Artificial Intelligence - The Final Frontier - Machine Learning Our Future

[Music] artificial intelligence the final frontier the AI singularity is a future period which technological change will be so rapid and its impact so profound that every aspect of human life will be irreversibly transformed there won't be a clear distinction between humans and machines computers are not going to be these rectangular devices we've put in our pocket they're going to be inside our bodies and brains and we're going to be a hybrid of biological and non-biological intelligence if we go back 500 years not much happened in a century now a lot happens in six months technology feeds on itself and gets faster and faster in about 40 years from now the pace of change is going to be so astonishingly quick that you won't be able to follow it unless you enhance your own intelligence by merging with the intelligent technology we're creating it's such a profound transformation that we've borrowed this metaphor from physics and called that event in human history a singularity people routinely underestimate what's achievable in long periods of time like one decade or two decades because they leave out the radical implications of exponential growth we can already sense how much more change occurs in a year than in the years before then if you speak to young people teenagers and even in their lifetimes they can see how much more quickly technology moves today than it did five years ago acceleration of technology is the implication of what we call the law of accelerating returns the nature of technological progress is exponential if I count linearly 30 steps one two three four five I get to 30 if I count exponentially two four eight sixteen then thirty steps later I'm at a billion it makes a dramatic difference forty years ago Gordon Moore saw that there was exponential growth in the power of semiconductors and that inside microchips basically every two years we can put twice as much components on a chip and because they're closer together they run faster and so computers get twice as capable overall for the same prize every year we'll make another billion fold increase in performance in the next 25 years shrink the size of these technologies a hundred thousand fold so we went from a computer building to something to fit in your pocket in 40 years and the next 25 years will go from something that fits in your pocket to something that's the size of a blood cell the reason that information technology grows exponentially is that we use the latest technology to create the next now each generation of technology grows exponentially in capability and the speed of that process accelerates over time this is true in general of the evolutionary process in fact even biological evolution long before even humans evolved chose the same phenomenon the very first paradigm the biological evolution was the evolution of DNA that took a billion years but then evolution adopted it and it's used it ever since so the next stage the Cambrian explosion when all the body plans of the animals evolved went a hundred times faster then it took ten million years after a few more steps Homo sapiens the first technology creating species evolved not only to a few hundred thousand years evolution then shifted from biological evolution to technological evolution took tens of thousands of years to evolve stone tools fire the wheel and then we always use the latest technology to create the next technology so the whole pace of technology has accelerated major paradigm shifts like searching evolved in just five or six years the reason we get to a singularity the point of astonishingly quick change is because it's going to go into hyperspeed over the next several decades many believe we can have a positive singularity within ten years by 10 years the precise numbers are not important it could be eight years or it could be twelve years but something on that order of magnitude is achievable none of us can know the future but it's a plausible hypothesis and an achievable goal ten years to a positive singularity if the right amount of effort is expended in the right direction if we really really tried the path to singularity via artificial intelligence can lead us to a positive singularity but there's a lot of other interesting work going on in working molecular nanotechnology which will allow us to reconfigure matter according to our well brain computer interfaces which will let us hook up our brains into computers radically enhancing our power of cognition any one of these paths could lead to a positive singularity way way sooner than most people think but the sad thing is the amount of resources the amount of energy that our society devotes to these things is very very small creating new forms of matter improving human cognition extending human life ending scarcity and inhuman suffering creating advanced artificial Minds this is what we really need to give to achieve our all positive singularity and it may well be achievable on the order of 10 years from now most people understand just how quickly machine intelligence is advancing it's much faster than almost anyone realizes even with in Silicon Valley and certainly outside Silicon Valley people really have no idea if there's a super intelligence particularly if it's engaged in recursive self-improvement or if there's some digital super super intelligence and it's optimization or utility function is something that's detrimental to humanity then it will have a very bad effect you know it could just be something like getting rid of spam email or something in the computer decides well the best way to get over spam is to get rid of humans this scenario has been played out in numerous movies do we get to an asymmetric situation where technology is so powerful that it extends beyond nation state it's not the nation states that have potential access to mass destruction but the individuals and this is a consequence of the fact that these new technologies tend to be digital we saw genome sequences where you can download the gene sequences of pathogens off the