Topic: [ANN] aiShare: Crowdfund the Decentralized Singularity (Read 11031 times)
One more ai)
Graham - what probability would you assign to AGI-5 in 5 years?
Alix, I have no opinion to offer. My perspective on intelligence is informed by cognitive psychology and it differs so profoundly from yours that the question makes about as much sense to me as if you'd asked “What wood makes the best divining rod?”
I'm outta here, no sense in talking past each other endlessly.
Cheers
Graham
Is this dead or alive ?
Alive - but still mostly in the early whitepaper/idea stage. The design on the wiki is still evolving and we are working on a demonstrator of some of the key initial tech (the neural engine). We are going to slowly spread awareness in a very limited fashion to targeted groups, refining the wiki along the way.
This is all leading up eventually to a main website launch which will have a better intro to the idea, a video or two, press release, etc. The neural engine itself - when appropriately demonstrated and validated - will also generate press, probably timed with the main website launch.
Graham - what probability would you assign to AGI-5 in 5 years?
0.1%? 1%? 10%?
Five years ago, it looked like human level visual recognition was impossibly far in the future. Many said that hitting that performance bar would require fundamental new insights into AI. It did not. All it required was minor improvements to existing ANN algorithms - combined with scaling up to appropriately large amounts of neurons and training experience. There are still challenges in vision - video, 3D object localization, etc - but almost nobody in ML now believes that solving these subproblems will require huge new insights.
Likewise five years ago, few believed that ANNs could be used to build a complete complex AGI. Pattern recognition is important, but there is much more going on in brains - complex sequential planning, attention, episodic recall, etc etc. Many believed AGI would require a mix of different algorithms. Then the deepmind guys trained an ANN to play atari and showed that you could achieve human level play across a wide spectrum of games using deep reinforcement learning across an appropriate initial architecture.
All of this has been achieved by small teams of researchers running models with 10 million neurons or less on just a few GPUs. It has actually required new insights to get this far - many of them in fact - but all fit within a general optimization theory of intelligence that has been solid for a while.
We can achieve AGI-5 by:
*scaling up to millions of GPUs each running billions of neurons (embedding in a video game, for example)
*scaling up the number of researchers to thousands, augmented with new global opt/research automation tools
*creating a new framework to organize the code/design exploration process and manage incentives.
The above combination will accelerate research by a factor of about a million, and will lead to a reasonable probability of AGI-5 in 5 years.
Accomplishing the above will not be easy, but there is literally nothing else more worthwhile. Even if it only has a 1% chance of leading to superintelligence in 10 years, 1% of a 100 trillion dollars (lower bound on the value of superintelligence) is still a trillion - and justifies almost any investment.
Everyone is making a bet, like or not. If you dont invest in AI (and in the right project) in the next 5 to 10 years, you will find your wealth/job/livelihood eroded away regardless.
0.1%? 1%? 10%?
Five years ago, it looked like human level visual recognition was impossibly far in the future. Many said that hitting that performance bar would require fundamental new insights into AI. It did not. All it required was minor improvements to existing ANN algorithms - combined with scaling up to appropriately large amounts of neurons and training experience. There are still challenges in vision - video, 3D object localization, etc - but almost nobody in ML now believes that solving these subproblems will require huge new insights.
Likewise five years ago, few believed that ANNs could be used to build a complete complex AGI. Pattern recognition is important, but there is much more going on in brains - complex sequential planning, attention, episodic recall, etc etc. Many believed AGI would require a mix of different algorithms. Then the deepmind guys trained an ANN to play atari and showed that you could achieve human level play across a wide spectrum of games using deep reinforcement learning across an appropriate initial architecture.
All of this has been achieved by small teams of researchers running models with 10 million neurons or less on just a few GPUs. It has actually required new insights to get this far - many of them in fact - but all fit within a general optimization theory of intelligence that has been solid for a while.
We can achieve AGI-5 by:
*scaling up to millions of GPUs each running billions of neurons (embedding in a video game, for example)
*scaling up the number of researchers to thousands, augmented with new global opt/research automation tools
*creating a new framework to organize the code/design exploration process and manage incentives.
The above combination will accelerate research by a factor of about a million, and will lead to a reasonable probability of AGI-5 in 5 years.
Accomplishing the above will not be easy, but there is literally nothing else more worthwhile. Even if it only has a 1% chance of leading to superintelligence in 10 years, 1% of a 100 trillion dollars (lower bound on the value of superintelligence) is still a trillion - and justifies almost any investment.
Everyone is making a bet, like or not. If you dont invest in AI (and in the right project) in the next 5 to 10 years, you will find your wealth/job/livelihood eroded away regardless.
Is this dead or alive ?
You describe yourself as a rationalist. Unfortunately, you're as mistaken in that as you are in pretty much everything else.
Higgins, I've read through some of your posting history and I can see that you are an intelligent - and normally civil - guy. There's no need to get angry. Yeah, you're right, I shouldn't even have posted. I've had 30 years to learn the essential futility.
Cheers
Graham
You describe yourself as a rationalist. Unfortunately, you're as mistaken in that as you are in pretty much everything else.
Higgins, I've read through some of your posting history and I can see that you are an intelligent - and normally civil - guy. There's no need to get angry.
estimate the performance necessary for replicating all of the cortex, as it has a uniform architecture bollocks
How much computational neuroscience literature have you read? 10 hours worth? 100 hours worth? 1,000 hours worth? I only ask this to speed up the conversation. Your reply of 'bollocks' indicates to me that it is unlikley that you are up to date on computational neuroscience literature.
The general uniformity of the laminar cortical architecture has been a central key finding of neuroscience since the days of Hubel and Weiss's discovery of columnar maps in the primary visual cortex of cats. The same general 6 layered microcircuit with its characteristic distributions of excitatory and inhibitory cell types and initial wiring patterns is repeated throughout the cortex of all mammalian brains, with varying degrees of small differences across cortical regions (motor and frontal cortical regions have some key differences) and between species. At the higher macrocircuit level the cortex is partitioned into (mostly) genetically predefined modules, each of which repeats the same general wiring strategy but has unique intermodule connectivity.
The most significant and striking direct evidence of cortical microcircuit uniformity comes from intentional or accidental rewiring experiments. If the normal 'auditory' cortex is rewired to receive input from the retina, the microcircuit develops the 'software' wiring for visual processing instead (although - as expected due to various reasons - not at the same level of capability).1
The detailed microwiring changes dynamically during development and learning - encoding acquired mind 'software' if you will - and thus ultimately is unique per individual.
In blind humans, the 'visual' cortical modules typically develop advanced audio processing capabilities, including even active echolocation (sonar) in some cases. 2 . 3
The general uniformity of cortical architecture leads to the 'one learning algorithm hypothesis', the key insight behind 'deep learning' (just another name for large cortex inspired ANNs), and the success of that field is strong indirect supporting evidence for the principle. The same general deep CNN architecture that has now achieved primate/human levels of performance in vision is also leading to similar breakthroughs in a wide variety of other pattern recognition challenges such as speech recognition and automated translation.
See this video by Andrew Ng, it summarizes the importance of the universal cortical learning hypothesis and how it has (and is) revolutionizing machine learning.
Also - just to be clear - there is still a great deal of research to do in learning algorithms. The brain uses a complex mix of supervised, unsupervised, and reinforcement learning, and in general its learning machinery is more complex and powerful than anything we have yet in DL/ML.
