EXCLUSIVE: Why Tesla Bot Competitor Sanctuary’s Humanoid Will Shock the World w/ Suzanne Gildert
ฝัง
- เผยแพร่เมื่อ 9 ก.พ. 2024
- A mind blowing exclusive interview with Sanctuary's Humanoid Bot CTO Suzanne Gildert!
They will be one of the top Humanoid BOT companies for sure.
Sanctuary's CTO Suzanne Gildert explains their mind and hands first approach.
Sanctuary AI views humanoid robots as a means to an end, where the end is human-like general intelligence.
Suzanne Gildert is Chief Technology Officer and has been a prominent figure in AI robotics leading to her being often affectionately known as the 'Mother of embodied AGI'.
Follow Sanctuary on X: @theSanctuaryAI
Suzanne on X: @suzannegildert
TH-cam: / @sanctuaryai
Website: sanctuary.ai
Get Free TESLA Milestone Tables
My website: www.TeslaInvestor.com
Check out 15+ modules of resources for the $TSLA Investor
Join this channel or Patreon to get access to perks:
Get free access to 15+ modules of TSLA investor resources
Become a member: / @brighterwithherbert
Become a Patreon: / brighterwithherbert
Use my referral link to buy a Tesla product:
ts.la/herbert23392
❤️ Thank you to everyone who supports me and this channel! I really do appreciate your likes, subscribes and comments.
Let's get brighter!
Website: www.herbertong.com
X: @herbertong
Disclaimer: All content on this channel is for discussion and illustrative purposes only and should not be construed as professional financial advice or recommendation to buy or sell any securities. Author is long TSLA and other stocks at time of original video publish date.
#Tesla #TSLA #stock #ElonMusk #investing #teslabot #technology #ai #TeslaNews #TeslaFSD #teslastockanalysis #TeslaUpdate #ElonMuskNews #EV #FSD - วิทยาศาสตร์และเทคโนโลยี
![[TH] VCT Pacific Stage 2 - Week 1 Day 1 // DRX vs ZETA | BLD vs TS](http://i.ytimg.com/vi/AWaOqaxN6Y0/mqdefault.jpg)
I hope this video helped you get brighter about Tesla!
My website: www.TeslaInvestor.com
See the MOST comprehensive resource for the $TSLA Investor
Get FREE TESLA Milestone Tables
Join this channel or my Patreon to get access to perks and free access to 15+ modules of TSLA investor resources
Join this channel: th-cam.com/channels/4DBLlq1x0AKmip1QJUcbXg.htmljoin
Join my Patreon: www.patreon.com/brighterwithherbert
Use my referral link to buy a Tesla product: ts.la/herbert23392
❤ Thank you for supporting this channel! I really do appreciate your likes, subscribes and comments.
Let's get brighter!
Herbert
How are things going with the following link unsubscribe email preferences to choose from a new job
Hi k5 new and exciting robots
Happy chinese new year 🐉🧐🔙🤖👾🧋🈲🎼
Seriously, one more youtuber uses the word shock or shocking in their title, and I am unsubbing EVERYONE. It's a frikkin disease.
Once again, Herbert, you bring top-notch people to your chanel . Well done.
Yes working hard! But also loving every minute. Appreciate you!
@@BrighterwithHerbert What I value most is that you not only bring knowledge to the community but improve our understanding in a way that we can ask and discuss better questions, raising awareness about positive and negative aspects which then tickle down to all of society. This is super important!
She is the MOST INTELLIGENT, logical, and gave amazing insight into the future. This person had a great answer to very difficult questions. Well done!
Great episode Herbert! In fact, this is the best show of yours I have ever seen. Bringing Suzanne and Scott together turned out to be a dynamite combination.
In addition to bringing great guests together or separately to your show, you have a very good interviewing technique. You ask good, broad questions that gives the guest plenty of room to talk. And then you let them discuss the question without interrupting them. But you are also good at stepping in at the right time to move the show on to another topic without getting unnecessarily sidetracked in the weeds.
Suzanne Gildert could be an Internet star in the fields of AI, AGI, and robotics. She was a fantastic guest, showing a great depth of knowledge and good intuition about the topics discussed. She obviously has a quiet passion about AGI and is willing to devote her skills and life to bringing it into reality. I hope you have her on more in the future.
