube in the video.

They also have a bipedal robot called NEO in production; we'll focus on EVE since you have noticed that it runs on wheels. Here is EVE's technical specs: https://www.1x.tech/androids/eve
  • 1.86m in height
  • 86kg weight
  • 14.4km/h top speed (since it uses wheels)
  • 15kg carrying capacity
  • 6-hour run time
Here's NEO if you wanted to look at its specs: https://www.1x.tech/androids/neo

1X Technologies wrote a piece about how they want to create gearless robots: https://www.1x.tech/discover/a-gearless-future

I couldn't find their company vision, so the closest thing was from this blog: https://www.1x.tech/discover/the-android-revolution-transforming-civilization-with-information

The potential benefits of widespread android deployment are vast. With their ability to assist in various tasks and augment human capabilities, androids have the power to raise the living standards of people across the globe. They can enhance productivity in agriculture, construction, healthcare, and countless other industries. Androids can take on dangerous and repetitive jobs, allowing humans to focus on creative and strategic endeavors. They can revolutionize education, providing personalized and immersive learning experiences. Moreover, they can foster a more inclusive and equitable society, ensuring that the benefits of AI and automation are accessible to all.

So again, very similar visions to the other companies.

I found the part I was looking for, in the product page for EVE, there is an FAQ and in that FAQ is this question:
Will EVE replace human labor?

Humanoid androids can automate repetitive, dangerous, or physically demanding tasks that humans traditionally handle. Naturally, this could mean fewer job opportunities in those areas. However, the development, programming, and maintenance of humanoid androids can create new jobs in robotics and automation industries. Humans can also be trained for more complex roles, as androids take over certain simple tasks.

Many human qualities are near impossible to replicate, so while labor as we imagine it today might change, 1X believes that androids will create new opportunities for everyone’s benefit.

There are a few problems I have with the claim that, "However, the development, programming, and maintenance of humanoid androids can create new jobs in robotics and automation industries. Humans can also be trained for more complex roles, as androids take over certain simple tasks."

First of all, what training? There are no jobs that exist where you get trained. When you apply for an entry level job, they're looking for someone that has minimum five years of experience in the role that you are applying for. Corporations don't train their staff anymore to upskill them, that hasn't happened for decades, what they do is poach workers from their competitors to fill those vacant roles. The corporation's role is to give you no upward mobility while you work for them, they are happy for you to stay in the same role paying you low wages for as many years as possible.

Second, the claim that while jobs will be replaced, new ones will appear in the robotics and automation industries. Isn't that fun? Your job just got replaced but don't worry! Just take on six-figures of debt and attend a university to get a degree with a major in robotics/automation.

By the way, because you are living under a nation run by a neoliberal government, they are not going to create an industry and jobs for you so that you can start earning money to pay off that debt, why? Because under neoliberalism, everything is left to the "free" market.

So, what happens next? You now have to compete for jobs against others but because there are so many people trying to enter the robotics/automation industry, you fail to get a job for who knows how long. You are left at the mercy of the private companies whose role is to earn a profit and one of the biggest obstacles to profit are wages so they're going to restrict how many people they hire.

The private companies will be happy with the supply of labor that has appeared and then they will select the few that will receive jobs and the excuse that they will make for why the others don't get jobs is, "It's your fault that we went for someone better. Your grades weren't high enough, you didn't go to the right university, we wanted someone with more experience, you don't fit in our team culture (In team culture they're looking for sheep by the way, they're looking for someone that just follows orders. No independent thoughts. If they tell you that you have to work 80 hours per week with no overtime pay, the only answer they expect from you is "yes" and to smile, no excuses.)", etc.

So now you have this massive pool of people who are trying to enter this industry because they wanted to not be paid low-wages and, they were told to enter this industry because their government kept saying, "STEM is the future of our economy" while the government does not even provide the jobs and they expect you to take on debt to get qualified to apply for those jobs.