Even so, our current learning algorithms are already powerful enough to rival isolated equivalent brain circuits in some cases - at least for tasks that are simple enough to construct an ideal training environment. The 'commonsense reasoning' that you mention below will probably require embedding an AGI for virtual years in a complex simulated 3D environment (running the simulation much faster than realtime is key - its a performance issue).
...
machine learning is already making steady exponential progress Patently untrue, do you know what the word means?
Performance is constrained by the complexity of ANNs in terms of size and speed. It takes about a few years of visual experience for a state of the art DANN to achieve human level performance, which requires running the simulation dozens of times faster than real-time so that training takes a more reasonable few weeks. Everything is performance constrained: with more compute power we can train and experiment with ever larger ANNs. Model complexity is thus increasing exponentially concomitantly with performance.
...
what future evidence would cause you to update that to a
Human level general intelligence will require very large ANNs with complexity similar to that of a human brain, along with a similarly complex and lengthy educational development process. Essentially we will need to actually raise and educate AGIs - somewhat like children. The key is accelerating the learning speed, which is already essential for training visual circuits.
There is still a great deal of research to do in unraveling many of the brains macrocircuits and principles. On the other hand, we also have a huge backlog of theories and wiring data - far more than we can test. Massively accelerating experimental throughput is the key to accelerating progress.
Say in 5 years, you experience an AI 'app' that has intelligence - but only that of a 5 year old. I've no idea what you think is to be gained by this orgasm of conjecture but it's certainly not a rationalist's argument, it's unfounded speculation about an as-yet-invisible near-term major breakthrough. And what do you mean: "only that of a five year old"? What towering, empty, casual, blind arrogance. Not a parent, are you? ROFL
The exact timeframe is irrelevant - the question is simply to gauge your current general intuition for where the principle locus of uncertainty/difficulty in achieving AGI resides. Is it in getting to infant AGI? toddler AGI? adult AGI? etc.
For example, once we achieve AGI-5 (toddler AI), it is still uncertain that AGI-30 (adult) is near. It could be that there is a huge remaining complexity gap. I doubt this - instead I believe most of the complexity/difficulty is in getting up to AGI-5. In that - at least - it appears we agree.
initially its just simulating infant brains that don't do much, and certainly aren't interesting yet You consider the cognitive plasticity of infants to be uninteresting yet anticipate the development of AGI in ML in just a decade or two? HAHAHAHAHAHA. Please stop, it hurts.
Now I sense that you are intentionally misreading me. By 'simulating an infant brain', I meant simulating a brain the size and complexity of a human infant - without yet understanding the correct seed wiring architecture. It would be a brain dead infant, essentially. There is still a huge amount of research work to do going from "now we can simulate an infant brain" to "now we understand all of the initial wiring/algorithms that infant brains need to become adults".
You might want to educate yourself a bit more about the G in AGI, at least do yourself the favour of reading the commonsense problem page and Vaughan Pratt's very accessible notes on the commonsense reasoning problems observable in Lenat's CYC.
A while back the AI/AGI community split into a few directions. One group believed that AGI could be achieved by advances in logic and traditional computer science. Another group believed that AGI will require artificial brains - based on understanding the learning mechanisms in real brains. The former group's thesis has essentially been disproven and is a dead-end, and all advances in AI at this point are actually coming from the latter group (neuroscience/machine learning).
General 'commonsense' reasoning requires a general internal predictive model of the world, which is something that comes around or after AGI-5. Its not an isolated problem you can solve independently before hand - it's essentially AI complete.
Cheers
-Alix
You describe yourself as a rationalist. Unfortunately, you're as mistaken in that as you are in pretty much everything else.
estimate the performance necessary for replicating all of the cortex, as it has a uniform architecture bollocks
...
machine learning is already making steady exponential progress Patently untrue, do you know what the word means?
...
what future evidence would cause you to update that to ahigh"higher", enough with the heavy-handed framing probablility? any evidence of advances in the understanding of the G in AGI (see below).
...
an AI 'app' that has true adult human level intelligence - like Samantha from Her Er, you seem to have gotten out of your depth, you offer fallacy in place of logic and fantasy in place of reality. You're either ignorant or careless of the "no true scotsman" fallacy and you reference a model of AI developed by a movie scriptwriter LOL
...
Say in 5 years, you experience an AI 'app' that has intelligence - but only that of a 5 year old. I've no idea what you think is to be gained by this orgasm of conjecture but it's certainly not a rationalist's argument, it's unfounded speculation about an as-yet-invisible near-term major breakthrough. And what do you mean: "only that of a five year old"? What towering, empty, casual, blind arrogance. Not a parent, are you? ROFL
...
initially its just simulating infant brains that don't do much, and certainly aren't interesting yet You consider the cognitive plasticity of infants to be uninteresting yet anticipate the development of AGI in ML in just a decade or two? HAHAHAHAHAHA. Please stop, it hurts.
You might want to educate yourself a bit more about the G in AGI, at least do yourself the favour of reading the commonsense problem page and Vaughan Pratt's very accessible notes on the commonsense reasoning problems observable in Lenat's CYC.
Frankly, I see only simplistic models of intelligence and media-fuelled infantilism; a modern version of the media's feasting off Mary Shelley's fantastical monster and the simplistic models of physiology that underpinned it.
Cheers
Graham
estimate the performance necessary for replicating all of the cortex, as it has a uniform architecture bollocks
...
machine learning is already making steady exponential progress Patently untrue, do you know what the word means?
...
what future evidence would cause you to update that to a
...
an AI 'app' that has true adult human level intelligence - like Samantha from Her Er, you seem to have gotten out of your depth, you offer fallacy in place of logic and fantasy in place of reality. You're either ignorant or careless of the "no true scotsman" fallacy and you reference a model of AI developed by a movie scriptwriter LOL
...
Say in 5 years, you experience an AI 'app' that has intelligence - but only that of a 5 year old. I've no idea what you think is to be gained by this orgasm of conjecture but it's certainly not a rationalist's argument, it's unfounded speculation about an as-yet-invisible near-term major breakthrough. And what do you mean: "only that of a five year old"? What towering, empty, casual, blind arrogance. Not a parent, are you? ROFL
...
initially its just simulating infant brains that don't do much, and certainly aren't interesting yet You consider the cognitive plasticity of infants to be uninteresting yet anticipate the development of AGI in ML in just a decade or two? HAHAHAHAHAHA. Please stop, it hurts.
You might want to educate yourself a bit more about the G in AGI, at least do yourself the favour of reading the commonsense problem page and Vaughan Pratt's very accessible notes on the commonsense reasoning problems observable in Lenat's CYC.
Frankly, I see only simplistic models of intelligence and media-fuelled infantilism; a modern version of the media's feasting off Mary Shelley's fantastical monster and the simplistic models of physiology that underpinned it.
Cheers
Graham
You can lose the “Brain-Engineering” white paper, it's tinfoil hat grade and you'd be horribly mistaken if you believe that it has a positive effect.
That article is not really a whitepaper and was misnamed: it really should have been called "Brain Hardware Parity" and now is. The actual high level network architecture for decentralized/distributed mind simulation is here: http://aishare.co/wiki/Cryptagoric_Cortical_Networks (warning: work in progress).
As to the "tinfoil hat grade", I'm not quite sure what you mean by that: this has nothing to do with conspiracy theories, it's just a basic review of recent computational neuroscience focused narrowly on performance comparisons. We now have ANNs that can replicate some specific cortical functionality very well (visual recognition), and we can use that - along with various low level metrics - to estimate the performance necessary for replicating all of the cortex, as it has a uniform architecture.
