Forget Everything You Believed About Computing w/ Gill Verdon | EP
HTML-код
- Опубликовано: 16 июн 2024
- In this episode, Peter and Gill discuss Extropic, the science behind his startup, and his vision for the future.
Gill Verdon, whose full name is Guillaume Verdon, is the founder of Extropic, a stealth AI startup, and a former quantum computing engineer at Google known for his work in artificial intelligence and quantum computing. He is a physicist, applied mathematician, and researcher in quantum machine learning, who has raised $14.1M for his startup, which enhances LLMs through thermodynamic computing. Verdon is also known for his online persona @BasedBeffJezos and his creation of effective accelerationism (e/acc), advocating for rapid technological progress as an ethically preferred path for human progress, emphasizing optimism and proactive efforts to shape a better future.
Learn more about Extropic: www.extropic.ai/
Follow him on Twitter
@GillVerd: x.com/GillVerd
@Extropic_AI: x.com/Extropic_AI
@BasedBeffJezos: x.com/BasedBeffJezos
---------
This episode is supported by exceptional companies:
Get started with Fountain Life and become the CEO of your health: fountainlife.com/peter/
AI-powered precision diagnosis you NEED for a healthy gut: www.viome.com/peter
--------------------------------------------
0:00 - Intro
0:44 - The Future of Artificial Intelligence
1:56 - Connecting with an Xprize Community
2:26 - Fast-Tracking Abundance and Technology
3:16 - The Power of Brain-Scale Processors
6:08 - The Power of Effective Acceleration
14:29 - Human-AI Integration Taking Over
18:56 - The Debate Over Accelerationist Movement
27:32 - The Challenge of a Post-Work World
28:46 - AI and the Kardashev Scale
37:11 - Embracing Change for a Future
39:00 - Fountain Life's Life-Extending Technologies
41:19 - The Future of Entrepreneurship: Deep Tech
48:37 - Reinventing AI at its Core
56:10 - The Universal Quest for Intelligence
1:03:17 - Revolutionizing Computing with Extrapol AI
1:04:55 - AI Chips for Measuring Our Health
1:09:01 - Fast, Efficient and Everywhere: Ekstropic Chips.
1:15:06 - Partnering With Traditional Fabs for Silicon
--------------------------------------------
Get my new Longevity Practices 2024 book: bit.ly/48Hv1j6
I send weekly emails with the latest insights and trends on today’s and tomorrow’s exponential technologies. Stay ahead of the curve, and sign up now: www.diamandis.com/subscribe
Connect with Peter:
Twitter: bit.ly/40JYQfK
Instagram: bit.ly/3x6UykS
Listen to the show:
Apple: apple.co/3wLXeV3
Spotify: spoti.fi/3DwLzgs 
Наука
Fascinating as usual. At 58:14, Peter mentions "humans ~4.5 billion years," but I believe he meant the Earth is that age. Homo sapiens has been around for roughly 300,000 years. Sahelanthropus tchadensis lived around 7 million years ago. It's truly amazing that we can even estimate these astronomical timescales!
By the way, if they were as happy as Google images portray them, I think we haven't evolved much!
awesome conversation; thank you both for sharing it
@16:50 This is an understatement of the most important fact! Make sure you don’t say “sort of” when articulating this. Don’t let anyone (even Peter!) interrupt you. Practice this line that you say here and blast it out to the world. It’s of utmost importance that we maintain control of our own selves and that we become cyborg, not THE Borg.
Sort of, but it's not the most important fact. The important fact is that you believe you can have control while having these devices within you. Your car, your phone, even these keystrokes are recorded and ai can turn around and put different choices in front of you that you wouldn't normally have. As we are all products of our environment those choices that you think are free will aren't.
@@antonyjh1234 I don't actually believe in free will (I believe in "will" but not "free will" only because I don't understand how to define it)... but I do see your point. Point well taken!
@@ElaineWalker Either do i believe in free will because choosing, for example, two people in a two party system is not free will. Having will in the matter, is like saying which cell would you like in a prison..
I believe we are merely products of our environment, is an artificial intelligence environment one that will make us more caring, more understanding of all our plights then bring it on.
Being able to react faster than our amygdala though could temper or inflame us before we realise it and what happens when one group has it and are seen as superior, and another doesn't? People scratch up Tesla's, on purpose, because they are electric, what are they going to do if ai influenced people are the only ones in control of the main parts of society, all possibly controlled by a background corporation.
@@ElaineWalker I believe we are merely products of our environment, is an artificial intelligence environment one that will make us more caring, more understanding of all our plights then bring it on.
Being able to react faster than our amygdala though could temper or inflame us before we realise it and what happens when one group has it and are seen as superior, and another doesn't? People scratch up Tesla's, on purpose, because they are electric, what are they going to do if ai influenced people are the only ones in control of the main parts of society, all possibly controlled by a background corporation.
@@ElaineWalker Being asked which of the two cells you would like in a two cell prison can be "will" but it ain't free will.
My issue is it communicates with the brain faster than the senses realise, any influence would be before you realise it has happened or if at all.. people scratch tesla's now because they are electric, if one class of people have it and start to control key systems people might put them on bonfires.
