Artificial Intelligence Explained | Interviews with Experts

Foreign. 20 years ago, you have one sensor for like air quality and it just says, okay, in Hong Kong, it's bad, done. I mean, and then you guess why. And you can change the, the thing to like making I know, 100,000 sensors. And then you understand, okay, it's just because of this, this and this. And we can actually solve the issue by just replacing this factory to like east coast from west coast. And you can actually make it better. And as soon as the data is like coming from multiple sources, thousands and millions of sensors, you need to analyze it. Not with humans. I mean, if you ended like with a woman reading the printouts from the sensors, it's like not help. And that's why you need a machine learning. What people name AI, but AI is like, it's probably for like science fiction, cinema and all the rest. Guys, they name it machine learning because they like, they tune computers to understand what the fuck is going on.

To be honest, I don't think it necessarily has. I think blockchain is still in its somewhat infancy stages and AI for sure is. AI still has a lot of room for growth and development. And to be honest, you really have to quantify what is AI first of all. I mean, you know, artificial intelligence isn't just necessarily robots going out and doing everyday tasks. It can be as simple as the Amazon Alexa or it can be as complicated as deep nlp, natural language processing of news sites in Chinese. We have a company called Chinascope that I'm friends with the CEO Tom Liu, and they do nlp. They're the first ones to do NLP in China. To decipher how can you actually, at a machine learning, deep learning level, how can you read Chinese characters? Because it's totally different from English letters. So blockchain is one, I guess, element of AI. But to be honest, we're still at the very, very, very infancy stages, to be completely honest.

Yeah, So I probably, I'd look at AI and blockchain from two different angles. So the first one is that Web2 has led to a series of kind of platform monopolies. And because of the sheer data that Facebook or Google or an Amazon have, or the kind of bats in Asia, they have data monopoly and therefore they have an AI monopoly. And so the current status quo of the world is that we're reliant upon increasingly fewer organizations for more of our lives. And that's only going to get, that's only going to increase as we're moving into a world with more autonomous, autonomous vehicles or whatever it may be. So that status quo is an existential threat, you know, if we end up, and it's going to be very difficult for us to catch up or for startups or new entrants to catch up with those AI monopolies. Now, the promise of something like blockchain and distributed ledger technologies in combination with AI means that firstly we can have a data commons, so we can have self sovereign ownership of our own personal data and we can choose how that's used, what organizations can use that data and for what purpose. And then we can begin to look at how we can tokenize the ownership of the AIs potentially that derive value on top. And so I kind of argue that we need decentralized AI, we need decentralized data markets if we are to kind of move away from this dystopic future that we're kind of currently on the trajectory of. So it's kind of fortuitous that this is happening at this time. And I always argue that, I think if you understand all this stuff as web3, the promise of web3 is that we can go back to the original vision of Web one, which was as a commons, an Internet commons that wasn't monopolized by a handful of global corporations. So that's kind of the promise. When you look at AI more generally and you think about job losses and this kind of stuff, obviously there's kind of different ways of looking at it and it's difficult to say which way it's going to go. The reality is that the characteristics of Web3 are Firstly, decentralization. The second one is automation, high degrees of automation leading to the point of Autonomy. And obviously DAOs and all that stuff are an example of how far that could go. But ultimately the way that I look at the potential for AI is really to augment what we do as people. So yes, there are going to be some jobs that become obsolete. And I think perhaps if you look at the negative end of that scale, the kind of whole gig economy has turned jobs into tasks. Tasks are much easier to automate. And so potentially we're seeing a shift where jobs are turned to tasks, tasks are automated. And again, one of the worrying trends with the gig economy is that the companies that employ these people, well, they don't employ them, right? So, so they don't have pensions, they don't do pension contributions, they don't insure them. So that's kind of a social, economic time bomb. If you think about large swaths of the population no longer having a safety net for jobs that are Currently tasks that ultimately going to be automated. If you look at Uber, well, it's very clear that their end state is to have driverless taxis. So if you're an Uber driver, you know, what's the shelf life of this new job that you've got where you've got no pension contributions? So I think these are the, that's the threat of the Web2 environment. But again, I think the potential and power for Web3 is that I think the organizational form or structure that's going to happen with Web3 is going to be something more akin to a cooperative model so, whereby we have very kind of fluid labor markets, very fluid capital markets. But people on an ad hoc or on a permanent basis can self organize and they can lend to one another, they can insure one another, they can mutualize assets and that could be mutualizing AI, it can be mutualizing fleets of autonomous vehicles. So I think probably the only hope that we've got of moving away from the trajectory we're on of Web two is coming back to this key point which is decentralization, mutualization of ownership, tokenization of the value of that and then the fractionalization of it and the fractionalization of it is key because what that means is that anybody anywhere in the world can join this new economic paradigm with, you know, one penny, one cent, one euro, they can become a stakeholder in it. But I do think we now need, we need to focus more on creating what I call the people layer in the web3 stack. So at the moment it's all very tech focused. How can we create code is law? How can we create totally autonomous organizations that aren't companies? I think we now need to make sure that we build in the people layer in this Web three stack before it's too late.

That does seem to be a problem a lot of times like tech and programming.