In essence, the coin aims to support the funding of AI research yet is based on an appeal to the popular but sadly doomed hope of the appearance of a prophetsingularity that will, as if by magic, solve all the currently-blocking AI problems.
As a field, machine learning is already making steady exponential progress. You can ignore the evidence for that progress - that's up to you.
This is simply a plan to greatly accelerate ML progress through a combination of:
* more efficient use of local compute resources (GPU neural simulation efficiency)
* an efficient distributed/decentralized architecture to leverage a wide network of GPUs
* a more efficient incentive structure
I consider myself a rationalist, and you probably do as well. The essence of rationality is updating on evidence.
I get the impression that you assign a low probability to the general possibility of near future AGI, and or this proposal in particular.
So, what future evidence would cause you to update that to a high probability?
Say in 5 to 10 years, you hear about an AI 'app' that has true adult human level intelligence - like Samantha from Her. You try it - it works, and it appears to be smarter than anyone you know. Do you still not believe?
Say in 5 years, you experience an AI 'app' that has intelligence - but only that of a 5 year old. How much would this evidence shift your belief on whether adult level AI is near?
Say sometime later this year, you read about some new neural simulation engine in the news which can simulate a full human brain equivalent with billions of neurons on a single GPU. The code/design is proven and its exciting to researchers, but initially its just simulating infant brains that don't do much, and certainly aren't interesting yet. How much would this shift your belief on whether 5-year old AI is near?
This fundamental contradiction between ends and means is the root of an incoherence that will prove ultimately fatal to the coin; the cracks are already beginning to show:
This project is in its early days - all feedback is welcome.
If your smarty pants friends have some specific smart criticism - we'd love to hear it.
That latter quote was sincere - not sarcastic - and any offense was unintentional. I actually really do want feedback and criticism.
Thank you for your feedback.
Cheers
-Alix
very nice Project..looking forward to this.
Our goal is to decentralize the Singularity
You can lose the “Brain-Engineering” white paper, it's tinfoil hat grade and you'd be horribly mistaken if you believe that it has a positive effect.
In essence, the coin aims to support the funding of AI research yet is based on an appeal to the popular but sadly doomed hope of the appearance of a
This fundamental contradiction between ends and means is the root of an incoherence that will prove ultimately fatal to the coin; the cracks are already beginning to show:
This project is in its early days - all feedback is welcome.
If your smarty pants friends have some specific smart criticism - we'd love to hear it.
Cheers
Graham
If you haven’t noticed the trends, the capitalist mode of production is in its death throes. Massively growing inequality and impotence among the public, inflexible concentration of political and financial power, continual encroachment of privacy, stupendous wastes of human capital, depletion of natural resources, rampant military conquest and violence, do I really need to go on? Surely, you realize this is madness. These are the gifts left to us by the capitalist ideology. We are absolutely doomed if we continue to think in these terms. Your assumptions about the nature of value are flawed for this very reason. This is the heart of the decentralization movement, as well as the cold predatory necessity of the cryptocoin initiative. It’s based on the logic of abundance rather than artificial scarcity; the cosmopolitan spirit of open collaboration instead of material greed and consumption. In this picture, the singularitarians and transhumanists are the minority.
Despite their many accomplishments, Google is still largely trapped in the old regime. Like all the other internet giants, their immense profits are not legitimate because they are parasitic on the value created by millions of users adding content to their infrastructure. You see what’s happened, don’t you? They've privatized the internet for their own selfish motives. And we’re supposed to be grateful if they throw us a few scraps of the prosperity in the form of free applications and services? That’s totally irrelevant. Google is totally irrelevant. Once we appropriate the web for ourselves we unlock all its value. This will be many orders of magnitude above your prediction of 30X. The internet is the global economy. All of us working together is super-intelligence. Maybe you’re too caught up in your personal projects to respect this fact, but that is the direction the internet is going. Lucky for us, it doesn't require any techno-magic or quantum woo to realize. We've already got the technology. And it’s only going to get better.
There’s no need to take offense or spoil the celebration with needless skepticism. No one is saying we don’t want super-intelligence. Yours is one of many projects that need to be pursued in some way or another. General AI that can benefit all of us will happen in its own time, and you have my full support in moving that along. However, we need to recognize this is our time. Humanity's time to right the fucked mistakes of our predecessors. In fact, it’s the first time since the last true regime change (Bourgeois capitalism) that something like this is even feasible. This is coming from someone who grew up with the internet. Never in my wildest dreams would I have imagined an opportunity like this presenting itself, and yet here I am fomenting revolution. I simply see no benefit in splitting hairs with you over the technical details of inefficient hashes and passive rent-seeking. We already acknowledge there are limitations, there are always limitations. That’s not going to change with fancier AI. We need to work with what we know. What know is that decentralized applications and blockchain technology work. Ethereum is part of that equation, but it’s not a complete solution. Something else needs to be built on-top of that foundation, something revolutionary. We’re all secretly dying for this happen, so let’s make it happen.
What definitely won’t make it happen is pie in the sky thinking about smarter robots. Let the crippled elite worry about that. They want to replace us all with machines anyway, so let’s let them build the infrastructure. By the time most of human society becomes automated, we’ll have won our revolution, and we can begin taking over their industries. There is no need to waste the communities energies on something that can be supported by wealthier patrons. If your work is sound, the rest of us will follow. What we definitely won’t follow, if our hearts and minds are in the right place, is someone who works to perpetuate the capitalist nightmare. Don’t pretend for a moment we’re not at war with the old regime. Enemies of the revolution are to be disrupted, tricked, and outsmarted. We have to be merciless enough to defeat them, but compassionate enough save them. It’s time to use real strategy and tactics.
Your concerns about new monopolies emerging are very real. I also dread the possibility of currency investors and miners squeezing out the common users. Every person needs to be given equal opportunities, but at the same time there needs to be incentives built into the system to drive further innovation. My proposal is that we create a new currency that will be interoperable with the Ethereum platform, let’s call it 'Prometheum'. If Ethereum is the oil, then Prometheum would be the spark to ignite the flames of our rebellion. We market this as a massively decentralized world-historical political experiment that can enable the general public to collaborate on all aspects of digital statecraft, including constitutional law, distributive justice, institutional engineering, and revolutionary activism. The objective would be to rapidly accelerate the development of an incorruptible stateless society for all human beings on this planet. Stake in the currency would be completely free and community funded providing most essential services. Upon registration, a universal passport would be officially certified. This entitling every person to a starting balance of the Prometheum currency in addition to an unconditional basic income, to be deposited into their accounts at the end of every economic cycle. All productive activity would in this way be mediated by an autonomous financial instrument responsible for the just distribution of profits. Effectively, participants split 50% of their earnings with a global accounting system that repays a fractional sum to each member. In this scenario, productive work is encouraged, but is not required in order to benefit from the wealth of the network.
