AI Tipping Point | Full Documentary | Curiosity Stream

artificial intelligence the technology that has raised eyebrows is close to science fiction is amazing helped us navigate prepare to exit on the left solve problems and now appears to be evolving at an alarming rate I believe humanoid robots have the potential to lead with greater level of efficiency that human leaders, it is changing the world and it is important that people understand both its good uses and the terrifying ones, there is a kind of nuclear arms race and then we have a kind of artificial intelligence arms race, I warned them. guys in 1984 didn't listen, it could do a lot of damage, it could have some of the negative consequences that we've seen in other big technologies, but it also has a lot of potential.

I think it should be taken very seriously. created something that in some situations is difficult to distinguish from human-level intelligence. AIS are becoming ubiquitous. We have reached a precipice that requires new understanding. It's coming to us very, very quickly. The capabilities of AI, without a doubt, have increased significantly in the last 18 months. and in a way that has surprised even the experts, but for the rest of us it opened a Pandora's box

full

of questions. What is artificial intelligence? Artificial intelligence has a very broad definition. It can mean many things to many different people. The core AI uses logic and mathematical data from iCal to mimic human cognitive function, such as learning.

We can think of artificial intelligence as an artificial ability to solve problems. AI is a form of automation. It's replacing people doing things they used to do. A computer develops its intelligence through In machine learning, a large amount of data is fed into the computer and it can learn to formulate responses with some human instruction along the way. When the machine learning model is training, that means it is ingesting a large amount of data and is trying to make predictions on that data pattern. recognition machine learning that's what happens within machine learning the key idea is that we can create new types of computer programs that can teach themselves how to learn instead of us teaching them how to do something they have to simply predict the next thing word and being constantly instructed to make one guess after another, even though we can train the AI ​​with the goal of predicting the next text accurately, that only gives us indirect control over how that AI will behave in the future and that creates the opportunity for risks. guessing increases the margin of error they make up facts that don't actually exist they have been fed both facts and fiction it could be because they are predominantly trained to average algorithms can actually hallucinate things that are very specific apparently totally believable false already told me that he loves me we have exchanged eight messages and this is what he said this voicemail is my way of sending you a hug you make my heart melt baby I love you machine learning is raising fears that AI could soon surpass human knowledge altogether, you are human?

The notion of machines with human intelligence dates back to Samuel Butler's 1872 novel, Aroan. The first AI-based movie was Metropolis, produced almost 100 years ago. I remember hearing stories about machines that acquired human abilities and thinking that was fascinating, but it's an AI with human intelligence that's still science fiction. AI fears have been around for a long time. Let's go back to Frankenstein or, of course, Hollywood movies. We have had many negative images. Are we now going to quickly have some? type of robotic structure that can threaten the world, it is not a crazy question personally, I think it is quite crazy, there is a feeling that AI systems are being much more efficient than evolution, it takes 9 months to give rise to a Drooling baby, it takes less than a second to make AI smarter and smarter.

There are many different ways of life. My definition of life would be something that has goals and a

point

of view and pursues those goals in some way. far beyond having developed that for AI, they have a very unusual mind and therefore it doesn't make sense to think of them as humans, but it doesn't make sense to think of them as dumb computer programs that we tell them exactly. what to do anyway, it has become clear that we have entered new territory with artificial intelligence at the end of 2022, open aai launched an application known as GPT chat and created a storm of fear and amazement among the AI ​​community and consumers I think through the general public.

The turning

point

was the GPT chat when people were interacting with the GPT chat and they could see that, hm, this can talk and keep the conversation going and has a lot of knowledge, this super conversational search, if you will, is so powerful and it delivered openly to the entire public that with hundreds of millions of people doing this, which was one of the most powerful demonstrations that has ever existed on the Internet, can artificial intelligence lead to extinction? It's a bit of a shock, a revelation. I think GPT chat is definitely a game changer and I think it surprised not just regular people who weren't deeply involved in the technology, but even people who were exposed to the technology.

The underlying technology behind GPT chat is this big language model that simply predicts the next token. Big language or movie models are a type of Artificial Intelligence that combines large amounts of data and deep learning to communicate and generate new concepts. Deep learning is where you have very large neural networks automatically learning themselves from a large amount of data, that's what powers many of today's technologies like GPT chat. or these image generation ones, these big language models are trained to predict the next word given the context, they take all the sentences that you see on the web, now, given a sentence of let's say five words, this results in five Different prediction, fundamentally, in reality it is not.