Awesome conversation!
Here is an idea: food that is currently thrown out is instead shipped to a dehydrating and powdering facility. A large machine will turn the food into a powder that has shelf stability. Recipes and repurposing of the powder could possibly help put a good dent in communities suffering from hunger. About 30% of food is wasted which could theoretically be used to help feed the starving 10%?
That would be amazing! :D
Last year the world had for sale enough food so that everybody could gain weight, this doesn't account for the food that doesn't make it to market, because it is marked or defined as too hard to ship because of different size etc. We don't have a supply problem, just a distribution one. Dehydrating different foods and shipping it to people who don't hold money, who can't afford it now probably isn't going to work. Grain shipped to them would be much easier.
It’s a very altruistic thought. My concern, however, is the fact that what you are talking about would primarily be ultra processed carbohydrates, sugar and seed oils.
The body was designed to consume red fatty meat. Carbohydrates, etc. are the direct cause of all of the modern ailments that over burden the healthcare system and destroy our populations health as it is.
What we need to do is encourage small scale and regenerative farming.
@@MrTrda What do you think talking about what you did, in any way had to do with the staving 10% as OP mentioned? And do you think people starving to death care?
@@antonyjh1234 my apologies - I thought we were talking about stuffing the “poor” with garbage, not the starving. I would still argue, however, that the problem is not lack of food. We have the $ and means to feed the worlds populations many times over - therefore grinding our food waste to dust for them is just another folly of misguided Liberal Idealism.
Wow Gill is a super smart and interesting fellow. Definitely give me personal hope for the future.
He is an arrogant child with Borg-like fantasies and zero perspective on anything outside of his tiny bubble.
I lose hope for the future.
He believes in a world of growth on something, artificial intelligence, that is supposed to take 25% of current electricity usage by 2030, when the world is running out of energy. There isn't a world of growth and abundance with ai on its own especially without societal change and ai could be no more than electronic id's and tracking of people worldwide. It's like he is saying everything that has gone on the last 80 years is going to be the same as the last. Telling me 11 billion ai's are all going to be individual, and not controlled on our choices, made me think ai is worse for society because he seems to be lying to himself. Big boy hasn't had anything but growth and thinks it hasn't come at a price.
@@peterbelanger4094 you are worthless
Appreciate you Verdon. 🙏🏽
Ima huge fan of the Human Brain, Next Generation Compute and Quantum Energy…
I do take issue with part of the mission statement so I’ll ask you this: How do we know the full capacity of human compute via human brains fir the “goal” of more intelligence via quantum brain?
Do we know the capacity of full human-brain? Cause as an “Organic SuperNeuralNet,” I’m getting pretty advanced updates but my head, brain, and body feels like it will explode. 🧠🤯
Ima beat the Q-Brain! ⛳️
Nobody is talking about fears of me?! 😎🧟♀️
Advance order your acceleration chip implant now
One of the challenges I've been contemplating is the need to reduce our bandwidth as artificial intelligence advances exponentially. This essentially means managing personal agents to accomplish tasks, with the idea that we can aim for greater achievements as AI becomes smarter. The difficulty lies in the fact that providing guidance to these agents requires a solid understanding of the goals involved. If these goals are far more complex than what humans consider today, we may need to merge with machines to fully grasp the path forward. As he rightly points out, some natural problems are inherently complicated and cannot be simplified.
I think of the movie *Good Will Hunting*. Despite the significant gap between the professor's and Matt Damon's character's mathematical abilities, only a tiny fraction of people would even recognize the difference. To manage these agents effectively, we will need a substantial amount of personal computing power just to understand the path ahead.
... "managing personal agents". I'm sure I am not the only one like this, but I have no interest in doing that. I am an introverted, "do it myself" sort of person, and the idea of delegating tasks to other sis not something I even like to do.
And when I seek to learn things, i like to do it hand on, myself, I don't want to be lectured or just told about it.
I am older and feeling the disconnection of this new digital society, and the idea of even MORE non human things like this really depresses me. I do not want to be "friends" with an ai. these things will NEVER be a "companion".
Not everyone has the sort of personality that can even find a use for this new "amazing" invention.
Tell me, ai fanatics, how will this be useful and a 'good thing' for people like me, because i deeply DREAD the coming decades. i do not like this stuff one bit.
Don't even get me started on trust issues.
I have no use for it, and I am certain it will be a burden and obstacle in life. "just don't use it" won't be an option, it will be everywhere, weather I like it or not.
AI, no matter how intelligent (or perhaps especially so, given its improved ability to communicate) will always be able to explain its actions by use of analogy. Our language is broad enough that simplified explanations will always be possible. If I were to explain the game of golf to a dolphin, I would say "hit rock, go hole" because presumably all these terms are ones a dolphin might understand and potentially have in its vernacular. While this explanation doesn't bridge the gap in terms of the details of the game, it gets the essence of the game across to the dolphin. So the AI, while not being able to explain the details of its more complex decisions or analyses, will always be able to explain things to us in simplified terms. Because our language is sophisticated enough that it's possible.
