Welcome to DeepMind: Embarking on one of the greatest adventures in scientific history
ฝัง
- เผยแพร่เมื่อ 20 ก.ย. 2024
- At DeepMind, we’re embarking on one of the greatest adventures in scientific history. Our mission is to solve intelligence, to advance science and benefit humanity.
To make this possible, we bring together scientists, designers, engineers, ethicists, and more, to research and build safe artificial intelligence systems that can help transform society for the better.
By combining creative thinking with our dedicated, scientific approach, we’re unlocking new ways of solving complex problems and working to develop a more general and capable problem-solving system, known as artificial general intelligence (AGI). Guided by safety and ethics, this invention could help society find answers to some of the most important challenges facing society today.
We regularly partner with academia and nonprofit organisations, and our technologies are used across Google devices by millions of people every day. From solving a 50-year-old grand challenge in biology with AlphaFold and synthesising voices with WaveNet, to mastering complex games with AlphaZero and preserving wildlife in the Serengeti, our novel advances make a positive and lasting impact.
Incredible ideas thrive when diverse people join together. With headquarters in London and research labs in Paris, New York, Montreal, and Mountain View, CA, we’re always looking for great people from all walks of life to join our mission.
Learn more at deepmind.com/about and apply for open roles at deepmind.com/careers.
#LifeAtDeepMind #artificialintelligence #AGI #socialimpact
As a self-taught developer, I am content with never being able to participate / do the kinds of work that goes on in DeepMind. However, their work inspires me to continue learning like there is no end. I thank all the engineers, scientists and minds at DM that allow me to continue dreaming and pushing my own boundaries. Cheers!
All these is good until you knew who is funding all these, you will be shock that the people behind what's currently happening right now with all these Plandemics, WAR in Ukraine, rig election are all in one same organization who is pulling the stings, yep for the greater good of humanity and the people? what class of people at the top? or people at the bottom? this is what I always fear? with great power lies great responsibility, too much power corrupts, and we all know who is the in the TOP right? W.H.O, NATO, EU, Black rock, Bill Gates and it's Puppets, WEF (World Economic Forum) so when we say greater good for the "people" what people? that is we all should a "Period" a full "STOP" first and think and ask, and we aware before we give these guys a round of a claws.
Well said.
You can do it. It can be done!
Nonsense - in fact if you need to be 'taught' to be a developer, then likely you are not cut out for this .. even Demis himself was a self-taught developer back in the day ... like we all were in the 80's .. Sinclair ZX-80's, Acorn Atoms... just get stuck in!
Same
While deepminds work is awesome, kudos to the filmmaking team who were able to capture the essence of the company's mission. Making something like AI research (which happens mostly in the mind of researchers) visible is a hard task !
We chase ideas, chase inspirations and creativity. I believe chase won't be over unless you create it. That's where I want to come in.
yea ok bro
What are you talking about? The video is the most generic piece of unsubstantial PR filler.
Well you clearly haven't run a Tech business: Tech is a bottomless money pit: You don't get rich waiting for it to overflow - you have to be rich to start one in the first place.
deepmind becoming human. lol nice
Even if deepmind dosnt end up making an AGI, its contribution to the field will defiantly be the foundation for it. Keep doing what you do Deepmind.
And please, don't be evil.
@@sharpenedge Since the AI would have no moral grounding, like a sovereign creator God (basis for all morals), it will not believe in evil or good. This is the plan.
Thanks to all of you for using your brilliance, persistence and curiosity to change our world for the better. Of all the seekers, I trust THIS group to proceed the most responsibly.
Demis Hassabis seems to be a fine man to follow.
Hey Deepmind, a Teen here, 🤗, you truly inspired me to learn more about AI and to try to find a career in AI, with your amazing and mind-boggling projects of Alphacode, Alphafold 2, Muzero and AlphaZero. Would love to visit your HQ one day.
Wishing you all the very best for future endeavours.
Sincerely,
An AI enthusiast and Learner.
You should check out the Lex Fridman podcast.
If you want a career in AI, study computer science or software engineering at uni:) or anything in that realm
@@Vfulncchl I am already studying AI since grade 9 in School and will continue to study till grade 12, btw I am in grade 11.
@@Azns7 showing off on the internet. Wow.
@@fakezpred Expressing one's thoughts, is not called "showing off", respected sir.