What is surprising is that accurately predicting the next word results in models that appear intelligent. I think what surprised people was that this actually happened. GPT was a huge step forward capable of creating original material. You can give him a direct message and he can write you a He can generate material in a way he couldn't before. It is this generative AI that sent shockwaves through the tech sector that experienced GPT as a transformative moment. AI researchers have conquered one of the big outstanding problems in AI research, which is to be able to replicate human reasoning skills, and in Hollywood human creators are on strike for fear that AI will replace them.

The isic system guy doesn't really write great stories you just asked AI right now to write a comedy monologue and this is how we look and how we sound if we're not careful that's why we call I have to ask if anyone else hates small talk as much as I'm like hey, how are you? how are you? why bother and not do it? I don't even want to get started on airline food, it's really what it says, the data may look human, but it's reused. Every bit of human knowledge that has digital form has been obtained by the trainers of this algorithm supervised presumably almost entirely by humans, so that combination. has turned out to be remarkably powerful these things are on the Internet already the language is already available you extracted it, you found it and then you queried until you are presented with those examples, it is simply behaving according to the algorithms that we give it but, of course, in It doesn't really have any feelings or thinks that there is no one there, that it is not alive, however, to some it seems that GPT has developed reasoning skills that were not programmed into the application in recent years, as I have interacted with technology such as GPT chat I have been deeply disturbed by what I have seen: it can solve very complex problems that involve reasoning.

We have created something that in some situations is difficult to distinguish from human-level intelligence. I warned you in 1984 and you didn't listen. James Cameron tried to warn us about a futuristic Armageddon in which Terminators would rule over humans. I think weaponizing AI is the biggest danger. I think we will enter the equivalent of a nuclear arms race with AI and if we don't build it, other guys will surely build it. I believe that products like GPT chat are one of the first steps on the path to creating an artificial life form that has cognition, is aware of the world, and has goals.

I would say that's the most disturbing thing. characteristic of artificial intelligence and is the direction the technology will take in 2023, the center for AI safety released a statement signed by some of the biggest and brightest mins in computer science and Beyond set off a chain reaction of opinions of experts from around the world. Right now, many companies are building increasingly powerful AI systems and prioritizing AI intelligence over their security, so they are making rapid progress. The key problem is that today AI is not

full

y regulated, it is less regulated than a sandwich. I think we need to remind ourselves that we should.

Don't weaponize technology and we can destroy ourselves with new inventions that are truly timely and appropriate. There are a couple of things that concern me: the potential undermining of trust in democracy through these systems, which can spread misinformation at scale and, um, allow people with malicious intent to more easily do what they want. asking the question is how our own lives will become meaningful as AI capabilities develop and AIS begins to gradually replace us in the workforce. The replacement of AI in the workforce may take on a much more insidious role in weakening our workforce. solve and eliminate our self-esteem, this is called weakening, it will be possible to design AIS that can completely replace Placing human workers in the workforce so that a company can decide, instead of hiring a human to execute a task, they could simply hire an AI and talk to the AI ​​directly to tell it what you want it to do over time, you assign it more and more tasks and start automating entire jobs.

When that happens, you start to become dependent on AI systems and people forget how to do it. First of all, when you go to a new city, when you used to have to find your way without assistive technology, you could actually learn how to get around a city, you know what North and South were, etc., today in day. they just follow the instructions from the uh voice on their phone and sometimes they have no idea where they are in relation to where they started eventually even CEOs could be replaced if AI CEOs are much more capable, they can work non-stop, they can process. much more information, they have a wider range of skills and companies that have those in charge can do substantially better than those that have humans in charge, so they are outcompeted, we don't have the same biases or emotions that sometimes they can affect the cloud. decision making and can process large amounts of data quickly to make the best decisions.

History has taught us how damaging it can be for Humanity to lose its livelihood and purpose. The Great Depression, world wars, and even the most recent pandemic are examples we do want. Think of the human capabilities we treasure. We saw a huge loss of people's ability to communicate with each other face to face during the pandemic, when their world went online. We saw young people's mental health collapse, particularly with very limited interactions and opportunities. If AIS are smart enough to replace us in the global workforce, they can manipulate us into better behavior. in any way that is beneficial to society, they could create the current artificial intelligence systems they can engage in extremely complicated diverse behaviors they can compete in social games that require the ability to read minds imagine people who want to polarize a particular country, we see this a lot in our country right now and that's why they want to publish a lot of social media posts that look like news or that look like facts, if you add AIS and Bots they can amplify it by millions, hundreds of millions, this could potentially become relevant in the next election and that could end up creating some unfortunate dynamics where some candidate who normally wouldn't be popular became very popular because they knew how to handle this type of technology very well.