This is nobody’s idea. This is the idea of the age. Whatever its ultimate practical implementation, the end result is the same: the egalitarian distribution of currency. From this simple principle we acquire the universal consent of the governed, and thus the support of every freedom loving person of the civilized world. As the project picks up momentum more and more people will be recruited to the cause, from the starving children of the third world to the charismatic celebrities of the Western nations, each playing a crucial role in selling this idea to the public. This will promote an exponential explosion of consensus-making. And together, we will design a new society. When this comes to pass, it won’t matter if an established power tries to liquidate its assets into the new regime, nor will it matter if someone acquires more wealth than others, as their investment is fairly absorbed and redistributed to everyone. Monopolies are mitigated by a modification of the difference principle, i.e. it only permits inequalities that work to the advantage of the system as a whole. We don’t need anything particularly complicated beyond this simple accounting measure. By this time we will have collectively determined the principles and architecture of the world we want to live in, and a new official currency: let’s call this collective development ‘civilization 1.0’. Society becomes like an operating system with its own versioning standards. We then continue migrating through these systems, retaining the value of all legacy contracts and currencies, endlessly evolving into the future. This isn’t mind-projection, it’s the ongoing logical conclusion of present developments.
Despite their many accomplishments, Google is still largely trapped in the old regime. Like all the other internet giants, their immense profits are not legitimate because they are parasitic on the value created by millions of users adding content to their infrastructure. You see what’s happened, don’t you? They've privatized the internet for their own selfish motives. And we’re supposed to be grateful if they throw us a few scraps of the prosperity in the form of free applications and services? That’s totally irrelevant. Google is totally irrelevant. Once we appropriate the web for ourselves we unlock all its value. This will be many orders of magnitude above your prediction of 30X. The internet is the global economy. All of us working together is super-intelligence. Maybe you’re too caught up in your personal projects to respect this fact, but that is the direction the internet is going. Lucky for us, it doesn't require any techno-magic or quantum woo to realize. We've already got the technology. And it’s only going to get better.
There’s no need to take offense or spoil the celebration with needless skepticism. No one is saying we don’t want super-intelligence. Yours is one of many projects that need to be pursued in some way or another. General AI that can benefit all of us will happen in its own time, and you have my full support in moving that along. However, we need to recognize this is our time. Humanity's time to right the fucked mistakes of our predecessors. In fact, it’s the first time since the last true regime change (Bourgeois capitalism) that something like this is even feasible. This is coming from someone who grew up with the internet. Never in my wildest dreams would I have imagined an opportunity like this presenting itself, and yet here I am fomenting revolution. I simply see no benefit in splitting hairs with you over the technical details of inefficient hashes and passive rent-seeking. We already acknowledge there are limitations, there are always limitations. That’s not going to change with fancier AI. We need to work with what we know. What know is that decentralized applications and blockchain technology work. Ethereum is part of that equation, but it’s not a complete solution. Something else needs to be built on-top of that foundation, something revolutionary. We’re all secretly dying for this happen, so let’s make it happen.
What definitely won’t make it happen is pie in the sky thinking about smarter robots. Let the crippled elite worry about that. They want to replace us all with machines anyway, so let’s let them build the infrastructure. By the time most of human society becomes automated, we’ll have won our revolution, and we can begin taking over their industries. There is no need to waste the communities energies on something that can be supported by wealthier patrons. If your work is sound, the rest of us will follow. What we definitely won’t follow, if our hearts and minds are in the right place, is someone who works to perpetuate the capitalist nightmare. Don’t pretend for a moment we’re not at war with the old regime. Enemies of the revolution are to be disrupted, tricked, and outsmarted. We have to be merciless enough to defeat them, but compassionate enough save them. It’s time to use real strategy and tactics.
Your concerns about new monopolies emerging are very real. I also dread the possibility of currency investors and miners squeezing out the common users. Every person needs to be given equal opportunities, but at the same time there needs to be incentives built into the system to drive further innovation. My proposal is that we create a new currency that will be interoperable with the Ethereum platform, let’s call it 'Prometheum'. If Ethereum is the oil, then Prometheum would be the spark to ignite the flames of our rebellion. We market this as a massively decentralized world-historical political experiment that can enable the general public to collaborate on all aspects of digital statecraft, including constitutional law, distributive justice, institutional engineering, and revolutionary activism. The objective would be to rapidly accelerate the development of an incorruptible stateless society for all human beings on this planet. Stake in the currency would be completely free and community funded providing most essential services. Upon registration, a universal passport would be officially certified. This entitling every person to a starting balance of the Prometheum currency in addition to an unconditional basic income, to be deposited into their accounts at the end of every economic cycle. All productive activity would in this way be mediated by an autonomous financial instrument responsible for the just distribution of profits. Effectively, participants split 50% of their earnings with a global accounting system that repays a fractional sum to each member. In this scenario, productive work is encouraged, but is not required in order to benefit from the wealth of the network.
This is nobody’s idea. This is the idea of the age. Whatever its ultimate practical implementation, the end result is the same: the egalitarian distribution of currency. From this simple principle we acquire the universal consent of the governed, and thus the support of every freedom loving person of the civilized world. As the project picks up momentum more and more people will be recruited to the cause, from the starving children of the third world to the charismatic celebrities of the Western nations, each playing a crucial role in selling this idea to the public. This will promote an exponential explosion of consensus-making. And together, we will design a new society. When this comes to pass, it won’t matter if an established power tries to liquidate its assets into the new regime, nor will it matter if someone acquires more wealth than others, as their investment is fairly absorbed and redistributed to everyone. Monopolies are mitigated by a modification of the difference principle, i.e. it only permits inequalities that work to the advantage of the system as a whole. We don’t need anything particularly complicated beyond this simple accounting measure. By this time we will have collectively determined the principles and architecture of the world we want to live in, and a new official currency: let’s call this collective development ‘civilization 1.0’. Society becomes like an operating system with its own versioning standards. We then continue migrating through these systems, retaining the value of all legacy contracts and currencies, endlessly evolving into the future. This isn’t mind-projection, it’s the ongoing logical conclusion of present developments.
This looks , quite intresting so much text , usually dev's write 5 sentences and hope they get some money , but this looks like a real dev team , with real intentions and not just looking to make money, Definetly gonna keep an eye on this , and hope it works out good for you guys! 
Alright Alix, I’ve digested what you’ve said.
What you and others are referring to as coup-completeness or regime change is interesting. I think this is a real possibility with present technology, as I’ll attempt to elaborate. But before I do that, I want to comment on some of your expectations for the future.
What you and others are referring to as coup-completeness or regime change is interesting. I think this is a real possibility with present technology, as I’ll attempt to elaborate. But before I do that, I want to comment on some of your expectations for the future.
You say you've digested what I've said, but in the linked posts I provided earlier I made a very strong argument (I would say a knockdown argument) that no PoW or PoS based cryptocurrency design can possibly be coup-complete. Coup-completeness requires immense value creation orders of magnitude beyond what existing designs can offer - indeed far beyond the value of any transaction system.
The dollar, obviously, has already accomplished coup: it became the world reserve currency. It did this because the united states became the world's superpower, by actually winning several wars against its competitors on the strengths of superior net technological intelligence (the source of economic power). The dollar reflects the value of the US economy.
Proof-of-Work (PoW) currencies like Bitcoin and Ethereum are like distributed governments that have very high early inflation which eventually tapers off, and they use this seniorage income to subsidize enormous worthless hashing computations.
Reductio ad absurdum: Imagine, for sake of argument, that a PoW currency was universally adopted by the citizens of a nation - as Bitcoiners dream. The national currency would then be abandoned, and tens of trillions worth in wealth (for US or Euro sized examples) would be transferred from the current 1% wealthy elite to a new 1% (or more likely, 0.1%) elite. The new elite would be composed of some mix of two groups: 1.) currency investors, and 2.) miners. Crucially, wealth is simply transferred from one group to another, without actually creating new wealth in proportionate trade - as in new valuable technology.