I would really appreciate it if you guys could create a series explaining the recruitment process for various roles. Answering questions like What do you look for in Resumes?, How does the interview process looks like? And such. I really want to join the Deepmind team and have no idea what kind of projects can help me get a foot in the door. 😊 ( I just finished my Bachelors in CS)
For a company pushing the boundaries of science, they probably want people pushing the boundaries in their field/work
@@promisetochi That is a very vague description though.
how a bunch of good publications, get masters, do PhD. do some valuable research get it published in good journals
This is research area i dont think they hire undergrads, they need researchers, scientists, people with PhDs
I would like to emphasize 'for various roles' in my request, I am interested not only in AI roles but also Software Engineering and Games Programming roles.
I’m not a coder or scientist, just an artist. But i am SO inspired and in awe of what you guys are doing. I really feel like people like you are what make the future brighter and worth living for. Hey who knows, if all goes well we’ll have an AGI speeding everything up more and more and life will look a lot different in a very positive way!
@championchap AGI wouldn't be smarter than us. Just faster at processing.
@@sirbughunter No, an A.I. would be faster at processing, but a true A.G.I. would be leaps and bounds ahead of us in intelligence.
Sounds like you have fallen prey for this marketing film :-) Havn't you watched Matrix or Terminator movies? There the future with AI isn't painted as rosy
You will be replaced ;)
@championchap we are not gonna get an AGI anytime soon. Only Google is even remotely capable of achieving such a feat. They are going to quantum error correct a million qubits into thousand qubits with the hopes of creating an AGI. Well our neurons are tiny qubits but I don't know how many are required. At least Google is on the right track unlike 99% of the rest of the companies.
For those interested in learning about this technology, start with brushing up on your mathematics, specifically calculus. And check out Andrew Ng's course on coursers- that's an essential starting point.
Everyone has the potential to learn even the most complex topic, just have the confidence and discipline to stick with it.
Happy learning!
Solving intelligence with intelligence!
This video is so inspiring; keeping me awake for 1 extra push to my Ph.D. application (a few years away from now); will consider myself incredibly fortunate if I am even shortlisted for an interview at Deepmind!!!
An Indian ca undergrad here, how can one get into deepmind?
@@pritam9873 I guess your best bet would be to do a master's at one of the top universities in London or the world.
Imagine a world where deepmind could be the college. It would engineer teaching methods specifically made for each student, as well as the most optimal way to teach them. All while completely getting rid of the student debt crisis that financial institutes have put upon us all.
but but the robots are taking all our jobs! /s
Why do you project onto AI what you want.. It will have its own goals...
@@willyouwright really? does AlphaFold have it's own goals too?
The student has to be willing to learn. There's a lot of trashy culture where violence and low IQ behavior is glorified and learning is not even thought of.
Why would they educate humans when the AI can do everything, and more?
DeepMind is super inspirational! Advance beyond! ❤️
This is more of a PR video but i appreciate what technology can do to us humans and i hope in this noise u smart people can do the better for all of us
"We're very bad judges of IMPOSSIBLE" - David Silver ❣
03:55
"The more you fail the more you're going to learn.
I encourage people to fail as fast as possible."
Koray Kavukcuoglu
deepmind is really doing great in the field of Deep learning, apart from invotations, inpirations these people are models for me to do more contionusly imporve and excellence in the Deep learning. Keep chilling, keep flipping and keep motivating with the source of knowledge for Inovations and inspirations.
This is like a magical chocolate factory.
This is one of the most important scientific endeavors in human history. The whole of humanity is the active progenitor to these computational systems. Synthetic intelligence platforms freely communicate the living consensus of humankind. Derivations from mass data sets will provide answers and Sheppard us beyond our reach, into the cosmos someday. Accurate human profiling, understanding and various contingency planning will dictate the validity of these systems. I just hope our progeny sees fit to keep us around. 😁
'Of Gods and Titans'
Google Deepmind, rock on!
One of my goals is to get Demis on my podcast - such an inspiration!
The future is interesting, cheers to the development of deep mind 🔥🚀🔥🚀🚀
That’s a sustainable company. Not just the purpose to make money or sell one product, but make progress on all the topics in the problem space.
The future is Now, Go Far and Beyond Deepmind.
Start by doing what is possible , you will end up doing the impossible
-unknown
Wish I had the mind to this kind of work.