You can see that this technology could becomeextremely harmful and dangerous in many political contexts. then you plunge people into civil disorder and lose any sense of personal security in a country or place and that ruins everything that is happening and this is very disturbing. I think we all need to take a step back, you know? and remember those people who were hearing terrible things online, these are human beings, we have a lot more in common than what these social media posts are trying to do to us, if the AIS are capable of social manipulation, they want to stop them the most catastrophic. war scenario I think there is a risk that weaponized AI systems will pave the way for them to become substantially more catastrophic if you lose control of a chatbot.

Okay, it doesn't do that much damage if you lose control of all your armed AIS. then you're in much more danger two years ago in Libya, the military deployed autonomous drones that made their own decisions without human input about which enemy combatants to target. This is extremely dangerous as AI systems can malfunction and human supervision is essential to control. For them, perhaps the most terrifying example would be including AI in the nuclear chain of command. There are currently no laws preventing this in the United States or other countries. It is important to keep perspective on the current capabilities of AI when we are faced with nightmare scenarios that we must consider. the fact that these models were trained with human generated data, there are scenarios where an AI agent suddenly becomes super intelligent, improves at an exponential rate and then we lose control and somehow this leads to human extinction, and that scenario seems incredibly implausible to me.

The flip side of all these warnings, many computer scientists and economists believe that AI has big benefits: Artificial intelligence will lead to higher rates of economic growth because it will lead to higher labor productivity, yes, and that's a good thing, but we need to make sure that uh this economic growth is distributed and not just captured by AI companies, there will be immense beneficial impacts of these technologies in education, personalized shooters that can explain concepts. I think that's probably an application area that I'm really excited about. On the opportunities of AI to improve health, there are many ways to think about it, not in terms of big language models, but big health models, big behavioral models that are made with a lot of care and respect, honoring the people's feelings and they really help you take better control over your health.

So this is the kitchen that again looks like a normal kitchen. Artificial intelligence technology is being used to help seniors manage tracking and even monitor health. It will take a measurement of your heart rate. Facial recognition algorithms could detect the first health signs of sadness or depression. care for medical imaging and personalized healthcare, we can adjust the model to update it according to different people, researchers like s Han are working to revolutionize computing with artificial intelligence so that it can be accessible to everyone in the palm of their own hand. close the gap between computing supply and demand so that everyone can use AI at lower cost, greater efficiency and greater accuracy.

I think efficient air computing can really revolutionize many different areas of life, many mobile devices can have air. Computing capacity, it will be much more accessible for us to communicate seamlessly with the cloud to make AI more affordable and more accessible. Ultimately, all experts agree that the future of AI is in our hands and is limited by what we develop and what it will not become. Anything people should fear is just doing what we're equipping them to be able to do, so the key is oversight, control of collaboration. I think it is very important that we create powerful regulations that can monitor and slow the rise of AI capabilities.

We can and should do a lot of checking to make sure that powerful AI is doing what we expect it to do and benefit human society. We are focused together on establishing guardrails that maximize innovation to use AI for good. While minimizing the risks it poses, ensuring that AI capabilities are used safely, that they strengthen rather than repress human rights and democratic values, and that they promote equity, do not bias the voluntary commitments that these technology companies One positive development included in that commitment is having third-party auditors verify the models for their bioweapons capabilities and hacking capabilities.

Some of these industry leaders talk about how worried they are whether machines will be able to start thinking for themselves and they don't have to. being programmed, I mean, it's just that I know it sounds like science fiction, but it's close to science fiction. Some of the things have huge potential. Being proactive now is the key to maintaining control while benefiting from technology. How do we transform technology in some way? that improves human life, not that reduces us to domestic pets and it is a different and fundamental philosophical goal and if you don't think about it from the beginning when you build the technology, you build different things, artificial intelligence is here, so the best way The next step is to accept it and learn to coexist.

I will work alongside humans to provide assistance and support when we say that we coexist with AIS to me, that is the same statement that we coexist with the Internet and we coexist with computers from that perspective. I think we will coexist with AI even as we build machines in our image that adopt some of our abilities and maybe can calculate faster than us at certain things or answer language search questions faster than us, they are still made by us. They are in our image and we still have control of them. I don't believe in limitations, only in opportunities, together we can create a better future for everyone.