From the perspective of the entire nation, this change is at best a marginal economic improvement, and adds significant new increased costs for tax collection, enforcement, etc. It's simply a wealth redistribution that favors anti-government libertarians (early investors) at the expense of pro-government liberal types (late investors). You may have better luck attempting to just pass legislation that redistributes all the dollars to libertarians. Those cryptocurrency advocates who naively think that everyone else will just 'wake up' to the obvious superiority of the new regime are suffering severely from the mind projection fallacy.
In a capitalist system, wealth distribution goes to those who provide value. We've had many partial coups - Google is worth $600B (3% of the entire stock market) because they actually created a bunch of new technology to dominate markets - they actually created value. A real coup requires something about 30X more powerful than google in terms of value added to the economy. Hashes are not valuable (PoW). Neither is passive rent-seeking (PoS).
Could a cryptocurrency be coup-complete? Yes - but it would have to also function more like a crypto-stock, where all of the new wealth redistribution actually creates an enormous plethora of immensely valuable new technologies. A PoW coin that sucks in $10 trillion doesn't create anything useful but $10 trillion worth of useless mining hardware and waste heat. If you instead spent $10 trillion on general AI, you could potentially takeover the world: because an AI economy is actually worth that much (and more).
Quote
I sympathize with your utopian dream of an all-powerful, all-valuable, all-technological intelligence bootstrapping itself into existence and saving humanity, but . .
You are now arguing against a strawman model of our future vision - one unrelated to our model. I ask instead that you attempt to steelman my arguments, and I'll attempt likewise.
My actual model - in simple form for our purposes - was defined earlier as:
Quote
We are essentially making a bet that at least general AI is possible in less than ten years (software intelligence better at humans at programming, science, etc. that runs on a single machine with a couple GPUs and thus costs < $1,000 a year). What do you think the probability of that is? Do you have a pretty deep understanding of machine learning and neuroscience? So pick some probability for that scenario.
I'd like you to actually answer those questions and focus specifically on that possibility.
To make an accurate assessment about the near term prospects of general AI, one needs to actually have real knowledge in the fields of machine learning and computational neuroscience, and especially specific knowledge concerning new developments in deep learning.
This article is a good introduction to that body of knowledge.
Understanding the future crucially hinges on understanding the state of the art in these fields.
Quote
The most significant reason is that it ignores or denies the scale of innovation required to solve the embodiment problem.
This we largely agree with, as will many other AI researchers. There is a growing awareness in the field that embodied development is important, and all the recent state of the art research involves a form of virtual embodiment during training.
Quote
While its true sophisticated quantum computers might provide the required granularity of material detail to house true AI,
This is pseudoscience that appeals to those who lack sufficient grounding in the requisite technical fields. Quantum computing is enormously misunderstood and overhyped by the general media, and the brain is absolutely, provably not a quantum computer.
Quantum computing would enable some large speedups, but only for some very specific types of computations. It could have relevance to AI oneday, but that day is far away, and we certainly don't need QC for superintelligence.
We already have achieved superhuman performance in many narrow domains (Deep Blue - chess, Watson - jeopardy).
The most recent key development is we are closing in on parity with the primate visual system, which is a much much bigger development than Deep Blue or Watson. The visual system is 25% of the monkey brain and about 10% of the human brain, and the cortex uses essentially a uniform architecture. So figuring out the visual system leads directly to figuring out the rest of the cortex.
Quote
Hence, we face the vicious circularity of requiring an all-powerful intelligence before we can design systems capable of producing an all-powerful intelligence.
So ... you don't believe in evolution?
Quote
Does that mean we can’t create extremely sophisticated forms of artificial intelligence? No, of course not. We’re already accomplishing that. . . .Systems of this sort will always fail to live up to our deeper expectations. For this reason, I would be extremely careful in deciding what kind of power we hand over to such systems.
There's some contradictions here. You argue first that strong AI is impossible because quantum-woo, but then according to you we can create "extremely sophisticated forms of artificial intelligence", but these somehow "fail to live up to our deeper expectations". And then you also seem worried about handing over power.
aiShare's goal is simply to create "extremely sophisticated forms of artificial intelligence". It doesn't matter whether they have 'true consciousness' or quantum-woo qualia, or whatever other non-scientific philosophical attributes you arbitrarily choose to associate only with human brains.
What matters in the end is economic value - can they do our jobs? And the answer, of course - as you have more or less already admitted - is yes.
Quote
With that said, let’s take stock of what we can actually do right now, or at least what we should be able to do in the near future. You argue there is an intrinsic limit to current blockchain design, in the neighborhood of 1,000-10,000 transactions per second. I’m not clear why you think this is a significant problem.
I already have spent a reasonable amount of mental energy attempting to clarify the need for scalability.
I already gave you one example of an interesting application that requires enormous transaction volume : a nano-payment based file sharing platform.
The other big future application category is distributed autonomous prediction markets. These are a key mechanism for a future machine economy and central to our vision. Imagine millions of agents each engaging in thousands to hundreds of thousands of nano-trades per second to predict anything and everything: the weather, traffic, crime before it occurs (haha - half joking), and of course: search and advertising. The key problem there is predicting what a user wants based on their text query, what actions they will take next, and how much influence is worth in each case. So essentially - a decentralized competitor to google.
These applications require or benefit from billions to trillions of transactions per second.
Stop thinking narrowly about just human scale transactions - there is very little untapped potential there.
Those prediction examples are all just external applications, where the most important long term applications are internal - distributed prediction and optimization markets to drive AI research itself.
A market is the most powerful global optimization engine - and its performance is measured in transactions per second. So TPS is everything. From this perspective, Bitcoin and Ethereum are like vacuum tube technology.
You are talking about going from 7 tps to 1,000 to eventually 10,000. I am talking about going directly to the range of a billion to a trillion tps. You seem happy catching up to Visa transactions levels, I would be only moderately satisfied with exceeding the entire transaction throughput of the current world financial markets combined (all the HFT stuff at NASDAQ, BATS - we can go far beyond that).
Quote
The limit on transactions is only going to continue decreasing with advances in computational power. We also know this can be solved using pegged sidechains and interoperable currencies, giving us the enhanced functionality, scalability, and efficiency you seem to think we’re lacking.
Sidechains/Treechains are still O(N) unless you have a sidechain/treechain per user (at least as I understand them) - which of course sacrifices all fraud robustness. They are a half-step in the right direction, but they are starting very far from the optimal.
We are claiming to have solved O(C) decentralized transactions - which of course is optimal. That's a strong claim to make, but that is the claim we are making.
Optimal Scalability
Optimality means there is no better possible design (in at least the measured criteria). The best that other networks can do is just eventually evolve into the same O(C) scalable design.
If Ethereum was based on the optimal scalable design, and there wasn't some large cost to using their network, we wouldn't have needed to work on the transaction problem and could have just used their system. But alas, we had to solve that subproblem
Quote
And it all begins with something as silly as dumb nodes trading bits on a network.
Indeed - but which bits on which network? That is the question.
Will watch this Project..a huge Project.
Alright Alix, I’ve digested what you’ve said.
What you and others are referring to as coup-completeness or regime change is interesting. I think this is a real possibility with present technology, as I’ll attempt to elaborate. But before I do that, I want to comment on some of your expectations for the future.