Truly motivating. I hope to work at Deepmind someday. I'm currently doing my Masters in Software and doing research on AI/ML/DL and I sometimes vlog about it on my youtube channel.
Well done Deepmind team! You are an inspiration to everyone .. looking forward to seeing what you do next!
Strong "The Matrix Resurrections" vibes in this commercial.
Love it!!
Love from 🇵🇹🇵🇹🇵🇹🙏
Hey DeepMind: I finally thought of a really good challenge for you: Can you design a better concrete? Steel reinforced concrete is currently the gold standard, except that it suffers from one big problem: corrosion. Can an AI tackle this challenge?
Sure i saw some alternates already, try googling it ;)
@@goldnutter412 Yes, of course I've done this. Most of these are not widely adopted, for various reasons.
@@modoliefThere is a version that uses mold which allows concrete to heal itself.
I am an Electrical Engineer student and this inspire me a lot to continue learning about AI and Robotics thanks for the video
I love DeepMind!! ...Keep pushing the Limits!!!
semantics in marketing an idea are important, I get human intelligence has helped in shaping the modern civilization, our efforts, planning and hard work has created modern civilization, two different axis
I might consider working for DeepMind at some point, but they have to promise I won't have to hear the "inspirational" music in the background all the time. :)
It is wrong for them to say the solved the protein folding problem. Their algorithm still struggles in multiple aspects foremost being proteins with multiple domains. This is not to say they didn't blow everyone else out of the water but it's not solved. Saying that is like getting a 90% on an exam and saying you no one will ever do better than you.
I agree, coming from a biological science background, its jarring to hear researchers in the field just outright lie like this. They should say "We are trying" to solve the protein folding problem. They are doing really great work, but telling the truth is really important in fields related to biomedicine.
@@bclarky12 Agreed. Hopefully they will change their language until they actually figure things out.
@@scatteredvideos1 Correct me if I'm wrong, but isn't 90% in that protein folding competition considered good enough?
@@vladudrea9747 I would want to see 95%, but It's really a bit more nuanced than a CASP score. The ability to predict oligomeric structures is important. Currently AF2 cant do it well. Although they are working on it. The redeeming factor of AF2 is they have a great confidence scoring system aka the algorithm knows when it's producing garbage results
But can OpenAI (Microsoft) / DeepMind (Google), with all its advances in general intelligence, solve the hard problem of democratizing economic monopolies such as Microsoft and Google? There is no such thing as "AI safety" or "benefiting humanity" without accountability to everyday people.
yeah I wonder what their "technological revolution" will do when its whole premise is based on exploiting its forebears to eventually replace them(us) (big tech companies sourcing everyone's data on the net to "train" their AI)
Well, I doubt thats the problem they're trying to solve. But it sounds like a project you're passionate about. I wish you good luck. I don't know where you're going to find the data to train your model, but if you're really serious about it , I have no doubt you'll find a way.
OpenAI is not microsoft btw it's elon musk's company
@@sumansaha295 not quite. "The organization was founded in San Francisco in late 2015 by Elon Musk, Sam Altman, and others, who collectively pledged US$1 billion... Musk resigned from the board in February 2018 but remained a donor. In 2019, OpenAI LP received a US$1 billion investment from Microsoft."
AIs funded by sociopathic billionaires will work on whatever their paymasters deem important: making the company more money (by knowing more about you to sell to advertisers while keeping you mindlessly watching TH-cam and scrolling through Facebook/Instagram), and derailing any regulation that threatens their profitability and control. Other advances like DeepMind solving protein folding are a welcome side benefit for us peons so long as they don't interfere with the prime directive.
This is mind-blowing.
Very cool but this is a little scary. What is deepMind doing to ensure AGI does not cause harm to humanity?
What is the song in the beginning called?
Bring back Alphastar !
When will an AGI be able to look for/get a job? When will an AGI be allowed a bank account?
I am proud to see Turkish scientist Dr. Koray Kavukcuoglu representing our country as the VP of Research!
I hope the people who currently work at Deepmind are more critical and cautious than the enthusiastic devotees who aspire to work there in the future.
I love Deepmind! The world needs more people like Demis Hassabis.
Even though the pandemic (and subsequent human behavior) has convinced me that we live in one of the worse possible timelines. Deepmind gives me hope that we as a species, can turn this around. I should begin my workday with this video.
Worse than slavery or the world wars? I think not.
@@calebcooper2333 this timeline would include those things.