internet individuals and small groups super empowered by access to these kinds of self-replicating technologies whether it be biological or other are clearly a danger in our world to limit the danger of these new things we have to limit ultimately the ability of individuals to have access essentially - pandemic power we also have to have a sensible defense because no limitation is going to prevent a crazy person from doing something but the troubling thing is that it's much easier to do something bad than to defend against all possible bad things the offensive uses really have any symmetric advantage I would say that you know we can't give up the rule of law to fight an asymmetric threat we have incredibly powerful computers but we don't have very good software for long and it's only in retrospect after the better software comes along you take it and you run it on a ten-year-old machine and say god the machine was that fast or when they took the Apple Mac in her face and they put it back on the Apple to the Apple it was perfectly capable of running that kind of interface we just didn't know how to do it at the time if we take carbon nanotubes recently discovered in 1991 they just have incredible properties and these are the kinds of things we're going to discover as we start to engineer at the nano scale their strengths are almost the strongest material of tensile strength material known they're very very stiff they stretch very very little in two dimensions if you make a fabric out of them there are 30 times stronger than Kevlar and if you make a three-dimensional structure like a Bucky ball they have all sorts of incredible properties if you shoot a particle at them and knock a hole in them they repair themselves they go with it and they repair the hole in femtoseconds which is super quick if you shine light on them they produce electricity in fact we flashed them with a camera they catch on fire if you put electricity on them they emit light if you turn through them you can run a thousand times more current on one of these than through a piece of metal you can make both P and n-type semiconductors which means you can make transistors out of them they conduct heat along their length but not across their other direction if you put particles in them and they shoot out to the tip they're like miniature linear accelerators the inside of the nanotubes is so small the smallest ones are 0.7 nanometers it's basically a quantum world what we see is with these and other new materials that we can do things with different properties lighter couple was stronger and apply these new materials to the environmental problems new materials that can make fuel cells work better new materials that catalyze chemical reactions that cut pollution and so on new ways of making ethanol new ways of making electric transportation the whole green dream because it can be profitable there are so many great technologies out there so if we can address and use technology to tackle education help address the environment does that solve the larger problem no because you can't solve a problem with the management technology with more technology if we let an unlimited amount of power loose then a very small number of people will be able to abuse it we can't fight it if it gets to a million to one disadvantage how could you keep the law I think the law would be a really good thing to keep well you have to hold people accountable the law requires accountability today scientists technology and technologists businessmen engineers don't take any personal responsibility for the consequences of their actions so if you tie that back with the law we have to do something we can't pick the future but we can steer the future our investment in trying to prevent pandemic flu is affecting the distribution of possible outcomes we may not be able to stop it but the likelihood that it'll get past us is lower if we focus on problem so we can design the future if we choose what kind of things we'd want to have happen above all we have to help the good guys the people on the defensive side have an advantage over the people who want to abuse things and we have to limit access to certain information and growing up as we have and holding the very high the value of free speech it's a hard thing for us to accept for all of us to accept it's especially hard for the scientists to extend to still remember Galileo essentially was locked up and who are still fighting this battle against the church but that's the price of having a civilization of the price of retaining the rule of law which is to limit the access to the great and the kind of unbridled power there will be new dangers from these new technologies many are convinced that we won't encounter painful episodes over all will be helped more than we think you don't have to look at the 20th century and say we had 180 million people died in the wars of the 20th century that scale of destruction was made possible by technology we really must mature when we think about these matters we have come a long way to reach the dawn of artificial intelligence the term artificial intelligence was coined in 1956 but AI has become more popular today thanks to increased data volumes advanced algorithms and improvements in computing power and storage early AI research in the 1950s explored topics like problem-solving and symbolic methods in the 1960s the US Department of Defense took interest in this type of work and began training computers to mimic basic human reasoning for example the Defense Advanced Research Projects Agency DARPA completed Street mapping projects in the 1970s and DARPA produced intelligent personal assistants in 2003 long before Siri Alexa or Cortana were household names this early work paved the way for the automation and formal reasoning that we see in computers today including decision support systems and smart search systems that can be designed to complement and augment human abilities while Hollywood movies and science fiction novels depict AI as human-like robots that take over the world the current evolution of AI technologies isn't