I sympathize with your utopian dream of an all-powerful, all-valuable, all-technological intelligence bootstrapping itself into existence and saving humanity, but let’s be realistic. We’re nowhere even close to achieving something like this. We could have every brain and computer on the planet working on this project and it still wouldn’t happen for hundreds if not thousands of years. The Singularitarian fantasy that we can construct a machine capable of radical intelligent self-evolution is based on the faith that we are approximating true intelligence through whole-brain simulation, and that the exponential growth in computational power is in some way a measure of our progress towards this goal. This expectation is deeply flawed for several reasons.
The most significant reason is that it ignores or denies the scale of innovation required to solve the embodiment problem. True intelligence is instantiated in an organic substrate that has evolved over millions of years, its underlying informational architecture (DNA) is therefore the product of complex quantum interactions occurring in the presence of deep time. The series of developments giving rise to human consciousness and intelligence is unimaginable orders of magnitude beyond what is conceivable with present and future technologies. While its true sophisticated quantum computers might provide the required granularity of material detail to house true AI, we are totally in the dark concerning the correct programming language to drive the system to true intelligence and self-evolution. The limitations on human knowability in determining the right computational algorithms to achieve this result dwarfs our mortal intellects. Hence, we face the vicious circularity of requiring an all-powerful intelligence before we can design systems capable of producing an all-powerful intelligence. Crowdfund all you want, this isn’t solving itself without massive quantum leaps in human understanding. Does that mean we can’t create extremely sophisticated forms of artificial intelligence? No, of course not. We’re already accomplishing that. However, these are chimeras for the predictable future, mere deterministic mimics of the real thing. Systems of this sort will always fail to live up to our deeper expectations. For this reason, I would be extremely careful in deciding what kind of power we hand over to such systems.
With that said, let’s take stock of what we can actually do right now, or at least what we should be able to do in the near future. You argue there is an intrinsic limit to current blockchain design, in the neighborhood of 1,000-10,000 transactions per second. I’m not clear why you think this is a significant problem. The limit on transactions is only going to continue decreasing with advances in computational power. We also know this can be solved using pegged sidechains and interoperable currencies, giving us the enhanced functionality, scalability, and efficiency you seem to think we’re lacking. To my knowledge, the primary singleton is not open to malicious attacks because the damage is confined to the sidechain itself. If the sidechain is insulated despite being bidirectional, where precisely is the threat coming from your perspective? If there isn’t one, the optimal strategy is simply to use Ethereum as the central precious resource (diamond if you will), implementing new instruments (platinum, gold, silver, etc) with the adaptive growth of the system. In this scenario, we only need to count on the collective intelligence of the users and the consensual utility they can build into the network to promote local maximums and global optimization.
Politically speaking, I think Ethereum is the beginning of a stateless open society. Essentially, we ignore world governments, corporations, and every other coercive entity. We take everything useful we’ve learned from previous social arrangements and crowdsource an inviolable (hardcoded) universal constitution, a set of basic rights, and essential institutions. This project is then backed by the Ethereum currency. From here we create a contract of citizenship and declare ourselves a digital commonwealth. Those interested in participating are registered and treated as anonymous members of a stateless society. At this point we start building decentralized foundations, organizations, and entrepreneurial innovations, replicating all the profit motivated platforms of the old web in a way that incentives common users to join in, effectively siphoning the enormous symbolic capital and market utility the internet makes possible. In this way, we capture the public imagination and everything starts going viral extremely quickly. No central government or corporation can exploit our cooperation. Out of the ashes of the old system we create a new one that is infinitely more valuable. Everyone will start using our applications and migrating to our systems, they will begin pouring their hopes and dreams into our righteous cause, establishing a new world social order founded on just principles, thus securing coup-completeness.
If this explodes to the level of intensity I think it’s capable, we will force the ruling powers into a bind they cannot overcome. Naturally, their mentality is motivated by possession of territory, physical property, military conquest, political engineering, media propaganda, and the means of productive industry; and through these goods they think they can own us too. This apparently impenetrable system of control is wrought from centuries of injustice, signed in the blood of age old contracts. Who are we to pry it from their fingers? What they’ve failed to recognize is the immense value of informational capital. Hence, we have something they desperately need to survive. If we appropriate the internet for ourselves and declare it a sovereign nation, they will not oppose us, in fact they will welcome us. This is the transvaluation of all value, the transmutation of material property into technological abundance and prosperity. Once that happens the scales will tip and they will be forced to undercut physical goods in exchange for virtual ones. Everyone on the planet will benefit from this development. And it all begins with something as silly as dumb nodes trading bits on a network.
What you and others are referring to as coup-completeness or regime change is interesting. I think this is a real possibility with present technology, as I’ll attempt to elaborate. But before I do that, I want to comment on some of your expectations for the future.
I sympathize with your utopian dream of an all-powerful, all-valuable, all-technological intelligence bootstrapping itself into existence and saving humanity, but let’s be realistic. We’re nowhere even close to achieving something like this. We could have every brain and computer on the planet working on this project and it still wouldn’t happen for hundreds if not thousands of years. The Singularitarian fantasy that we can construct a machine capable of radical intelligent self-evolution is based on the faith that we are approximating true intelligence through whole-brain simulation, and that the exponential growth in computational power is in some way a measure of our progress towards this goal. This expectation is deeply flawed for several reasons.
The most significant reason is that it ignores or denies the scale of innovation required to solve the embodiment problem. True intelligence is instantiated in an organic substrate that has evolved over millions of years, its underlying informational architecture (DNA) is therefore the product of complex quantum interactions occurring in the presence of deep time. The series of developments giving rise to human consciousness and intelligence is unimaginable orders of magnitude beyond what is conceivable with present and future technologies. While its true sophisticated quantum computers might provide the required granularity of material detail to house true AI, we are totally in the dark concerning the correct programming language to drive the system to true intelligence and self-evolution. The limitations on human knowability in determining the right computational algorithms to achieve this result dwarfs our mortal intellects. Hence, we face the vicious circularity of requiring an all-powerful intelligence before we can design systems capable of producing an all-powerful intelligence. Crowdfund all you want, this isn’t solving itself without massive quantum leaps in human understanding. Does that mean we can’t create extremely sophisticated forms of artificial intelligence? No, of course not. We’re already accomplishing that. However, these are chimeras for the predictable future, mere deterministic mimics of the real thing. Systems of this sort will always fail to live up to our deeper expectations. For this reason, I would be extremely careful in deciding what kind of power we hand over to such systems.
With that said, let’s take stock of what we can actually do right now, or at least what we should be able to do in the near future. You argue there is an intrinsic limit to current blockchain design, in the neighborhood of 1,000-10,000 transactions per second. I’m not clear why you think this is a significant problem. The limit on transactions is only going to continue decreasing with advances in computational power. We also know this can be solved using pegged sidechains and interoperable currencies, giving us the enhanced functionality, scalability, and efficiency you seem to think we’re lacking. To my knowledge, the primary singleton is not open to malicious attacks because the damage is confined to the sidechain itself. If the sidechain is insulated despite being bidirectional, where precisely is the threat coming from your perspective? If there isn’t one, the optimal strategy is simply to use Ethereum as the central precious resource (diamond if you will), implementing new instruments (platinum, gold, silver, etc) with the adaptive growth of the system. In this scenario, we only need to count on the collective intelligence of the users and the consensual utility they can build into the network to promote local maximums and global optimization.