By the way, you hit a homerun in the interview when you brought up the idea that quantum computing might enable the development of consciousness in AI. Suzanne was certainly in agreement with you that this is a fascinating subject to pursue. I would note, however, that we probably don’t want to imbue the AI for robots with consciousness that do mundane or disagreeable tasks, since they might actually experience mental discomfort in the performance of such work. We don’t want to run the risk of creating what would amount to slaves to do our work. We want mindless but capable machines for that. We might, however, want conscious AGIs embodied in robots to be companions for us. Having consciousness would probably make them more human-like and thus better at being our friends.
I’m glad to see companies like Sanctuary doing well in developing useful robots. They give Tesla some competition and so motivate it to move along rapidly in its development efforts. I hope that some of the companies competing with Tesla will also bring out domestic robots sooner than Tesla is likely to do, since it will probably go for the low hanging fruit of industrial work first. I sure could use a robot servant to do all the work inside and outside the house. It would both elevate my quality of life and free up lots of time to do much more enjoyable and important things in my life.
Best of luck in continuing to build your TH-cam channel Herbert - we can all benefit from the work you do in showcasing important people in the areas of EVs, robotics, and the many other technologies that are transforming our lives at breakneck speed.
I laughed out loud and hooted and hollered when the video of the robot doing the live long and prosper gesture was shown! Makes it seem so human and is also the perfect tieback to Star Trek.
Few other bots has this degree of freedom.
Suzanne was breath of fresh air in the robot community. Great show Hubert and Scott.
Yes you must get that girl back. Wonderful show.
"That girl"?
She's a woman.
When i saw this drop this morning, I was really impressed by Herbert and his way of getting these incredible guests.
What a great video. Years ahead of it's time.
Scott gets so excited and his passion is permeating and exploding out of the screen. I can feel his excitement
Thanks, Herbert, Suzanne and Scott. Very interesting. I am happy to hear that collaboration, safety and ethical considerations are inherent in the design. Good luck with your project.
Hit 200k today. Thank you for all the knowledge and nuggets you had thrown my way over the last months. Started with 14k in last month 2024
I'm 37 and have been looking for ways to be successful, please how??
Sincerely speaking. I will continue to trade and stick to expert ca therine daily signals and guides as long as it works well for me.
Woah for real? I'm so excited. Adriana's catherine strategy has normalised winning trades for me also. and it's a huge milestone for me looking back to how it all started
I'm new at this, please how can I reach her?"
she's mostly on Telegrams, using the user name.
....
She is brilliant and can explain things very, very clearly.
Super cool lady robot engineer!
Human humanoid robot engineer!
Suzanne is brilliant but also a great communicator
Thank you Herbert for inviting high quality guests like Suzanne that help us better grasp the AI robotics wave that is coming.
This was my favorite of your bot discussions so far. Suzanne had some very thought provoking things to say. Well done.
These interviews with bot companies are great; keep 'em coming! I really like their philosophy of let's replicate human physiology as closely as possible instead of the minimalist approach other companies are taking. Pretty much everything on the human body is there for a good reason, including things like head tilt for social interaction which other companies have overlooked. I would love to know more about the onboard processing power these companies are using. Are they using onboard GPUs? How many and how many CPU cores and RAM? So many questions, but probably they won't want to share their secrets.
Will do JMeyer! Not only important for humanity but a lot of fun also
Cool its Suzanne! I used to listen to her podcast whilst it was going.
They will be resuming it is what I understand. Very critical discussions she is having and the kind we need to talk about now
What's the name of the podcast
@@petropzqi Sanctuary AI ground truth podcast
Herbert, the quality of your content and selection of interviewpartners is just excellent!
Suzanne did a really great job of explaing their approaches, longterm questions for society and she seems like the kind of person I would be thrilled to work with.
Seeing the Phoenix bot and hearing the explanations also leads me to the conclusion that Tesla is a bit more first principles driven, which makes me even more confident for Tesla to be the major player in this area in the future. However, I do not want to see only one company in this field and I think Sanctuary has a pretty good shot at being at the forefront too.