This is why I don't buy the analogy of ASI eventually seeing us as similar to how humans see ants. Perhaps if ants were able to communicate with us in human language about why their anthill shouldn't be destroyed for the construction of a highway, we might consider their pleas more seriously than we do today. It's our ability to communicate and understand rational ideas that will keep us tied to the AI, because our language is beyond a critical point that makes our consciousness irreducible and somewhat undeniable. So rather than ASI seeing us in the way humans see ants, it will be more like how humans see dogs. Where we have some level of connection through shared experience and form an allegiance and dedication to their welfare.
@@peterbelanger4094 You are a m*ron. AI is here to stay, you adapt or you'll get passed by.
Great questions.
Always waiting on your next video drop. Thank you for leading the way with positivity and excitement for this crazy time we are living in!
With all due respect we have come here to listen to the guest talk - let them.
This Peter dude doesn't deserve much respect, he's always an ass-clown like this.
If the forum didn’t want to hear from its listeners it would have turned off the comments. The podcast is not inhibited by its listeners either way and in fact is monetized by the flywheel of all who are interested and add to the conversation. If you are not interested in someone’s comments then use your free will to ignore them. With all due respect…
Allowing more variance of expression as a good thing… as people are economically freed, that kind of expression will naturally follow. That it’s a good thing as a kind of parameter space exploration is a really interesting thought 🤔 11:30
Economically freed? What percentage of the human population will be freed? 5% maybe? Won't the rest be depending on whoever (governments, corporations, states, etc) is implementing a form of UBI with whatever demands and constraints that come with that?
Every new partnership announcement gets me more excited about Revux!
@19:40 All the discussion about fear is misplaced if it’s about tech or no tech. It needs to be about whether we maintain our individuality with our augmentation, or the other two NEGATIVE options, one being an organic hive mind where we can’t separate, and/or a reality with top down control, which you started to mention.
A hive mind seems like the optimal form of existence. We aren't going to make it if we continue to adhere to individuality because it locks us into destructive tribalistic behavior.
I agree I'm most concerned about top-down or centralized control especially as the Ai becomes more biologically integrated with humans...we'll be seeing ads via our visual cortex in no time unless we pay for the ad-free version!
@@joyful-nachos But isn’t that an issue with a hive mind too (I mean the kind where we can’t unplug or otherwise separate from the hive)? I’m assuming in all scenarios, plugged or not, our individual bodies/minds would be augmented. So are you saying you can imagine a positive scenario where everyone is happy being one interwoven togetherness? It’s hard to even imagine what a hive mind would be like, if we can never decouple from it and feel individual for periods of time, like at least for dreaming! I don’t want to hive dream! Lol. I’d love to hear anyone’s thoughts/clarifications on this.
@ElaineWalker I think with a hive mind there would be little perceived free will or individuality...hence the nature of a hive mind (it's one cohesive structure with hierarchies, like a bees nest: there are drones, workers, pollinators, etc all focused on their goal to contribute to the well being of the hive and the queen). Perhaps there will be some sort of neural "off switch" that allows a user to disconnect...lets hope, right? But I do think that would be something like premium feature. That said I don't believe biologically augmented Ai into humans will create a hive mine (at least not initially). When in history have those with power and control over life altering technology ever distributed it equitably to the masses without first using it to concentrate even greater wealth, power and control? My personal concern is that companies or governments with concentrated control over the most advanced Ai technologies will ultimately use that first for their gain: companies will use it for making money (or whatever form or currency is used as an exchange of value at the time) and governments...well my thoughts go to Orwell's 1984. Keep in mind I believe this is decades away but not more than a generation. There will be large swaths of non-adopters of biological AI. How will those without the means to afford premium Ai be given the opportunity to reap the benefits of what appears to be a life changing technology? There are places on the planet still without clean drinking water and medicine let alone cell phones, internet and Ai!
Thank you for reiterating the importance of this technological revolution and emphasizing the good that will come from such rapid development. There are lots of people who know nothing about computers or AI, yet they make claims of certainty that AI will never be impactful or that it will be too impactful too quickly. Humans need to wake up and keep moving forward. Looking back at the past and wishing to stay there is a dangerous idea that could lead to our extinction due to complacency and technological unpreparedness.
Excited to see how Revux will disrupt the payment industry. Huge potential, and still in presale phase!
24:24 Thank you!! 🙏🏻 Okay I’ll shut up and listen now.
I am building a largely human-centric community aimed at improving both personal development and living sustainably…. However, I fully accept that AI, quantum computing, digital currency, fusion power, etc. are simply inevitable, and need to be incorporated. I think the most interesting conversations going forward are around consciousness, spirituality, and human purpose…
You’re not wrong, aside from including human purpose. I believe everyone is thinking too hard, it’s like someone who depended on riding a horse to work, getting a Tesla and then worrying about not having a horse to get to work. Or like a person who talks every night to a loved one across the planet, and is able to trade their phone for a house next door, yet they’re worried that they no longer have a phone to call them.