Politically speaking, I think Ethereum is the beginning of a stateless open society. Essentially, we ignore world governments, corporations, and every other coercive entity. We take everything useful we’ve learned from previous social arrangements and crowdsource an inviolable (hardcoded) universal constitution, a set of basic rights, and essential institutions. This project is then backed by the Ethereum currency. From here we create a contract of citizenship and declare ourselves a digital commonwealth. Those interested in participating are registered and treated as anonymous members of a stateless society. At this point we start building decentralized foundations, organizations, and entrepreneurial innovations, replicating all the profit motivated platforms of the old web in a way that incentives common users to join in, effectively siphoning the enormous symbolic capital and market utility the internet makes possible. In this way, we capture the public imagination and everything starts going viral extremely quickly. No central government or corporation can exploit our cooperation. Out of the ashes of the old system we create a new one that is infinitely more valuable. Everyone will start using our applications and migrating to our systems, they will begin pouring their hopes and dreams into our righteous cause, establishing a new world social order founded on just principles, thus securing coup-completeness.
If this explodes to the level of intensity I think it’s capable, we will force the ruling powers into a bind they cannot overcome. Naturally, their mentality is motivated by possession of territory, physical property, military conquest, political engineering, media propaganda, and the means of productive industry; and through these goods they think they can own us too. This apparently impenetrable system of control is wrought from centuries of injustice, signed in the blood of age old contracts. Who are we to pry it from their fingers? What they’ve failed to recognize is the immense value of informational capital. Hence, we have something they desperately need to survive. If we appropriate the internet for ourselves and declare it a sovereign nation, they will not oppose us, in fact they will welcome us. This is the transvaluation of all value, the transmutation of material property into technological abundance and prosperity. Once that happens the scales will tip and they will be forced to undercut physical goods in exchange for virtual ones. Everyone on the planet will benefit from this development. And it all begins with something as silly as dumb nodes trading bits on a network.
Nexus 9, we have a similar long term goal. You say :
In particular, see our reply here concerning coup-completeness.
Ethereum has great marketing, but the core of their technology design - what we actually care about - is not a huge advancement beyond Bitcoin and alternatives compatible with it (Mastercoin, Counterparty, etc). They may end up having the best practical implementation of a scriptable blockchain, or maybe just better marketing - but its still just a scriptable blockchain.
It will be able to process what - 1,000 transactions per second? 10,000 if they really compress transactions and or limit themselves to fiber nodes - maybe? The core problem is O(N) transaction cost, so performance doesn't scale as you add more nodes. That's not an improvement versus current technology, its a step back to the 80's.
Will you be able to build some sorts of applications on that? Yes, but only by dropping almost everything off the blockchain and using some side database - essentially abandoning the design - which then opens up fraud and attacks again. This just isn't interesting from a technology standpoint; it's a deadend.
To take just one simple example: BitTorrent is kinda cool, but it'd be way cooler if it could be integrated with a pirate-bay style database and a cryptocurrency, to create a market for valued file data. That requires enormous quantity of transactions per second. Its just not going to happen on anything like a blockchain.
But that's not the biggest problem. Imagine if a country/region (say europe) actually adopted ether as their reserve currency 8 years from now. The inflation of about 10% that year would amount to about a trillion euro being spent on pointless hashes. The world is probably not that stupid.
Nor does this problem go away with PoS schemes - for the distribution problem is not technical, it is economic. For a coin to go to reserve status, it has to redistribute wealth massively - and just about any such redistribution will be an enormous economic malinvestment. PoS just redistributes it in a very concentrated fashion to the early stakeholders - who don't add any additional value thereafter - hardly ideal. To avoid that problem, you actually must have some sort of strategic plan for the economy: something to invest all that monetary energy in which results in actual return and real GDP growth.
I know this probably sounds surreal and kooky, but we need to stage a bloodless revolution.
In particular, see our reply here concerning coup-completeness.
Ethereum has great marketing, but the core of their technology design - what we actually care about - is not a huge advancement beyond Bitcoin and alternatives compatible with it (Mastercoin, Counterparty, etc). They may end up having the best practical implementation of a scriptable blockchain, or maybe just better marketing - but its still just a scriptable blockchain.
It will be able to process what - 1,000 transactions per second? 10,000 if they really compress transactions and or limit themselves to fiber nodes - maybe? The core problem is O(N) transaction cost, so performance doesn't scale as you add more nodes. That's not an improvement versus current technology, its a step back to the 80's.
Will you be able to build some sorts of applications on that? Yes, but only by dropping almost everything off the blockchain and using some side database - essentially abandoning the design - which then opens up fraud and attacks again. This just isn't interesting from a technology standpoint; it's a deadend.
To take just one simple example: BitTorrent is kinda cool, but it'd be way cooler if it could be integrated with a pirate-bay style database and a cryptocurrency, to create a market for valued file data. That requires enormous quantity of transactions per second. Its just not going to happen on anything like a blockchain.
But that's not the biggest problem. Imagine if a country/region (say europe) actually adopted ether as their reserve currency 8 years from now. The inflation of about 10% that year would amount to about a trillion euro being spent on pointless hashes. The world is probably not that stupid.
Nor does this problem go away with PoS schemes - for the distribution problem is not technical, it is economic. For a coin to go to reserve status, it has to redistribute wealth massively - and just about any such redistribution will be an enormous economic malinvestment. PoS just redistributes it in a very concentrated fashion to the early stakeholders - who don't add any additional value thereafter - hardly ideal. To avoid that problem, you actually must have some sort of strategic plan for the economy: something to invest all that monetary energy in which results in actual return and real GDP growth.
Alix, I don’t disagree with the value these types of projects can bring. However, I think you’re downplaying the ramifications of these so-called “dumb nodes” and totally missing the big picture. The protocols we use to connect hosts on a network were also comparatively simple, and yet we’ve built the entire internet on that simple foundation. Likewise for all the other web standards we take for granted. Nowhere did we start with the idea of a massively distributed resilient network, it emerged spontaneously as a consequence of simple addition. You seem to ignore the fact we’re already perfectly capable of making intelligent and even superintelligent use of new technologies.
What we didn't have prior to Ethereum was a way of organizing our existing tools into an ecosystem of decentralized consensus-making applications with real economic power behind it. All we had before was BitTorrent and cryptocurrencies; file-sharing and monetary value outside the control of governments and big business. The game is changing now. This is the first time in generations we actually have the tools to destabilize existing powers, completely and without violence, but only if we’re awake enough to seize the opportunity. So yeah, I’m fairly persuaded and morally invested in the possibility that Ethereum (or something very much like it) IS this opportunity. To my knowledge, nothing in the Ethereum code will give anyone exclusive control over the technology, as it’s entirely open-ended and democratic.
America and its imitators were merely the first Great Experiment in contractual statecraft, one dominated by a plutocratic and increasingly self-destructive elite. We can’t physically oppose these people, they have a monopoly on force, they've taken our weapons, turned us into indolent docile consumers, and left us completely despondent over the fate of the planet. It’s time to snap out of the trance they've put us under and recognize the tables have turned. If the Occupy movement taught us anything it’s that aimless protesting in the streets and faux-democracy aren't going to save us. I know this probably sounds surreal and kooky, but we need to stage a bloodless revolution. It’s our turn to re-imagine the world as we believe it should be, not as they say it must be. The future is now and we can literally do this if we all work together.
What we didn't have prior to Ethereum was a way of organizing our existing tools into an ecosystem of decentralized consensus-making applications with real economic power behind it. All we had before was BitTorrent and cryptocurrencies; file-sharing and monetary value outside the control of governments and big business. The game is changing now. This is the first time in generations we actually have the tools to destabilize existing powers, completely and without violence, but only if we’re awake enough to seize the opportunity. So yeah, I’m fairly persuaded and morally invested in the possibility that Ethereum (or something very much like it) IS this opportunity. To my knowledge, nothing in the Ethereum code will give anyone exclusive control over the technology, as it’s entirely open-ended and democratic.