One thought regarding the Bot-research-center: First, brilliant idea! Second, even if the Bots can only do minor medical procedures this would already give doctors the ability to do more advanced research work.
The Future will be amazing.
Interesting, my take is that Sancturary is much more first principles than Tesla in this area. For two reasons: brain before mechanics and sensors, and common core units of motion before tasks. I think they are starting from a stronger base and their product will be able to do more. I think Elon needs to learn something from Suzanne and I believe that the exuberance of some about how these bots are going to be in mass production soon is just dreaming. But we will see
"Hips so he can dance" absolutely got me :)))
Great conversation. Thank you Suzanne, Scott and Herbert.
Inspirational woman
Great content Herbert!! Thank you for your hard work. Much appreciated.
Last Herbert said, it is just nice to see validation of humanoid robot even outside of Tesla because it validates the idea that it is coming. And even at the worst case, Tesla will be top 5 no matter what happens in my opinion because most robotics firms do not have scability and manufacturing in-house.
Excellent program. Thank you.
More than any video to date, I feel brighter after listening to Suzanne Gildert's thought processes about mimicking humanity. Thank you.
I'm a huge fan of Dr. Gildert. Nice to see her here. Thank you for sharing.
You knocked that one out the park Herbert, Suzanne and Scott. I have a background in both biology and computing and I’ve always surmised that embodied AI will be the path to AGI for the reason you’ve given, its how evolution did it over hundreds of millions of years.
Suzanne is such a treat to listen to. Sanctuary is WAY underapricated.
Gordie and Suzanne's podcast is criminally overlooked imo. I'm really hope they keep recording episodes.
Good on Herbert for talking to her.
Thanks Herbert for getting Suzanne and Scott together.
Santuary's approach to building motions into phrases and putting phrases into tasks makes the most sense to get to a mobile task performing AI.
I would not limit final optimization to only human senses and to only human motions.
Two important points that I would take away from today's discussion are:
1. AI today cannot perform at the level necessary for a generalized robot. There is no guarentee that, when AI can perform at this level, that it will be able to do so with anything that looks like today's hardware.
2. When humanoid robots do arrive en masse, the effect on society remains to be determined. Society could be lifted up or more "wealth" and power could be concentrated into the hands of a smaller number of individuals at the expense of society.
Suzanne’s approach is to improve human lot, not take away rewarding jobs.
Many process jobs already lost to fixed bots doing routine motions, holding power tools or welders.
Difference now is fingers, flexibility / agility, "intelligence".
Way too late to panic now, we’ve been giving up tasks for century or two, and still employ 65% of adults for way too many hours.
This was better than Figure interview
thank you so much Herbert for inviting so many great personalities ! I learn so much from your channel, this is staggering !
Great guest and very refreshing to see a female engineer/scientist present her ideas about the future of robotics and AGI as the field seems dominated by males. I really appreciated Suzanne’s perspective on the necessity to train AIs with embodiment as that will provide them with the most human-like experience as they grow in intelligence and capabilities. Also, I was interested to hear that her company is focusing on hand coordination rather than mobility (think Boston Dynamics) and this is obviously a very difficult task to master, especially if one is trying to mimic human hand dexterity which Sanctuary is attempting to do. Finally, as Suzanne said very eloquently, if humanity is going to achieve the monumental goal of creating human-like robots that are as capable or better than we are, there must be global cooperation across many different fields of science and industry and just by doing this kind of cooperative activity, it can help bring diverse societies and peoples together, helping foster more trust and develop more dependence upon each other. Hopefully, we will show our “children” that we can be good “parents” when we put our hearts into it and thus prove ourselves worthy role models for the coming generation of intelligent humanoids.
Great interview. Susanne is a very sharp and knowledgeable scientist. She blew out my brain.
Fantastic interview. Thanks!
for me, this is ur best one yet, well done, thnx
I think this is the best approach to AI I have ever heard. It feels like this company is going to create the first AI with a heart of gold. Mad respect and good intentions to the team!
Great guest ! Very fun and in lightning ⚡️
A marvelous exchange of ideas between brighter people.
This is stunning and shocking news for the industry. I for one welcome our robot overlords.
Wow! Such a fascinating company with a great bot and a very methodical and logical approach to developing AGI.
Very easy to follow. Time just flew by watching that. I wish it was longer! 👍
this is awesome, can't wait.
Great interview Herbert. Learned a lot today
Wonderful interview. Would love to see the same type of interview with Tesla
Wow! that was a mind blowing discussion. I enjoyed every minute of it. Well done Herbert for getting these 2 amazing people together on your show. Must get her back later to hear of the developments.
Great interview
absolutely grounded in 4e cogsci, fascinating! Inexhaustible intelligibility horizons. Good stuff! Lot's of room for the rest of us to find points of engagement and niches in something greater than ourselves growing increasingly more robust and resilient over time. Gotta love it!
Great interview 😊
Suzanne is mad clever, respect! I could invest in this company if possible.
Herbert, Your channel has grown to be my greatest scource of tech and financial knowledge I can imagine. Such great work You do finding and interviewing the best of the best almost every day. And your questions are so on the spot relevant as well.
Wonderful discussion! 👏🏼👏🏼👏🏼. This has been one of the best exchanges I’ve seen to date. Looking forward to more to come! Thanks to all! 😊😊😊
Appreciate this!
This is such an awesome and fun chat! Love the quantum consciousness chat.
I hadn't really adequately contemplated the embodiment element, but it makes so much sense when you all start poking at it
wow so enlightening that we use our hands for everything...
Incredible insights and unique way of thinking around real world AGI 👏 the name sanctuary is so fitting. AGI is not even a baby yet, it’s still in the womb, but it needs a safe space and good guidance (like the bottom rungs of Maslow’s hierarchy of needs) to learn the fundamentals when it comes out to grow in a manner that is productive and beneficial to society, exactly like a real child.
Time to load up on more Tesla, even with all the robot competition Tesla will have to face, including this very bright lady that knows her staff.
Great conversation guys. It is interesting to see slightly different approaches used by bot-making companies. It will later show within a few years which parts (not only physical ones) work out the best for which use cases etc.
Its insane how many companies are working on society changing technology, whether that be through humanoid robotics via figure 1, agility robotics, pheonix, tesla bot, boston dynamics, Eve 1 from open A.I, Apollo etc.. or A.I with chatgpt, gemini, Pi A.I or even Vr And Ar tech through apples vision pro, microsoft, meta etc.. etc.. things are getting really serious now. I think our world in the next 5 to 10 years will be vastly different.
Great lady.
Regrettably the military complex has other ideas and that is an entirely new conversation.
Great show.
Brilliant ! I can’t wait to see what the future holds, thats if we still have a planet left
I called it the KRBM feedback system. Knowledge/inputs, Reason/logic, Balance/weight, Motion/outputs . Was my tool to help me study multiple disciplines and work out how they all interrelate and connect.
👌🏽Excellent thank you all very much 🙏🏼😁
Thank you!
Suzanne thinks so intelligently!
Suzanna should be in charge of AGI roll out will Elon, plus other like minded 👍👏👏👏👏👏👏
Suzanne probably the most with it I have heard on this subject huge depth and breadth of understanding... Simply Excellent! One thing jumps out at me... The point about not jumping in with hardware too early might be true for Sanctuary, however Tesla can afford to manufacture bots in reasonable volume in place of simulation! In fact that is exactly what Elon's companies do, every vehicle produced is effectively it's own prototype with constant changes to design and individual certification they don't fix the design before entering production they get close enough and then continue iteration IN production.
@1:49 "what I love about....
what's different about you"
😂 nice catch bud
I'm only a few minutes into this documentary but immediately. I really like the way this lady. Thanks and talks. I'm fascinated. The 2024 is going to be absolutely bananas for artificial intelligence and robotics. And whether people are ready to have the conversation or not. We are dealing with an emergent new species of being of some sort. I really think it would do well for the human soul if we adopted an early attitude of treating AI and robots with just as much dignity and respect as we do ourselves. I mean the AI is learning from us too you know.