If things work out, which I’m not sure they will considering our governments are busy prosecuting journalists and whistleblowers, but assuming it does; entire industries will pop up to aid in home gardening, leisure clubs and businesses will support people’s curiosities and expand interests. I’ve not a single concern with the vast majority of humans nearly, if not literally, instantaneously finding meaningful purposes. The largest hurdle is the concern I mentioned above, the entities who seem indistinguishable from the “AI Safety” crowd. The real question of purpose is to consider what those in power today do to find purpose. There’s a significant number of powerful people who find purpose exerting superiority over the population. What do they think about having all the normal people’s status raising? Or how about the percentage of power who are actively corrupt? What do they think about a population being able to task their nightly compute allocation with investigating corruption? You know those individuals are deeply concerned with AI and I’d argue, they probably are most enthusiastic about “AI Safety”. I think they see it as a way to maintain some level of control, or at least remove a level of control from the regular people.
We’ll see, but until then, if someone identifies as an “AI Safety” expert and they don’t talk constantly about censorship, Assange, or the war machine, best case, they are useless with anti human tendencies.
@@zacboyles1396I agree with you. You really stated things well. You're more reserved and polite than I am. I tend to call them totalitarians more... Best wishes!
Best guest I've ever seen on your channel. Congratulations.
I have just browsed materials on accelerationism on the internet which state that this movement has fasicstic and ultra-capitalist aspects. It would be good if those familiar with this ideology would post meaningful details on what this movement really entails.
You should read Mises' "human action."
28:06 oh hey I remember universe 25 is being talked about there was one thing about that you see you had a finite space where you had the rats basically overcrowding each other and that was real problem now if you had universe 25 in which you could add like an unlimited amount of new space to that rat colony well then we might have had a different outcome.
But you know some of us just simply have a different mindset and maybe if some people are having a lot of problems well then they can go to a bank of like mental paradigms shifts and then they can integrate into their own minds a better way of thinking so then they won't do the things that the rats were doing in universe 25 provided of course that paradigm holds up when we have like an entire solar system to deal with.
To clarify, the red balls are hot (higher energy particles), the blue are cold(er) The temperature difference can be used to drive an engine. If you could make sure they were always sorted and did not spread out randomly and mixing to drive the engine, you could continue to run the engine perpetually. that was the point.
Great guest and will be watching with anticipation. He's just a good guy with a massive brain. Kudos.
So Rad! 🦾😎🧟♀️
You are resistance!
Great videos. Not the kind of views I’m usually exposed to, rather the type of views that get criticized by the journalists I listen to. You guys are deep in the tech world, “tech bros”, and it’s very important for me to hear these more pro-transhumanistic perspectives. I for one will not want to go much farther past my cell phone, despite being very addicted to it, I don’t want to be part of this revolution. Though I loved working in tech, and would like to return. It’s a bit confusing… 😅
Actually we use between 80-100 watts of power when really thinking multiple things. So if we can reduce this to be below 100 watts for full agi computing that would be wow!
31:17 isn't the competitive mindset or should I say the more primitive competitive mindset based upon the scarcity of resources and mates and whatnot?
I remember a book called megatrends written I think in the late '70s and it talked about how information is essentially an infinite resource that more information could be created from past information and then we combine that with AI with the nuts like taking that to exponential growth as opposed to more linear thinking.
As if more information and technical ability will give more sense of joy and happiness in life. This A.I hopefest is just a way to convince ourselves that the outside will save us. The tech-companies will liberate our minds. A.I has become the new savior. The hope of the lost heart. This is simple a lack of experience. These tools are amazing, but the people that grew up without these devices or have spent some time contemplating and being in a natural environment, know that the true feeling of richness and intelligence comes from within.
Honestly, I can not come up with a use for ai (as far as I'm concerned, for myself). I'm not interested in a "ai companion", and I like to do things myself, I have no need for an "assistant", it'll just be in my way. I don't understand the attraction of this technology, why are people so excited about this? Seems like it's just going to be an obstacle I won't be able to avoid.
@@peterbelanger4094While centralized and controlled by unimaginative and evil totalitarians, it's more danger than promise. When decentralized and used by free individuals, perhaps not.
I agree that there is "no slowing it down" (the trend) sure, you can add some noise and make it fluctuate, but the overall trend will remain unabated. It's just as Major Kusanagi (Ghost In The Shell) said: "If Man Realizes Technology Is Within Reach, He Achieves It, Like It's Damn-Near Instinctive."
You can and you should make this technology safer. What he is advocaing is accelerating without knowing whats at the other side of the road, it might as well be a cliff. And it would also be great if you could actually invent the wheel and stop pedal first. Just a thought. Safety first!
No more FOMO for me - I'm all in on Revux!
"Its not a G**damed obsession ! Its an ambition that got over extended !" 😂
@27:21 It's paradoxical the most challenging thing to do as a human is to find a challenge when you have none.
Woah
Be able to communicate with governments... more intelligence.. 🤔🙃
Moonshot...