that scary or quite that smart instead a AI has evolved to provide many specific benefits in every industry the theoretical idea of AI or more specifically neural networks has been around longer than you might believe the first neural network was conceived of by Warren McCulloch and Walter Pitts in 1943 they wrote a seminal paper on how neurons may work and modeled their ideas by creating a simple neural network using electrical circuits this breakthrough model paved the way for neural network research in two areas biological processes in the brain the application of neural networks to artificial intelligence AI AI research quickly accelerated with kunihiko fukushima developing the first true multi-layered neural network in 1975 the original goal of the neural network approach was to create a computational system that could solve problems like a human brain however over time researchers shifted their focus to using neural networks to match specific tasks leading to deviations from a strictly biological approach since then neural networks have supported diverse tasks including computer vision speech recognition machine translation social network filtering playing board and video games and medical diagnosis as structured and unstructured data sizes increase to big data levels people developed deep learning systems which are essentially neural networks with many layers deep learning enables the capture and mining of more and bigger data including unstructured data why are neural networks so important neural networks are also ideally suited to help people solve complex problems in real-life situations they can learn and model the relation ships between inputs and outputs that are nonlinear and complex make generalizations and inferences reveal hidden relationships patterns and predictions and model highly volatile data such as financial time series data and variances needed to predict rare events such as fraud detection as a result neural networks can improve decision processes in areas such as credit card and Medicare fraud detection optimization of logistics for transportation networks character and voice recognition also known as natural language processing medical and disease diagnosis targeted marketing financial predictions for stock prices currency options futures bankruptcy and bond ratings robotic control systems electrical load and energy demand forecasting process and quality control chemical compound identification ecosystem evaluation computer vision to interpret raw photos and videos for example in medical imaging and robotics and facial recognition the terminology machine learning is pretty much well known now it's a method of data analysis that automates analytical model building it is a branch of artificial intelligence based on the idea that systems can learn from data identify patterns and make decisions with minimal human intervention because of new computing technologies machine learning today is not like machine learning of the past it was born from pattern recognition and the theory that computers can learn without being programmed to perform specific tasks researchers interested in artificial intelligence wanted to see if computers could learn from data the interative aspect of machine learning is important because as models are exposed to new data they are able to independently adapt they learn from previous computations to produce reliable repeatable decision and results it's a science that's not new but one that has gained fresh momentum while many machine learning algorithms have been around for a long time the ability to automatically apply complex mathematical calculations to big data over and over faster and faster is a recent development researching interest in machine learning is due to the same factors that have made data mining and Bayesian analysis more popular than ever things like growing volumes and varieties of available data computational processing that is cheaper and more powerful and affordable data storage all of these things mean it's possible to quickly and automatically produce models that can analyze bigger more complex data and deliver faster more accurate results even on a large scale and by building precise models an organization has a better chance of identifying profitable opportunities or avoiding unknown risks the hype around artificial intelligence is pretty much centered on the next step in the evolution namely deep learning deep learning is the type of machine learning that trains a computer to perform human-like tasks such as recognizing speech identifying images or making predictions instead of organizing data to run through predefined equations deep learning sets up basic parameters about the data and trains the computer to learn on its own by recognizing patterns using many layers of processing deep learning techniques have improved the ability to classify recognize detect and describe in one word understand for example deep learning is used to classify images recognize speech detect objects and describe content systems such as Siri and Cortana are powered in part by deep learning several developments are now advancing deep learning algorithm ik improvements have boosted the performance of deep learning methods new machine learning approaches have improved accuracy of models new classes of neural networks have been developed that fit well for applications like text translation and image classification we have a lot more data available to build neural networks with many deep layers including streaming data from the Internet of Things textual data from social media physicians notes and investigative transcripts computational advances of distributed cloud computing and graphics processing units have put incredible computing power at our disposal this level of computing power is necessary to Train deep algorithms at the same time human to machine interfaces have evolved greatly as well the mouse and the keyboard are being replaced with gesture swipe touch and natural language assuring in a renewed interest in AI and deep learning everything we love about civilization is a