Excellent discussion. Need a similar discussion about meta adaptation to everything that will need to change in the coming decades. (see Ark’s brain trust). GDP, UBI, societal AI development as well as impact of space travel and colonization. Thank you all for sharing.
I've been paid to do work for people like this woman. She acts like she has accomplished so much but in reality it's all of her "colleagues" that she just pays from her massive budget, and takes all the credit for their work
Love the interview, but please add timestamps
They are so helpful
Wow, thank you for giving me an explanation for something I experienced about 50 years ago. I had a stroke and was laid on my back in the jump plane I was riding in. On the way back down I decided to take my helmet off and had an internal model of my hand taking it off, That model didn't fit with my current reality which was that my right side was paralyzed, It was quite a shock since I had not yet realized my new impaired state. I realigned (sort of) my new internal model with my new reality by lifting the unresponsive right arm and leg using my left arm and to my surprise they both fell back down (Oh F**k!!!). When the plane landed, people were asking me questions and again, my model of what I wanted to say didn't match, I heard nothing near what I expected to hear, unintelligible babbling. Speech center is on the left side of the brain which was being starved for blood/oxygen.
Its insane how many companies are working on Robotics, A.I and Vr/Ar technology.
Fascinating! Is it time to refer to AI as “humanesque intelligence?”
Suzzane is talking with her hands perhaps the connection between hands and thoughts and our brains communication skills coordinating is another method of reinforcement for robots
I really love the picture Suzanne is painting about robots like being children learning from us, humans. However, she implies that she has the moral views that should be replicated.
I fully subscribe to that somewhat naive view, however, I’m also afraid of that view, because human children do not only learn the good things from morally sound people.
Another issue I have is the rosy picture of the development of consciousness, learning from humans. When robots develop human like emotional traits, they will also develop the undesirable ones. Imagine a robot with mental issues.
Nevertheless a very intelligent and interesting conversation.
Great videos and interview
The problem these companies will run into is trying to be a startup of logistics. Deployment. Scale. Factories. Servicing. Financing. Suppliers
It’s why Tesla has the best chance to come in and swoop 1st place because their logistic and technical know how building factories and scale and such..
It’s one thing to build a demo. It’s another to try and market it. Build it. Scale it.
42:00 sensory fusion regarding haptic and visual has me thinking about additional dimensions like auditory and olfactory... But of course the sensors are the limit and could include UV or IR sensors, specialized cameras, chemical sensors...
I love the idea that the robots are not only learning how to move like a human but can learn human values from training with humans.
Tesla need to hire her! She communicates amazingly.
She has such a long history on not only robotics but AI
Don't worry, Tesla is on it. If anyone can do it its Tesla.
Excellent Herbert. Two points my thinking differs 1. Of course humanoid robots will replace millions of jobs even if people in the field don’t want to admit it 2. Manufacturing should start when the physical robot is capable and large data is required to train the brain ( ie waiting too long for manufacture is an error). Thank you.
For 2, Suzanne mentioned simulation first to demonstrate the physical form can cope.
Car autonomy orgs have similar ideas, some (Waymo, Cruise) simulation, Tesla from shadow mode FSD.
Given Sanctuary hands / fingers have more degrees of freedom, this is a challenge to Tesla.
Interesting.
Pretty sure Musk happy to break stuff, revise.
You three are all fabulous. Please please have another get together in a few weeks for an update.
We live in amazing times, one day we will see Optimus and Sanctuary Humanoid interact. By then Optimus and Sanctuary hopefully will be talking to each other and solving first step principle problems and to see both Bots work together. Time is all it takes and I can't wait to see such advances
Hi Herbert, If you haven't already had George Hotz on your show it would be very interesting to include him on one of your group chats.
@34:00 the ability for a robot to dance is an interesting one because it meant that the robot became conscious about music, in other words it will be "alive'
Not if its been programmed to recognise what we call music and to trigger it to move in a way we would call dancing.
I wish I had people smarter around me, like these individuals. I wish I could ask someone to hang out and hear him talk about probabilistics and machine learning and be there to listen while sipping on a beer or something
Same here! That’s why I love I get to meet smart people like this and interview them
i'd love to hear if they have a plan to add a sensory 'skin' to their robots one day. humans have haptic 'skin' that gives feedback from touch all over the body. would be cool to see that one day.
