Beyond ChatGPT: What a robot creator says about the idea of ​​machines with human intelligence

[ad_1]

The so-called general artificial intelligence does not yet exist, but it is being pursued by companies such as Google and OpenAI. Expert warns of potential job losses and the need for something like a universal basic income. Ameca, the robot that impresses with its resemblance to humans Patrick T. Fallon/AFP The launch of the conversational robot ChatGPT at the end of 2022 led many people to wonder what will become of their jobs in the future. That’s because the tool can write, program and even give advice. But ChatGPT could be just the beginning of something more complex. For Ben Goertzel, one of the creators of the Sophia robot, which became famous a few years ago as the most intelligent in the world, it is possible to expect machines that really reach human intelligence. The so-called artificial general intelligence (AGI) does not yet exist in practice, but it is a term used to refer to systems that learn and perform actions like human beings. This is being sought, for example, by DeepMind, Google’s artificial intelligence area, and OpenAI, creator of ChatGPT. ChatGPT is a far cry from being a true thinking machine. He lacks the ability to do complex multistage reasoning or the creativity to invent a new genre of music or literature, for example. Still, he warns that many jobs will be at risk with slightly improved versions of these robots, also called large language models (LLM). “The successors of ChatGPT, even those without human intelligence, would probably do 80% of the human work”, opined the researcher, in an interview with g1 during the Web Summit Rio. “The lesson is that most of what people get paid to do doesn’t require complex multi-stage thinking or creative originality.” Robot creator Ben Goertzel during Web Summit Rio Eóin Noonan/Web Summit Rio via Sportsfile Will machines replace humans? Ben Goertzel’s bet is that machines that think like humans will become a reality in the near future, within a period of three to eight years. For him, if that happens, there will need to be a kind of universal basic income that compensates for job losses. “The world has enough money to support everyone, but people in rich countries don’t want to share their money with other countries,” Ben said. “It’s purely a problem of greed and poor social and economic organization, which I think the world will solve. The problem is that the world will be called upon very quickly to adapt.” Ben Goertzel is one of the creators of Sophia, a robot that became known in 2018 as the smartest in the world He also claims that pausing the development of artificial intelligence – as requested in a letter signed by Elon Musk and thousands of others – is not the best way, since it would be practically impossible to guarantee that this order would be fulfilled. “An alternative is to accelerate the development if you think you can do it in a beneficial way,” said the researcher. Despite the warnings, Ben believes that there is no reason to be afraid that machines will end up with humans, as happens in the movies. That’s because the technology is mostly being developed by companies that “just” want to sell things to people. “You can’t rule out a ‘Terminator’ or something like that as a result. But I think it’s quite unlikely, it’s not something anyone is trying to build,” he said. ALSO READ: How the robot capable of programming a game from scratch in a few minutes works ‘Singles Summit’: participants at a technology event in Rio create a ‘makeout’ group Ameca: meet the realistic robot that looks human

[ad_2]

Source link