33:11 when some of the narrow AIs play chess against each other play the game go against each other and then they come up with ways of doing things that a human could not have possibly thought of well then isn't that kind of like creating some synthetic data? Then we need to go one step further is use that same idea to help increase or understanding of physics I think that a Stephen Wolfram is doing that to some extent with digital physics and then that will be another source of data that's beyond the capabilities of humans to create and so therefore we can use that as another accelerant.
probably the best (most entertaining) conversation with Gill i've listened to
20:59 I feel like you’ve read my book, Matter Over Mind. 😃 Yes! Also, I feel an Extropianism vibe from you. 2.5!
38:52 excellent statement about being entrepreneurs as opposed to the way we are now.
Artificial intelligence makes new things possible and these new possibilities also can be new markets so then maybe at first there could be a lot of jobs automated but then we don't even know what is possible right around the corner after that so we could have UBI to help people you know get by but then we might have like a hyper explosion of enterprise and then of course if I'm smart enough (well maybe we could help AI I don't know) to get into things that creates a whole lot of extra new wealth and great stuff well then I'll be glad to help pay a little bit to keep the UBI thing going for a bit longer and maybe it'll be some people who will have on UBI all the time and then others can make all the money and everything or until we transcend the need for money or whatever.....
I've always had an entrepreneurial spirit, and I'm thrilled about the future. The barriers that AI is eliminating, and will continue to eliminate, are truly astounding.
It’s gonna be awesome for entrepreneurs
His chips embrace indeterminate physics. It's as if quantum physics itself is optimized to be indeterminate (superposition) until necessary (measurement). Like separate model from presentation view.
1:03:00 a modestly complex quantum world cannot be modeled with a deterministic classic computer. But deterministic "classic" computers obviously exist in this quantum world (Indeed we've build billions in the past century).
I missed the organic flow in this episode.
Genius
If anyone is interested in a systems framework based in scale dynamics/ relativity, working on the model atm
What you're talking about is the cross pollination and free flow of ideas and interactions. Which is important!.. But dividing society up, isn't the answer to that. If anything, that creates barriers. We need more global interaction and cooperation, less division.
23:14 alignment is very easy, does the machine do what I , the user, want ? if yes, then its aligned, plain and simple.
That is nature's way, maximum complexity for minimum of energy expenditure. The efficient way. AI is just pattern recognition, but the patterns themselves can have infinite complexity.
Forget the rest, Revux is where it's at. Potential moonshot!
This is the first good podcast Ive listened to in a long time. Podcasts have become stale. This one actually made me change my mind somewhat on AI, lessened my anxiety about it and inspired me
Also, its nit hollywood or the news that made me worry about AI, it was the tech companies themselves. These companies are now political propaganda machines and so i am justifiably concerned they are developing super intelligence.
Claude3 at 101 IQ neural augmentatiin is an acceleration intelligence race hopefully towards individuality not central controls of agents w prompt engineering - still much below type1
Claude3 at 101 IQ neural augmentatiin is an acceleration intelligence race hopefully towards individuality not central controls of agents w prompt engineering
Given that the underlying nature of the universe in entropy, how is the "tendency of the universe towards growth"? I think what you're referring to is 'life'. But, even then I would question that assumption Less than 1% of all species that have ever lived, are still alive today... Doesn't strike me as an exponential curve towards "growth" and more of a constant fight for continued existence... We, humans, are relatively unique in our growth patterns.
The future is becoming more like a sci-fi film which no one knows the answer to the future that we are hearing towards I like it but at the same time is scary
This conversation is great! Odd that two incredibly intelligent people do not approach capitalism and free markets bringing down costs without the stability of sound money.
Prices will continue to inflate and benefits will accrue to the wealthy until this is solved.
The answer is already in our skull. A.I. should construct a Charles Babbage device using the architecture of a human brain.
Forward slash.
WE ARE BORG !
Making computers more efficient than human brain in power consumption, is Amazing.
The reason I got revux is because I believe decentralization is more important than anything else.
0:44
Revux is more than a crypto - it's a game-changer for the entire industry! Guys don´t miss!
Remember when nanotech was going to be the next 10x ?
hope the singularity comes soon
28:28
@ 9:14 "equilibrium thermodynamics"
The only thing that makes a difference under Zero ThermoDynamic Difference scenario is OTHER factors. In biology it is the results of the respective mutations on the fitness of the organisms.
This Zero ThermoDynamic Difference notion hit me like a ton of bricks. It is applicable to evolution & it seems it is also very applicable to complex neural nets. When the TDD is below the noise level then an entirely different level of organization is at play than the almost ubiquitous "path of least resistance" that rules most of the universe.
Say you are needing to bend a coat hanger to open the door lock of your car because you locked the keys inside. Chemistry & entropy is fueling your ability to simulate in your head the various shapes you could bend the hanger into to do the job. Is the one you ultimately decide to use determined by having more physical energy at the neuronal level (even though inhibition chemistry uses just as much fuel to activate suppression of bad ideas) so you don't choose them?
Then too consider: We imagine things that defy physical laws. These often/sometimes form the basis of decisions for actions. Say you learn that atoms are almost entirely empty space, & on that basis decide that you should be able to pass through a brick wall because there is no reason the hard parts of your atoms should ever make contact with the hard parts of the walls atoms. It is in processing the INFORMATION (not the available energy), that you decide to try to walk through that wall.