America and its imitators were merely the first Great Experiment in contractual statecraft, one dominated by a plutocratic and increasingly self-destructive elite. We can’t physically oppose these people, they have a monopoly on force, they've taken our weapons, turned us into indolent docile consumers, and left us completely despondent over the fate of the planet. It’s time to snap out of the trance they've put us under and recognize the tables have turned. If the Occupy movement taught us anything it’s that aimless protesting in the streets and faux-democracy aren't going to save us. I know this probably sounds surreal and kooky, but we need to stage a bloodless revolution. It’s our turn to re-imagine the world as we believe it should be, not as they say it must be. The future is now and we can literally do this if we all work together.
EDIT: fixed changed URL link
Thanks! We agree!
Most new big technology starts out as fiction - literally in many cases, such as satellites. And AI was first conceived in fiction long before it became a field.
As for the actual science, its mostly all there, it just takes some time to read and digest. We have a more 'fun' introduction to the science which compares the brain to computing technology (with cool pictures) and has links to the important research:
http://aishare.co/wiki/Brain_Hardware_Parity
If your smarty pants friends have some specific smart criticism - we'd love to hear it.
As for the big team/funds - read more and give it time to set in, and eventually to snowball. The project certainly doesnt need massive funds/developers right away, it just needs to get an initial network started, get the coin traded, incrementally prove some of the key innovations, make some headlines, and then eventually growth is organic and can continue as awareness spreads.
Holy macarole this is exciting !
Thanks! We agree!
Quote
EDIT : I talked to some friends more intelligent then myself and they called this "fiction".
Until you have some actual science to back this up and a big team of dedicated programmers, and a lot of funds.... you guys have no chance.
Until you have some actual science to back this up and a big team of dedicated programmers, and a lot of funds.... you guys have no chance.
Most new big technology starts out as fiction - literally in many cases, such as satellites. And AI was first conceived in fiction long before it became a field.
As for the actual science, its mostly all there, it just takes some time to read and digest. We have a more 'fun' introduction to the science which compares the brain to computing technology (with cool pictures) and has links to the important research:
http://aishare.co/wiki/Brain_Hardware_Parity
If your smarty pants friends have some specific smart criticism - we'd love to hear it.
As for the big team/funds - read more and give it time to set in, and eventually to snowball. The project certainly doesnt need massive funds/developers right away, it just needs to get an initial network started, get the coin traded, incrementally prove some of the key innovations, make some headlines, and then eventually growth is organic and can continue as awareness spreads.
Nexus 9:
We understand you are a believer in Ethereum and are invested in that project (at least emotionally), and we sympathesize with the decentralization movement and thus have goals in common, however . . .
We are somewhat doubtful of both premises. Our focus on scalability and performance is motivated by a very different vision for the use and applications of the network. Ethereum - as we understand it - is mainly focused on building decentralized Web applications. Aionic is focused on enabling new applications of AI, deep learning research, and so on: a very different immediate niche. There is naturally overlap - both designs encompass programmable smart contracts - but the visions are very different.
Like it or not, Ethereum started without a solution to scalability/efficiency. To his credit, Vitalik Buterin honestly admitted this from the beginning.
Our vision/plan for Aionic begins with a fully scalable design (driven by the specific needs of machine learning) - and it turns out that this requires a radically different foundation; it can't be based on a blockchain, even the principles and philosophy are quite different. Scalability necessarily implies a partially observable environment, and that is exactly the situation where traditional computer science techniques break down and you move into the realm of AI and reasoning with uncertainty.
No - but it is reasonable to expect a strong foundation that at least points in the right eventual direction. What hard problems exactly did ethereum solve from the beginning? Do you consider adding a scripting language for contracts solving a hard core problem?
So basically you are saying : "There can only be ethereum, even if we didn't solve any of the hard problems to start out with, ah sorry but we are the monopoly now." Well sorry, but this is market evolution.
From the perspective of AI research, Ethereum is a non factor - not even on the path.
We are essentially making a bet that at least general AI is possible in less than ten years (software intelligence better at humans at programming, science, etc. that runs on a single machine with a couple GPUs and thus costs < $1,000 a year). What do you think the probability of that is? Do you have a pretty deep understanding of machine learning and neuroscience? So pick some probability for that scenario.
Now, conditional on the above, what is the market value of general AI? It's close to the market value of the entire economy : we are talking trillions to quadrillions. It's technically more than that, because general AI will rapidly increase GDP - making the entire economy more valuable.
Web X.0 whatever doesn't lead to general AI.
We understand you are a believer in Ethereum and are invested in that project (at least emotionally), and we sympathesize with the decentralization movement and thus have goals in common, however . . .
I understand your concerns over scalability and computational efficiency, but this is something that will work itself out in future developments of the Ethereum platform.
We are somewhat doubtful of both premises. Our focus on scalability and performance is motivated by a very different vision for the use and applications of the network. Ethereum - as we understand it - is mainly focused on building decentralized Web applications. Aionic is focused on enabling new applications of AI, deep learning research, and so on: a very different immediate niche. There is naturally overlap - both designs encompass programmable smart contracts - but the visions are very different.
Like it or not, Ethereum started without a solution to scalability/efficiency. To his credit, Vitalik Buterin honestly admitted this from the beginning.
Our vision/plan for Aionic begins with a fully scalable design (driven by the specific needs of machine learning) - and it turns out that this requires a radically different foundation; it can't be based on a blockchain, even the principles and philosophy are quite different. Scalability necessarily implies a partially observable environment, and that is exactly the situation where traditional computer science techniques break down and you move into the realm of AI and reasoning with uncertainty.
Quote
We can’t expect to have every problem worked out from the beginning.
No - but it is reasonable to expect a strong foundation that at least points in the right eventual direction. What hard problems exactly did ethereum solve from the beginning? Do you consider adding a scripting language for contracts solving a hard core problem?
Quote
Splitting our attention into so many different currencies and projects ..
So basically you are saying : "There can only be ethereum, even if we didn't solve any of the hard problems to start out with, ah sorry but we are the monopoly now." Well sorry, but this is market evolution.
Quote
Ethereum is the next logical step on that path.
From the perspective of AI research, Ethereum is a non factor - not even on the path.
Quote
Crowdfunding superintelligence, on the other hand, while highly desirable is not immediately feasible, nor is it something we can all give universal support. I have no doubt what you are hoping for is going to happen, but it is far in the future, perhaps web 5.0 or 6.0. If you want to dedicate your life to something that is potentially decades away, fair enough and good luck to you. Personally, I think the singularity is already happening, it begins with the open society Ethereum makes possible.
We are essentially making a bet that at least general AI is possible in less than ten years (software intelligence better at humans at programming, science, etc. that runs on a single machine with a couple GPUs and thus costs < $1,000 a year). What do you think the probability of that is? Do you have a pretty deep understanding of machine learning and neuroscience? So pick some probability for that scenario.
Now, conditional on the above, what is the market value of general AI? It's close to the market value of the entire economy : we are talking trillions to quadrillions. It's technically more than that, because general AI will rapidly increase GDP - making the entire economy more valuable.
Web X.0 whatever doesn't lead to general AI.
Jump to: