Suzanne Gildert is the CEO of Sanctuary AI which is making Synthetic Humanlike Robots. This is like the kind of human-mimicking robots of the movies Bladerunner and the TV Show Westworld.
Prior to Sanctuary AI, Suzanne was co-founder of Kindred AI. She oversaw the design and engineering of the company’s human-like robots and was responsible for the development of cognitive architectures that allow these robots to learn about themselves and their environments.
Kindred AI is currently mostly involved in tele-operated robots performing work in warehouses.
Before founding Kindred, Suzanne worked as a physicist at D-Wave, designing and building superconducting quantum processors, and as a researcher in quantum artificial intelligence software applications.
Suzanne received her Ph.D. in experimental physics from the University of Birmingham (UK) in 2008, specializing in quantum device physics, microfabrication techniques, and low-temperature measurements of novel superconducting circuits.
Sanctuary AI wants to make general purpose human-like robots
The goal of human-like robots can be seen as being on a spectrum of some existing product.
There are sex-dolls without any actuators or movement. There are more expensive versions that have more realistic texture and feel for the skin and which appear to look more human. More realistic sex dolls have a market of a few hundred per year and some makers in China and Japan might make a few thousand per year.
The robotic actuation sex dolls are rare. There are perhaps produced in the tens per year. Most often they are built for marketing purposes.
However, Sanctuary AI will go far beyond those kinds of products using and applications using advanced robotics, new materials and advanced artificial intelligence.
Applications like human-like elder care companionship and synthetic human alternatives will become possible. Situations that were partially explored in fiction like Bladerunner will become truly possible. The negative situations explored in fiction will be what should mostly be avoided with careful work in real life.
Suzanne Gildert and Sanctuary AI wants to not sell their product but charge for them based on task and time. This kind of renting is how we charge for the gig or by the hour for people.
Fluidic muscle, Human like skin and better actuators
Sanctuary AI is taking fluidic muscle created by the company Festo. Festo has developed a pneumatic muscle that is recreated from a biological muscle. It enables sequences that closely approximate human movements.
The bionic muscles consist mainly of a hollow elastomer cylinder embedded with aramid fibers. When the fluidic muscle fills with air, it increases in diameter and contracts in length, enabling a fluid, elastic movement.
Sanctuary AI is taking advanced 3D printing of carbon fiber to make advanced lightweight skeletal structure in half a day to a day.
They are taking new actuators with magnets.
They are taking the best forms of synthetic skin.
Synthetic skin is one of the areas which will require a lot more improvement to enable human-like synthetic robots.
Suzanne believes we could ten years away from fully human-like synthetic robots.
Applications for human-like synthetic robots
Suzanne discussed virtual teleportation, where people with different skills could beam into a human-like robot for different tasks.
AI could learn from the teleoperation and eventually the humans would not need to be in the loop for some tasks.
I mentioned that the highly valuable skills for remote assistance are currently in telemedicine. Having human-like robots would likely enhance telemedicine. A doctor, surgeon or psychologist could interact more effectively using human-like robots.
There is also the enhancement of video games and theme parks with human-like or fantastical but high fidelity robots. Characters from movies and fiction could be brought to life.
Some of the fantastical could make up for being not quite perfect human imitation. Superhero or magical characters could be created.
Disney had animitronic characters at their theme park for many decades. Human-like robotics could be superior and more flexible interaction than the animitronic routines.
Suzanne indicated that specific people could be modeled for the human like robots. One of the next robots could look like her.
Certain wealthy people could have a use for robots that looked and acted like them. Celebrities could have a need for such robots.
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.
Anthropomorphism can be useful in some contexts, but it’s darn hard to do it right. The Boston Dynamics bipeds (and quadrupeds) are most impressive though, and the closest we are to have a workable platform for humanoids. These mannequins on the other hand, are just parlor tricks and convention room curiosities in terms of applications. They do nothing that a Disney animatronic didn’t do since the 50s.
Cute scientist lady… Making the robots look like people is not really useful though. The Boston Dynamics quadruped “doggies” look pretty natural and useful…. rubber face on a biped – not so much.
Anthropomorphism can be useful in some contexts but it’s darn hard to do it right. The Boston Dynamics bipeds (and quadrupeds) are most impressive though and the closest we are to have a workable platform for humanoids.These mannequins on the other hand are just parlor tricks and convention room curiosities in terms of applications. They do nothing that a Disney animatronic didn’t do since the 50s.
Cute scientist lady… Making the robots look like people is not really useful though. The Boston Dynamics quadruped doggies”” look pretty natural and useful…. rubber face on a biped – not so much.”””
I’m pretty sure lots of the first jobs for these mannequins will be something Mr. Disney would not approve of. So yeah. In the meantime I refer you to my previous statement that “not all great filters are asteroids”.