product of intelligence so amplifying our human intelligence with artificial intelligence has the potential of helping civilization flourish like never before as long as we manage to keep the technology beneficial Stephen Hawking Elon Musk Steve Wozniak Bill Gates and many other big names in science and technology have recently expressed concern in the media and via open letters about the risks posed by AI joined by many leading AI researchers why is the subject suddenly in the headlines the idea that the quest for strong AI would ultimately succeed was long thought of as science fiction centuries or more away however thanks to recent breakthroughs many AI milestones which experts viewed as decades away merely five years ago have now been reached making experts take seriously the possibility of super intelligence in our lifetime while some experts still guessed that human-level AI is centuries away most a AI researchers at the 2015 Puerto Rico conference guessed it would happen before 2060 since it may take decades to complete the required safety research it's prudent to start it now because AI has the potential to become more intelligent than any human we have no surefire way of predicting how it will behave we can't use past technological developments as much a basis because we've never created anything that has the ability to wittingly or unwittingly outsmart us the best example of what we could face may be our own evolution people now control the planet not because we're the strongest fastest or biggest but because we're the smartest if we're no longer the smartest are we assured to remain in control FL eyes position is that our civilization will flourish as long as we win the race between the growing power of technology and the wisdom with which we manage it in the case of AI technology FL eyes position is that the best way to win that race is not to impede the former but to accelerate the latter by supporting AI safety research digital life is augmenting human capacities and disrupting eons old human activities code driven systems have spread to more than half of the world's inhabitants in ambient information and connectivity offering previously unimaginative and unprecedented threats as emerging algorithm driven artificial intelligence continues to spread will people be better off than they are today some 979 technology pioneers innovators developers business and policy leaders researchers and activists answered this question in a canvassing of experts conducted in the summer of 2018 the experts predicted networked artificial intelligence will amplify human effectiveness but also threaten human autonomy agency and capabilities they spoke of the wide-ranging possibilities that computers might match or even exceed human intelligence and capabilities on tasks such as complex decision-making reasoning and learning a sophisticated analytics and pattern recognition visual acuity speech recognition and language translation they said smart systems and communities in vehicles and buildings and utilities on farms and in business processes will save time money and lives and offer opportunities for individuals to enjoy a more customized future many focus they're optimistic remarks on health care in the many possible applications of AI in diagnosing and treating patients or helping senior citizens live fuller in healthier lives they were also enthusiastic about a eyes role in contributing to broad public health programs belt around massive amounts of data that may be captured in the coming years about everything from personal genomes to nutrition additionally a number of these experts predicted that a I would have bet long anticipated changes in formal and informal education systems yet most experts regardless of whether they are optimistic or not expressed concerns about the long-term impact of these new tools on the essential elements of being human all respondents in this non-scientific canvasing were asked to elaborate on why they felt AI would leave people better off or not many shared deep worries and many also suggested pathways towards solutions creativity may be the ultimate moonshot for artificial intelligence already AI has helped write pop ballads mimic the styles of great painters and informed creative decisions in filmmaking experts wonder however how far a I can or should go in the creative process experts contend that we've barely scratched the surface of what's possible while advancements in AI mean that computers can be coached on some parameters of creativity experts question the extent to which AI can develop its own sense of creativity can AI be taught how to create without guidance can it truly understand what is beautiful perhaps by looking at pixel arrangements or color palettes experts point out that teaching computers to be creative is inherently different from the way humans learn to create although there's still much we don't yet know about our own creative methodology many examples of creativity involve learning and exploring in a hierarchical style neural and multi-layer network systems can help us construct different frameworks to better understand those hierarchies but there's much more to learn and discover if you have a computer that comes up with random com of musical notes a human being who has sufficient insight and time could well pick up an idea or two a gifted artist on the other hand might hear the same random compilation and come away with a completely novel idea one that sparks a totally new form of composition 95% of what professional artists and scientists do is exploratory and perhaps the other 5% is truly transformational creativity a lot of the processes behind creative thinking is still unknown and AI has a big role to play here thought leaders ponder whether AI innovation will ultimately yield technology that can create without supervision or direction but the bigger question remains should this be the goal of AI even if it is technically feasible it's not our goal to recreate the human mind that's not what we're trying to do what we're more interested in are the techniques of interacting with humans that inspire creativity in humans that requires that we spend time thinking about that creative process what do we do to help people come up with new ideas on a much more regular basis than they do today why does artificial intelligence scare us so much then when people see machines that respond like humans or computers that perform feats of strategy and cognition mimicking human ingenuity they sometimes joke about a future in which humanity will need to accept robot overlords but buried in the joke is a seed of unease science fiction writing and popular movies from 2001 a Space Odyssey 1968 to Avengers age of Ultron 2015 have speculated about artificial intelligence AI that exceeds the expectations of its creators and escapes their control eventually out competing and enslaving humans or targeting them for extinction conflict between humans and AI is front and center in AMC's sci-fi series humans in the new episodes conscious synthetic humans face hostile people who treat them with suspicion fear and hatred violence roles as scents find themselves fighting for not only basic rights but they're very survival against those who view them as less than human and as a dangerous threat even in the real world not everyone is ready to welcome AI with open arms in recent years as computer scientists have pushed the boundaries of what AI can accomplish leading figures in technology and science have warned about the looming dangers that artificial intelligence may pose to humanity even suggesting that AI capabilities could doom humanity but why are people so unnerved by the idea of AI is it really an existential threat Elon Musk is one of the prominent voices that has raised red flags about AI in July 2017 musk told attendees at a meeting of the National Governors Association I have exposure to the very cutting-edge AI and I think people should be really concerned about it I keep sounding the alarm bell says Elon Musk but until people see robots going down the street killing people they don't know how to react because it seems so ethereal earlier in 2014 must get labeled AI our biggest existential threat and in August 2017 he declared that humanity faced a greater risk from AI than from North Korea physicist Stephen Hawking who died March 14th also expressed concerns about malevolent AI telling the BBC in 2014 that the development of full artificial intelligence could spell the end of the human race it's also less than reassuring that some programmers particularly those with MIT media lab in Cambridge Massachusetts seem determined to prove that AI can be terrifying a neural network called nightmare machine introduced by MIT computer scientists in 2016 transformed ordinary photos into ghoulish unsettling health games an AI that the MIT group dubbed shellye composed scary stories trained on 140,000 hours of horror that reddit users posted in the forum negative feelings about AI can generally be divided into two categories the idea that AI will become conscious and seek to destroy us and the notion that immoral people will use AI for evil purposes one thing that people are afraid of is that if super intelligent AI more intelligent than us becomes conscious it could treat us like lower beings like we treat monkeys he said that would certainly be undesirable however fears that AI will develop awareness and overthrow humanity are grounded in misconceptions of what AI is Weinberger noted AI operates under very specific limitations defined by the algorithms that dictate its behavior some types of problems map well to AI skill sets making certain tasks relatively easy for AI to complete this means that while a I might be capable of impressive feats with carefully delineated boundaries playing a master level of chess game or rapidly identifying objects and images for example that's where its abilities end the other worrisome idea that an unscrupulous human would harness AI for harmful reasons is unfortunately far more likely Weinberger added pretty much any type of machine or tool can be used for either good or bad purposes depending on the user's intent and the prospect of weapons harnessing artificial intelligence is certainly frightening and would benefit from strict government regulation perhaps if people could put aside their fears of hostile AI they would be more open to recognizing its benefits Weinberger suggested enhanced image recognition algorithms for example could help dermatologists identify moles that are potentially cancerous while self-driving cars could one day reduce the number of deaths from auto accidents many of which are caused by human error Elon Musk wants the US government to spend a year to understanding the problems before they consider how to solve it his recommendation for the longest time has been consistent saying I think we ought to have a government committee that starts off with insight gaining insight spends a year or two gaining insight about AI or other technologies that are maybe dangerous but especially AI and then based on that insight comes up with rules in consultation with industry that give the highest probability for a safe advent of AI from musts perspective here's what's going on researchers especially at alphabet google deepmind the AI research organization that developed alphago and alpha0 are eagerly working towards complex and powerful AI systems since some people aren't convinced that AI is dangerous they're not holding the organization's working on it too high enough standards of accountability and caution Elon Musk's AI concerns are not an out of character streak of technological pessimism they stem from optimism the people who expect AI to make the biggest splash who've concluded that working to get ahead of it should be one of our urgent priorities [Music] [Music] [Music] [Music]

Channel: Viper TV - Science

Views: 20,100

Rating: 4.5357141 out of 5

Keywords: documentary, artificial intelligence, robots, machine learning, science, technology, deep learning, neural networks, internet, cloud computing

Id: QwzsoG_lSxs

Length: 34min 47sec (2087 seconds)

Published: Thu Dec 05 2019