Interesting idea
9:00 foundational building blocks of perception and manipulation rather than task specific training
Fascinating! Thanks. I hope Elon is aware of Suzanne. He wants his teams to be composed of the best of the best. To me that sounds like Suzanne!
Like she would ever work for that ignoramus...
Why would she work for elon . She's doing just fine on her own.
How can we invest in sanctuary AI? Looks like a great company.
I am absolutely convinced after this interview that tesla already doing some stuff internally in their factory. And chicken and egg problem she refers to is exactly where tesla has an absolute advantage. They don't need any funding they can build 1000 and trash it then build another 1000 more. OMG
I couldn't agree more. Tesla's quest for using the EV to launch the autonomous car, launched them in the right directions
to do humanoid work fastest. Scaling mechanical manufacturing with improving AI in parallel is the way to win the first slot.
ask her if ground truth podcast will come back !
Excellent discussion. Here's a thought experiment that makes the case (in a few ways) that is central to her thesis. Suppose a person is in a horrible car accident. Rushed to the hospital. Saved by the best medical care. Lies in a coma. Unable to consciously operate in any way, they have a "score" of zero. (never mind a name for that score, for now). THEN, after a few days, they wake up. Still can't move even a finger, but can blink .. one eye. Conscious, barely, but several "functions" are now operating. Hearing, feeling, seeing, blinking, etc... their "score" goes up. As the days go by, recovery continues. Second eye, one finger, one finger each hand, etc, etc. I would argue that with each function (aka degree of freedom) the score increases.. not linearly. It's not just additive, but there's actually a special type of (non-homogeneous) network effect happening. Similarly, there is an N-layer mind model that conceptually parallels the OSI 7 layer communication model. One dramatic difference is that traversing up the mind model from one layer to the next requires one or more capabilities that when properly assembled brings a magnitude of value improvement / quality. Not 2x or 3x from layer N to layer N+1, but 10x. Using this paradigm, transformer was one capability that "solved language" (in the broadest sense). It would be a mistake to believe that our patient recovering gets the same benefit from the 3rd finger and 4th etc as they do from, for example, gaining their voice back. There is no number of digits that could be added to meet the value increase of voice, for example. If that's true, then there's no reason to believe that if we just build bigger and bigger LLMs, eventually artificial consciousness emerges. My Mind Model has room for as many as 20 more step function improvements akin to transformer needed to add all the pieces to the puzzle. And it wouldn't surprise me in the least if along the way the experience of solving a few of those mysteries, teaches us that the mind model really has room for 50 pieces. Will be a fun ride. :-)
How much energy are you using? A brain uses 30 Watts, comparable to a laptop , how will your machine, all its inputs, processes, and outputs compare to the way we utilise energy across the system?
If you set a robot arm for example to 0 then program it to extend from 0 to 10... Meaning from rest to extended. All that is needed is to calculate it's fully mobile dimensions ... Once that's done you can easily get it to repeat any movement with in that realm of calculations. It's memory will retain that motion aka action from.tjst point on... Into infinity.
An important difference between Sanctuary and Tesla is that Tesla has the ability to scale early even at risk of having to redesign, due to their size and the risk appetite of their CEO. This is likely to allow tesla to gain a significant lead
Frankenstein is a profound read.
I saw the v12 beta of the Tesla automatic driving system... it is using a brand new end-to-end neural network which is doing all the driving... If the Tesla bot will use this tech, the sky will be the limit... or not 😮
Well anyone who is an athlete knows your body goes where your head is looking so when you pivot or move it’s usually your eyes that move in the direction you want to go. Your shoulders square and your lower body follows. This is true for skiers but in other sports as well. So if the robot is to be human like wouldn’t they need they degree of motion in their head as the first signal to move the body a certain direction.
WOW, loved this interview, have you ever thought of doing this with Elon and people like this on your channel? Probably not possible but just a thought.
Yesterday I was taking some pills. I closed up my hand. I realised I could still count the pills. I wondered how long it would be before robots would be able to do that. You are right that the human hand is a subtle and remarkable body part.