Change the g so your quadruped-centaur remotely operated robot walks around. When you are done you park it and take a break. Really no reason to have astronauts walk around in bespoke suits who are only productive between certain bio breaks.
I mean ‘g’ not ‘G’
Imagine if all they had to do was change the gravitational constant in the Boston Dynamics software for the (now 6-8 legged) robot to move well on the moon. I say ‘imagine’ because I bet there is a lot of empirical stuff in the control software based on the ‘screwdrivers’ in the codes that I use (the ones I am aware of).
Much better would be a remotely operated quadruped robot with 2-4 arms. Replace spacesuits so people don’t go out on the surface of the Moon/Mars. Just saying.
Lets make Bladerunner in real life” Yeah hard pass on that.
I’m pretty sure lots of the first jobs for these mannequins will be something Mr. Disney would not approve of.So yeah. In the meantime I refer you to my previous statement that ot all great filters are asteroids””.”””
Change the g so your quadruped-centaur remotely operated robot walks around. When you are done you park it and take a break.Really no reason to have astronauts walk around in bespoke suits who are only productive between certain bio breaks.
I mean ‘g’ not ‘G’
Imagine if all they had to do was change the gravitational constant in the Boston Dynamics software for the (now 6-8 legged) robot to move well on the moon. I say ‘imagine’ because I bet there is a lot of empirical stuff in the control software based on the ‘screwdrivers’ in the codes that I use (the ones I am aware of).
Much better would be a remotely operated quadruped robot with 2-4 arms. Replace spacesuits so people don’t go out on the surface of the Moon/Mars. Just saying.
Lets make Bladerunner in real life””Yeah hard pass on that.”””
As the movie effect guys found long ago, the more human something is, the easier the audience will spot a fake. Robot sharks and dinosaurs? Easy. Robot mammals like bears and lions? Not easy. Robot familiar mammals like dogs and cats? Hard. Robot humanoids like goblins and orcs? Very hard. Robot humans? Not possible yet.
No, you don’t understand. It’ll be more like Westworld! What could possibly go wrong?
As the movie effect guys found long ago the more human something is the easier the audience will spot a fake.Robot sharks and dinosaurs? Easy.Robot mammals like bears and lions? Not easy.Robot familiar mammals like dogs and cats? Hard.Robot humanoids like goblins and orcs? Very hard.Robot humans? Not possible yet.
No you don’t understand. It’ll be more like Westworld!What could possibly go wrong?
Look at those blow job lips!
Look at those blow job lips!
How about a pleasant looking face on future robots in the service industry? I’d think that would be better than trying to interact with a disembodied voice or a cylinder with mechanical arms and cameras for that matter.
How about a pleasant looking face on future robots in the service industry? I’d think that would be better than trying to interact with a disembodied voice or a cylinder with mechanical arms and cameras for that matter.
How about a pleasant looking face on future robots in the service industry? I’d think that would be better than trying to interact with a disembodied voice or a cylinder with mechanical arms and cameras for that matter.
Look at those blow job lips!
As the movie effect guys found long ago, the more human something is, the easier the audience will spot a fake.
Robot sharks and dinosaurs? Easy.
Robot mammals like bears and lions? Not easy.
Robot familiar mammals like dogs and cats? Hard.
Robot humanoids like goblins and orcs? Very hard.
Robot humans? Not possible yet.
No, you don’t understand. It’ll be more like Westworld!
What could possibly go wrong?
I’m pretty sure lots of the first jobs for these mannequins will be something Mr. Disney would not approve of.
So yeah. In the meantime I refer you to my previous statement that “not all great filters are asteroids”.
Change the g so your quadruped-centaur remotely operated robot walks around. When you are done you park it and take a break.
Really no reason to have astronauts walk around in bespoke suits who are only productive between certain bio breaks.
I mean ‘g’ not ‘G’
Imagine if all they had to do was change the gravitational constant in the Boston Dynamics software for the (now 6-8 legged) robot to move well on the moon. I say ‘imagine’ because I bet there is a lot of empirical stuff in the control software based on the ‘screwdrivers’ in the codes that I use (the ones I am aware of).
Much better would be a remotely operated quadruped robot with 2-4 arms. Replace spacesuits so people don’t go out on the surface of the Moon/Mars. Just saying.
“Lets make Bladerunner in real life”
Yeah hard pass on that.
Anthropomorphism can be useful in some contexts, but it’s darn hard to do it right. The Boston Dynamics bipeds (and quadrupeds) are most impressive though, and the closest we are to have a workable platform for humanoids.
These mannequins on the other hand, are just parlor tricks and convention room curiosities in terms of applications. They do nothing that a Disney animatronic didn’t do since the 50s.
Cute scientist lady… Making the robots look like people is not really useful though. The Boston Dynamics quadruped “doggies” look pretty natural and useful…. rubber face on a biped – not so much.