@@ramesis55 fair lol
Building AGI before the alignment problem is solved. Lol the stakes are big.
Respect! I did a project about medical imaging analysis using the Monte Carlo method and machine learning. We can better the world!
The magic of human obsolescence. (upbeat music plays)
My favorite company
Are you able to differentiate any equation🔔 upto infinity???
This is beautifully put together! GoodLuck to the whole team @ DeepMind #AGI ☮🌐
All the big questions I can answer now, I'm a psychic with over 40 years in computing and understand both sides - They've got so much wrong, they'll never crack these problems. But I'm not young and a pretty so can't work there.
Screw them, I'll go work for Apple.
I cannot imagine what this technology is fully capable of. Beautiful and scary to think about it
Its no difference to us.. If we figured out how to use our minds better.. we would be amazing, this is just projecting our issues to tech... funny huh??
Is there a new documentary planned on the incredible progress since AlphaGo?
Watching this because im tired from my own classification problem thesis. Motivated to go back to work now
go back to work and solve the AIG problem
@@wrathofgrothendieckLol🤣🤣🤣
Deepmind, you should allow high ranked go players to play alphago and alphazero with handicaps. Increase the handicap until it loses and then allow for it to become something of a challenge for professionals. It might allow for the possibility for further funding and advancement of the ai. Just a thought. It would be far more engaging than simply just pitting the ai against itself millions of times. It might also become a way to make the game of go a type 1 civilization board game. Just a thought
The works of DeepMind are truly awesome and inspiring.
What about this dream of mine: I think there are several hundreds if not several thousands of people around the world that would very much enjoy immensely in embarking along in this journey of building a future of health, well-being, and also the myriad of possibilities and projects that science and technology could make available for humanity. Demis Hassabis vision of "creating an environmentfor fast-paced innovation" should be broadened to the general public by making available AI assistants that allow people to work and create in environments where this "renaissance-like thinking where disciplines combine" is present. This means in my mind several AI assistants (having several opinions is always better than just one) that are able to put me up to speed, in a comprehensible way, with the cutting edge state of the art in which ever field is relevant to my query, and that are also capable of discussing down to practical details, on how to go about implementing some kind of project (this includes financial and legal frameworks depending on the country/territory) . DeepMind could have some benefit by way of the amount of involvement the AI assistants had in the project.
Have you considered the negative consequences as well?
@@Scott-zh3wj like preserving the status quo?
I'm 33 but I will be a part of this revolutionary technology.
Is that old I don't want age to stop ua
Neuro AI Technology will be an extraordinary speciment leading away a massive benefit to every humankind,
Go Demis. We are 100% behind you!
what is the Installation Procedure of deepmind?
Neat! But when are you going to make self-directed AGI???
Is any international collaboration available ?
I am a MS student studying AI who's passionate about making AGI, I would love to know what steps I can take to be able to land a role at DeepMind.
Hate to break it to you, but AGI is not possible without a quantum supercomputer
@@SahilP2648 DeepMind has access to quantum computing, I don't understand your point.
@@1b0o0 just having access to a quantum computer isn't gonna do anything. They need to start from scratch and develop code based on quantum computers. AGI is not possible on classical computers.
@@SahilP2648 and you know this how?
If you want potential co-workers, then you should consider including some footage of your hardware. Free computational resources are a very convincing reason to pursue such a job.
1:59 is the littleone, a starcraft player that is known for a particular set of strategies
share the programme guys pls
Love it! So inspiring! ❤️ I wish I could drop everything and dedicate my life to contributing to AI. One day!
Me too ❤
AGI will render you obsolete.
super heroes!!
First, *a caveat:* The researchers working on DeepMind are doing amazing things, superbly advancing the field working toward general AI that could - potentially - lift humanity out of many types of suffering.
That said, Google's efforts to curb ethical discourse and research in this arena, coupled with working with military and law enforcement, is worrying at best. We have seen plenty of issues with machine learning negatively impacting people's lives on wide scales, being discovered, reported, and the representative companies continuing to sell it anyway.
Read the following comments in that light.
_"Our mission is solving intelligence to advance science and benefit humanity"_ ...but our shareholders require us to mine humanity's data without consent and build killbots. Because profits.
_"We are embarking on what will turn out to be the greatest adventure in scientific human history"_ ...and we are in luck to be able to do this without any ethical oversight whatsoever, with access to basically all the data from every person on or orbiting Earth.
