The Tesla Teslabot and the Agility Robotics Digit will both be mass produced in the thousands in 2024. Agility Robotics announced a factory they had started last year would be completed and in production late in 2023. Mass production takes a few months to ramp up. The Agility Robotics factory will have a capacity of 10,000 Digit robots per year.
Digit is the first human-centric, multi-purpose robot made for logistics work. It is designed from the ground up to go where people go and do useful work safely in spaces designed for people, starting with bulk material handling within warehouses and distribution centers.
Tesla is completing a factory to mass produce the critical actuator component and the entire Teslabots in substantial volume. Tesla showed off recent video showing the Teslabot learning from video and copying human actions. Teslabot is moving freely without cables and is performing precise sorting and even yoga balancing motions.
Elon Musk’s advantage with Teslabot is similar to the advantage for Falcon 9 launches. Elon can be his own primary customer for Teslabot just as SpaceX became the primary customer for Falcon 9 rocket launches. 70-80% of the SpaceX launches are used to launch Starlink satellites.
Tesla is increasing the number of human staff at the Gigatexas factory to 60,000 people. Tesla already has about 127,000 worker across its global operations. SpaceX has 12000 workers in 2023. If Tesla and SpaceX moved to have Teslabots perform half of the human factory, warehouse and delivery tasks and both companies grow 50% per year, then there would be internal demand for 300,000 Teslabots over the next two years.
If the Teslabots cost $10,000 each to produce then the capital cost to Elon’s companies would be $3 billion in hardware. The cost to Tesla would more likely be about $5000 per bot for less the 4% of the batteries for a car, chips, cameras etc.. Tesla already buys the chips, cameras and batteries in volume for its cars. This would reduce the hardware cost for 300,000 Teslabots down to $1-1.5 billion.
There is a need for neural net software training with a lot of Bots. $3 billion over 2 years would be less than half of the Tesla’s R&D budget over two years. Tesla spent $3.1 billion in R&D in 2022.
Brian Wang is a Futurist Thought Leader and a popular Science blogger with 1 million readers per month. His blog Nextbigfuture.com is ranked #1 Science News Blog. It covers many disruptive technology and trends including Space, Robotics, Artificial Intelligence, Medicine, Anti-aging Biotechnology, and Nanotechnology.
Known for identifying cutting edge technologies, he is currently a Co-Founder of a startup and fundraiser for high potential early-stage companies. He is the Head of Research for Allocations for deep technology investments and an Angel Investor at Space Angels.
A frequent speaker at corporations, he has been a TEDx speaker, a Singularity University speaker and guest at numerous interviews for radio and podcasts. He is open to public speaking and advising engagements.
We are all screwed when the bots come out, working class anyways. They will be cutting jobs left and right. I think they’ll be a lot of civil wars in the future. The rich are just too greedy. The tricky part will be when the robots start learning, bc that’s coming too. We might be 5 years or more away from job replacement, but its coming.
In an industrial environment, where repetitive tasks are performed at stations along an assembly line, it might be most efficient to have a central controlling “Puppet Master” AI assigning individual bots tasks.
No need for complete autonomy for each bot under those conditions, which translates to less expensive individual bots.
Outside of such confined use cases, go full Rosie the Robot.
A reverse use scenario would be to have all the needed AI capacity in the bot, which can chauffeur me to and from work in an otherwise “dumb” electric car, then go off to perform other mixed tasks.
Taxi service, shopping, running my household, etc.
Two levels of Teslabot architectures.
Near term military uses for a fully capable Teslabot could be driving fuel and supply trucks in combat zones, and flying existing fighter jets as a short term, half step in the direction of fully autonomous drones.
Nothing scary about watching a robot climbing into a fighter cockpit.
Nope.
There are Greedy Controlling people leaders the Rich .. out here and robots are power n ya can program them to do what ever . N I say No. History repeats it’s self. Bad idea. To make these robots. Lol just saying .
Why didn’t the Baxter robot by Rethink Robotics take off? The cost was OK. But who was going to program it for the very specific tasks that businesses need. If it is going to work with humans, it needs to be as fast, as agile, as strong, and as accurate as humans. These are just less friendly, scaled up Baxters.
Digits still does not have fingers… The Tesla bot is obviously designed to make more tasks than carrying mid size loads. Not sure that the best way to achieve this is by building a human looking robot. With some creativity a simplified and easier to build shape can be created that can make more tasks cheaper.
Electronic signage is a technology that can save a lot of time and effort instead of humans putting up sales tags and changing prices every week. The technology has been around for years. Very few stores have even started to use them.
There are medication dispensing machines in existence—have been for years—that a handful of pharmacies use and those which use them only use them an a small fraction of meds, ones that are commonly prescribed but patients get a lot of them counted out per dispensing. Everything else gets counted by technicians and checked by pharmacist as usual.
Many stores, offices and businesses that are pretty profitable and competitive use IT systems that were probably obsolete when they were implemented decades ago.
Robotic floor cleaners exist. So do employees with brooms and mops. One guess which you’re more likely to see in a store or factory.
So when we start to see humanoid robots in work places it will seem really futuristic but they will be only in a tiny fraction of businesses, they will only be doing a marginal amount of work and owners, managers and CEOs will not be rushing to avoid being left behind by being the second or third in the industry to have them. They will put it off as long as possible.
If I understand your point, businesses keep people around doing things that automated systems can do, so we shouldn’t be expecting humanoid robots to be widely adopted either.
But the advantage of humans over the automated systems you list is their flexibility, allowing them to shift between multiple tasks.
Humanoid robots will also have that advantage over fixed-purpose machines, just not to the degree of humans for some time yet. So if a store needs 10 workers at a time, maybe they’ll get by with 9 humans at a time plus 1 robot continuously doing simple tasks.
Once mass produced, robots will be far cheaper than human labor – probably around $3/hr. ($10K/yr lease, $2K/yr repairs, trivial electricity cost, replacing 4000+ hours of human labor). So it doesn’t need to be as fast or flexible as a human – just fast and flexible enough to be cost effective.
Where do you come up with those numbers?
How long will it take to train a humanoid robot capable of doing a given task?
Specialized robots have always been better than generalized robots at a given task
What are the use cases?
Meet the future MSR operations/maintenance crew – circa 2060 (by then they might actually function well enough).
All those movie take a bad turn. Just saying.
Butlerian Jihad time!