So, what are you left with? You either go back to whatever low wage jobs are available as you continue to struggle getting into the robotics/automation industry of which you may end up never getting into, or, you become an "entrepreneur" and start-up your own business and convince VCs to invest in you, or, you have to look abroad and convince a company from another country to hire you so then you can move to that country to work for that industry finally.

The answer is never so simple as, "new jobs will be created". You removed the repetitive, monotonous jobs and now have jobs that require more qualifications which expect a higher wage attached. It's pretty obvious that only a few people are getting those higher wage jobs, "the rest can go and rot".

At this point you are probably thinking, "This is why we need Universal Basic Income."

You're not going to get Universal Basic Income; you've probably already come to the conclusion why if you have been reading all of my threads. To make it even work these are a few things you would need to do: de-privatize the infrastructure and utilities that were privatized so that cost-of-living standards go down, you need to move the tax burden away from the working class and put it onto the rentier who has all the tax benefits, you have to stop the asset price inflation of housing and produce housing that isn't privately owned because: landlords.

If you do not bring the cost-of-living down to make universal basic income work, then it's going to end up being a "conditional" basic income such as, "We can't give you more money to cover being able to live in this expensive region, consider moving to a remote area where housing is cheaper".

Boston Dynamics

Boston Dynamics are the odd one out since their products are not humanoid robots, they have one robot that looks like a dog and another that looks like a giant arm with suction cups attached to it. They look more like robots you'd encounter in a sci-fi video game.

Spot: https://bostondynamics.com/products/spot/

Stretch: https://bostondynamics.com/products/stretch/

They are also trying to make a general-purpose humanoid robot, it's in R&D:

Over 30 Years of Expertise​

We set out to build a robot that could go where people go.
The commonly referenced “dull, dirty, and dangerous” tasks of robotics don’t always occur on a neatly organized factory floor. They pop up in the natural world and human-built environments. Our designs are ultimately motivated by functionality—to move and balance dynamically in unstructured, unknown, or antagonistic terrain.

Today, we combine an established legacy, roots in R&D, and an entrepreneurial mindset to create the foundation for our commercial organization.

Robots working in the real world​

Spot, our quadruped robot, is unlocking digital transformation for hundreds of customers. Stretch, a versatile case handling robot, is automating back-breaking warehouse tasks. And Atlas, the world's most dynamic humanoid robot, is pushing the boundaries of robotics R&D and taking the first steps toward a true general purpose robot.


How I'd win

That's enough of the competition to look at. You can see how they're pretty much going for the same goal of replacing the dangerous/undesirable jobs. Then they'll obviously expand beyond that.

You've also seen how a common bottleneck in these general-purpose robots are: runtime of 5-6 hours, hands that are not humanlike due to engineering complexities.

Then the other bottleneck that occurs, the price. There are only so many companies that are willing to fork out $150k+ per unit (reminder, that's the manufacturing cost) to purchase a fleet of robots. So this idea of reducing unit cost through "high-rate volume manufacturing, working towards a sustainable economy of scale" isn't going to happen as magically as they expect especially with the number of competitors targeting the exact same market of dangerous jobs.

If they want to reduce unit costs, then they have to go for the consumer market and literally everyone, man and woman would want a robot that replaces the role of the housewife. Women would especially love getting a robot because they would no longer have to do work at home that they don't get paid for. All the tasks of cooking, cleaning, taking care of kids/grandparents are put onto the robot. The moment one of these competitors produces a robot that can do all this, they will easily gain a monopoly in the robotics market among their competitors.

Why would I pick the consumer market? Because I've already shown how I would beat robots in the service sector thanks to the Prosumer model. There's a reason the Roblox company has 1600 product or engineering staff being paid wages but 7.5M "Creators" who have received an average payout of 13,500 Robux (and the median payout was 50 Robux) in 2022: https://www.installbaseforum.com/fo...part-4-rise-of-the-prosumers.2393/post-237686
 
Last edited:
Back
Top Bottom