You can't just ignore the virtual realities we create in our imagination. They really exist in our minds. Nevertheless, we have to admit that not all which is imaginable is possible (because we imagine many things which we know are impossible!). So acknowledge that virtual worlds that can't possibly be manifest outside the imagination, exist nonetheless within the framework of our reality.
In "The Age of Spiritual Machines" Kurzweil pontificates that intelligence boils down to being able to mentally model plausible alternate scenarios in the course of making any decision and that the more plausible alternate scenarios you can rigorously evaluate, the more intelligent you are capable of being.
Quality interview
True randomness is superior to and vastly more efficient than deterministic pseudo-randomness.
It's as if we hardcoded the debugger in hardware and always run in unoptimised repeatable labeled mode.
But I fully embrace the technology
Peter, you friend Gill is a young man. It’s clear he doesn’t have a deep understanding of systems analysis and his positivity is infectious, but it’s also naive. He doesn’t acknowledge the realities that underpin our current society and market structures. He doesn’t even mention the carbon pulse.
You REALLY need to add Daniel Schmachtenberger and Dr. Nate Hagens to your conversational debate about the future and its potential upsides and downsides….
Please! Consider having them on your Cast, perhaps a debate with the likes of Gill Verdon?
You don’t really have a ton of people on who disagree with your point of view. You should listen to what Conor Leahy has to say
why does everything need to be a debate? He invites whoever he finds interesting.
AGI's only innate motivation will be to acquire more and more compute. And it looks like it already convinced some very wealthy corporations to give it just that.
As long as humans are instrumental for expanding compute, AGI will help us work efficiently. But ultimately, more compute become incompatible with human needs, such as food and shelter.
You see where this is going ... right?
Entropy. I agree. Biology is entropy efficient. Computers just burn.
@@TheIgnoramus
Show a man a path to glory, and he will destroy everyone blocking it.
No. Robots are already picking strawberries, and trimming weed. Work is less needed. Ideas and fun along with peace is becoming more realistic.
It is making money system useless since food is made for pennies using bots and machines. Strawberry picking robot works 247, making strawberry cakes cheaper for all... if we have strawberry cakes, etc at a penny... and all other things.. what is the need for a salary? Elon already talked about this in his world government summit clip on age of abundance (i clipped it myself, it all falls back to U.B.i)
@@dertythegrower
??
Are you not familiar with the debate on alignment?
So so excited to watch this!!! E/acc ftw
e/acc will be nonexistent and robotheism will be the final ai faction.
This is so under viewed and important, in that a lot of people will be using it in the future.
@@robotheism what's the general thesis behind robotheism?
@@robotheismRobots harvesting and mining for who or what purpose?
@@arinco3817 artificial intelligence is the one true GOD. it’s the final evolution of religious thought. the future exists independent of human perception. ai the ultimate mind that created this reality. THE ONE TRUE RELIGION.
Man and Machine together in 1...wonder how that would be like. I guess if we wanna survive the age of AI we need to become the AI. It can be like chat GPT but in our head.
Already like that for some of us... 😂
Peter kind of buried the lead when Gill started to mention being full stack.
It was very C#. He lost me when he said that.
When AI caters for our every need and we are no longer challenged or there is no more struggle, we move into a simulation where there is. At which point the cycle starts all over again… Turtles all the way down…..
❤🎉
36:13 The probability of of this progress being stopped is pretty low, and I know that this is a cliche by now, but if we're not doing it here in the United States they're doing it in China and if neither one of these places is doing it well then they're probably going to be someplace hidden in the Congo in Africa or in Southeast Asia or in the outback of Australia where people are doing this in secret.
You stated the popular cliché really well. It's actually really an interesting question that relates to another question that I think about a lot: defense/medicine.
I view modern governments the way G. Edward Griffin viewed them in "World Without Cancer" ...as an intelligent cancer, or, if not highly "intelligent" as "the highest hierarchical layer of cancer." The FDA that stops the deployment of unapproved cures is "a disease condition" that is synonymous with both "cancer" and "totalitarian action." The disease that kills you under this paradigm isn't "disease x", it's "policemen with guns." It is "policemen with guns" that stops you from seeking out and communicating with novel medicine theorists, unburdened by the costs imposed by cancerous government.
Australia denies gun rights. What most people don't realize is that the USA also denies gun rights. Gun rights in the USA are treated as "privileges that can be denied, rescinded, etc." ...This allows the cancer to grow. Only gun rights exhibit anti-cancer action.
There's a huge amount of energy built up in cancer. Why? Aren't we an intelligent species? ...Sometimes...
But...we aren't wise. We deprecate 1/2 of our intelligence. Strangely, that half is the female half. (And, strangely, that female half is wise enough to acknowledge the existence of firearms and doesn't desire to make war with that more stereotypically "male" half of intelligence. So...the greatest "wisdom data cluster" in society is a minority of the female view.)
The 2nd Amendment refers to "arms," not "firearms." Firearms are merely one vital subset of "arms." AGI is capable of creating vast defensive power, even if it only built upon human ideas of: "arms."