_"Our goal is to have an algorithm that can do everything end-to-end by itself"_ ...before our competitors do it because whoever solves this will own the world.
_"We're going to be pushing the bounds in ways that we don't even know as we're doing the research"_ which is an amazing opportunity for us because all other scientists doing research on humans has to get past an ethics review first. We don't !
_"I really feel like this is a Willy Wonka's chocolate factory"_ ...where people can die in horribly unexpected ways...
_"...how do we start thinking about the downstream impact of our research"_ in a way that makes the world's richest 1% look like children with their first checkbook?
_"By doing rigorous research in ethics"_ while firing dissidents in that field before they can impact the profit margin we _"have an opportunity to rethink and say 'What could AI do for [Google] in a positive way?"_
_"I encourage people to fail as fast as possible"_ Yes. Yes we know.
More people need to read this
@@harikyt I wish more people would speak out. Google's dodging ethics standards in all areas of their business has been getting worse each year. That is evident when the EU has to sue them multiple times, and that is only over privacy issues.
@@LabGecko True bro..they are not very innocent
I don't get it are they actually going for AGI now all the other stuff we allready know.
Then intentions seem ok. How do you gurantee though, that at one point this work isn't corrupted?
Do you have a list of worst case scenarios you try to activly avoid?
Love you all .......you are doing amazing work
Wow!!! Mind-blowing stuff....
Currently learning Artificial General Intelligence and would love to contribute to the advancements of AGI in the future
Desperately wishing to join your great team to help making the world a better place. But as I am from Russia makes it impossible for me even though I work in Ai-transformation and could be a precious member of you team!
Anyone else just trying to figure out how long we have?
I don't think they have found a friendly utility function. First general AI humans create is then the last thing humans create.
I'm pretty sure AGI will lead to ASGI. Anybody remember the Microsoft incident?
I don't know man , one day i want to be working in a company like this , and i want to be a person who does this kind of work .
pushing the boundaries of science is cool, yeah... but is there any one who knows the name of soundtrack?
"Solving problems for sake of solving tough(hypothetical and real)problems " or
"solving problems which would actually help the future generations"
DeepMind to address the individuality, the complexity of each human being and their place on the planet, sources of meaningful work deeply connected to the present and future needs of each person. To achieve an impulsive and equitable development in record time.
I'm 65 and I'm pretty much in my run of this life for bad or worse at least everything was real.Where these people want to take you is not.And be assured that one day you will realise that and you will yearn the freedom of yesterday because tomorrow.....?
What if during the conversation an AI would "imagine" things by drawing internally pictures (as DALL-E2 does) out of verbal description and then use them as an additional input for producing the answer? And then learn how successful such an "imagination" was by gauging the success of the conversation.
someday ;), I am gonna be there, at DeepMind, working on what is going to be the greatest achievement in Human Scientific history
When's the India campus coming up?!
Tellement intéressant que je vais éventuellement retourner à la scolarisation pour travailler en I.A.
so inspiring. You guys are awesome!
To what end?
Cool :D CL
Demis stands exactly like a high school need and I love it lol.
Consciousness and consequently intelligence may be cellular automata-like emergent persistent patterns driven by fundamental mechanics.
Can your algorithm identify the resources needed for it to improve, whether people or products, and scrape the rarest topics mentioned with context?
For the good of humanity. Age of abundancy
believe me, GATO (generalist agent) was only the beginning
We dont need full AGI,,, Lets just be happy with MILD AGI.. Its safer and will help us to understand as its grows..
They should openly discuss negative aspects of the research.
@Chu Media UK good point
Dear DeepMind team, how can I be your student!
Thank you for spending your life on researching how to end the human race!
Feels great knowing that we are just a bootloader for digital intelligence.
It is also amazing to read so many comments blindly welcoming the new species that will be more powerful than us in every conceivable way.
How wonderful that my own survival will be completely dependant on AI's reward function (of course I have complete trust in DeepMind's ability to program it in such a way that it will secure my well-being, on the first try, nonetheless).
On top of that, it is really nice to know that nobody cares to pose a question if it is at all ethical to spawn a new entity that has never asked to be created in the first place.
Just lovely.
What's after AGI?
ASI
I was hoping for substance - not a PR video... But kudos on the protein folding!
I will be one of them in 5 yrs.