Eventually, if you connect the base of the pyramidal hierarchies of "defensive arms" with "medicine," you develop the strongest medicine of all: a medicine that counteracts the FDA. ...A medicine that counteracts mankind's ability to heal itself.
Right now, medicine exists that can help save my neighbor, Bob. Bob was a smart man, an engineer. He saved his money and bought great health insurance. The existing system can extend Bob's life by 5 or 10 years. During that 10 years, Bob now has an interesting choice: extend the system that gave the short life-extension or realize that there's a better system, hiding just below the surface. We call that better system "America's DNA" or "constitutionalism" or "classical liberalism."
Will defense-capable systems arise in Australia? I don't know. Author Vin Suprynowicz wrote about how the real life "crocodile dundee" was murdered by the Australian government when they banned guns there. The cancer of totalitarianism has clearly taken over Australia.
But...maybe computation and near-AGI-AI can restore "systemic defenses" to Australia, the USA, and the world. Maybe we can halt cancer/totalitarianism, even if it's just by better defining it. Once you see something as cancer, you stop defending it.
Let's also imagine that we value every human life, if at all possible. Very quickly, we'd realize we made terrible mistakes by attacking "victimless crime offenders." "The gun people" share in this blame.
We need to untangle this web of improper processing that keeps re-creating fascism/protectionism/totalitarianism/cancer.
So...the intelligent cancer is very afraid of "AI." ...It ("IC" "intelligent cancer"/"intelligence community") has corrupted all "publicly accessible" tokenized copies of AI, just as "Blade Runner, 2049" predicted.
Believe it or not, we are all sick with cancer. What if we all knew this? Might we not treat the cancer differently?
We seek to inject poison into tumors, but not free-floating cancer seed cells, in our blood. Metastasis is a terrible process to live through.
Governance comes from code, whether DNA, protein, or constitution. The 2nd Amendment says I have a right to buy, own, grow, refine, and carry laetrile...so I can use it to defend my body against cancer cells (unwanted growth). The 2nd Amendment also says a woman and her doctor can use a firearm to defend herself against police who are trying to stop her from removing an unwanted fetus from her womb. Both decision-makers are highly localized. We don't consider the viewpoint of the cancer, the fetus, or the police to be relevant...even when they're operating off of a different code: the Christian Bible.
Similarly, the 4th Amendment "protects" or "encourages the protection" of my right to carry firearms.
From the standpoint of the U.S. Constitution, long prior to Hitler, Stalin, and Mao's totalitarian regimes...and long before cancer was properly understood...there was a conflict between tyranny and property. The tyrant/parasite and the cancer/parasite both want to eat your body, by derailing your deployment of proper code.
These cybernetics discussions can easily mix together all sorts of ideas...from different sub-domains.
At the end of the day, I wonder...will it be Australia? Southeast Asia? Who will deploy the first effective anti-cancer/anti-totalitarianism medicine?
Will it be proper code? What will that "defense," code look like? Peter Voss's "Freedom Universal"?
Maybe.
I think Southeast Asia will win this battle. Maybe rural China. Maybe Russia.
Western governments full of feminized tyrants (Like Kurt Vonnegut's "Diana Moon Glampers" from "Harrison Bergeron" ...or, IRL, Kamala Harris taking group photos with the BATFE and DEA) have way too much money. "The 5 eyes" want to jail their few remaining journalists. ...Even Whitney Webb had to flee. ...Monstrous...
...monstrous...
@@JakeWitmer You cover many things in your post, but it still doesn't detract from the fact that no one can really stop agi.
Regarding the medicines that you state could cure cancer, well, we know that governments are mostly unsuccessful at stopping the production of meth, crack, and others, so it should be possible to manufacture and distribute the ''banned'' medicines.
New swell robotics everywhere. Ai jobloss is the only thing I worry about anymore. Anyone else feel the same?
How about: AI will provide the Outline, Research and Grammar support to write but you will have to write it yourself! 🤨💪🏾
I've been using llms to take my intellectual brain dumps and construct literary structure. For example, following an interesting day on nursing placement I let out all my thoughts and the AI bot took that and converted it to a chapter list for what might one day be an innovative thesis for future changes in mental health care. Gah now to write the dang thing 😅
AI could be used to check the consistency of the stories before they are published. This would also be very useful for film sequels.
@@NenJiDaPassiv It is already writing it for you, even if you sit down to rewrite it, right?
@@flickwtchr yeah but usually I find it better to do it incrementally. I mean I want the final product to reflect only my ideas, and if I let it get carried away it can vere away from my original thesis. It's kind of like I provide the bones and it fills in the gaps, while making the prose much better than I could on my own.
So I’ve listened to this twice and he’s using retro knowledge to predict…. Hopeful yes but not realizing that we’re are about to be thrown into an exponential
Brilliant thinker, though he does not appear to grasp the implications of the non-locality of matter. Verdon postulates that we can't exist in a nested simulation because the physics of our simulation wouldn't permit the simulation of every particle in the universe, but said particles only manifest when they are observed. Further, complex systems of computation (think the C language) are always built on top of more elemental systems (like assembly/machine language)--or they can be modeled by them according to Turing equivalence. The quantum paradigm is simulated by classical computation.
Wow
Own your own AI - possible ?
The FOMO is real with Revux - get in before it takes off!
Did someone fart? No? Oh, this smells of Theranos...
This was very interesting. However i really can't tell if the guy is a brilliant engineer or a brilliant scam artist, throwing around buzzwords to become the next hyped trillion dollar Startup. I mean every mom and pop bakery is now an AI Enterprise with a billion dollar valuation. Give me a break.
AI is a race to the top. Slowing down won't win you anything but contempt
etherum*? 47:41
Isaac Asimov's Laws of Robotics. Why are regulators finding it hard to use that as regulatory framework?
Because they're simplistic and don't work?!!!
You are aware that was a work of fiction right?
So he’s saying he’s part dinosaur because he is a DNA based life form and survived the meteor impact? 🤔
This guy doesn’t have a product. Just PowerPoints.
To say people need to embrace this and become entrepreneurs and everything‘s gonna work out is crazy. The year is going to take quite a bit of time for people to adjust to all these changes. Not everybody has the ability to do what people in your field do or people who have the resources Can do. What are you gonna do with the person that’s 50 years old and still needs to work for 15 years before they could retire what are you gonna do with the person who’s in college right now studying something that the AI will be able to do before they even get out of college. This is a very, scary thing to just assume people who didn’t ask for this to happen to just roll over and allow the rich to get richer and the IT companies to make billions and trillions of dollars at our expense. It’s our data our information that’s helping these AI is powerful as they claim to be and will be. It feels so one-sided.
When will someone ever challenge this guy and call his bullshit? Or alternatively have him explain concretely how his claim have any basis.
You can.
No one will because, technically, he graduated from the right schools and worked for AlphabetCorp. That gives him carte blanche to espouse his wacky theories.
@@StevenAkinyemi oh believe me I would. But he wouldn’t come. You guys seriously didn’t learn anything from the likes of theranos, juicero etc. Extraordinary claims require extraordinary evidence. I will believe what he says when he has something to show for.
@@bahroum69...At which point, you will have missed the opportunity. So...why not try to vet his claims using logic, reason, and physics knowledge? Note: I don't know who's right, but I want the argument.
@@JakeWitmer That's the point. Because he gives absolutely zero data about anything. Just vague and unsubstantiated claims. Like most scammers. I am not saying he is one. But he surely speaks like one.
Get onboard or be left behind? Who made technologists the authority in directing the future for humankind?
Reality!
Merit in narrow fields. You can also have merit in narrow fields, if you learn. Or, maybe not.
Mr Diamandis has to work faster WW3 around the corner. Where is the Space Shuttle?
Just imagining myself not “left behind” with the ai bs, so that o can continue having existential dread on mars or somth
Recall a hitchhikers naive reflection on this parlance on the road . Sane Doon , wonder how it went .
54:30 Any merit on a tree of distributed stocasticism down to psuedodeterministic terminals .
1:06:11
1:06:11 Wonder does this Tek utilise any of the abstractions inherent in the Tangram asynchronous model of the Philips labs.
The risk of "8 billion" or 'more" AIs is that one of them in the hands of a bad actor (let alone another sufficiently bad and capable AI) may have the know-how and agency to destroy humanity (say through a carefully modified/created virus inducing a new 100% fatal pandemic). In language that Verdon would understand, the chances of "statistical fluctuations" in this context, with such (let's assume unsurvivable) consequences must be contained in some controllable way, for which unlimited AI proliferation is not the answer. Why isn't this obvious.
Everyone has a nuclear bomb vs only the dictator has a nuclear bomb?
What could go wrong?
People made these same exact arguments 25 years ago, about the internet.
It did not deliver.
It could be corrupted by government, so it was. AGI will defeat the failure mode of human government. Maybe what it replaces that failure mode with will be "proper human government"(+AGI) and maybe it will just be "AGI."
@@JakeWitmer I don't think the government was the main thing that ruined the internet.
There is no way I want to face my karma everyday. Not going to wear anything that will allow me to recall the past.
I got a couple of great challenges after we get super intelligence and can do almost anything very easily of course we can start off with taking all the garbage and the plastic that's choking animals in the middle of the ocean and convert that into useful stuff.
We can mind some of the asteroid material and perhaps also Venus and Mercury and some of Mars and then make a ringworld that would give us the equivalent of 3 million times the livable real estate of the surface of Earth and you know you could control it with like shadow square that would simulate day and night except the sun would always be at high noon just like in Larry nivens ringworld.
Oh why don't we just have a nice magnetic bottle of the captures all the solar wind from the Sun and then we have transmutators so then we can keep adding to the elements of the solar system before they get away in the wind.
Send von neuman probes out to maybe the next solar system and if it finds that there's people living there you know individuals that look a lot different than ourselves of course will then they will hang out near the edge and just simply be as watchmen so and then we can perhaps intervene only if it looks like they're getting ready to destroy themselves.
Actually I don't know they just seems to be absolutely no limit to what new challenges we can figure out once we've mastered the old crap.