The real threat of AI - ethics, exploitation and the erosion of truth

人工智能的真正威胁——伦理困境、滥用风险及真理的侵蚀

New Scientist Podcasts

2025-05-30

34 分钟
PDF

单集简介 ...

Episode 305 As artificial intelligence grows into more and more aspects of our lives, it seems we’re just at the beginning of the boom. Hundreds of billions of dollars are being pumped into advancing AI capabilities, making it the best funded area in science. But, just like the dot-com revolution, is it a bubble waiting to burst? In this special episode of the podcast, we explore the growing promise of AI - and also the existential threat it poses. Despite the amount of money going into AI, chatbots are still making glaring mistakes, plagued with hallucinations. All the while students are relying on them to do their homework for them, and others are using them to replace very human tasks, like writing wedding speeches. So we hear from two authors who have been thinking hard about AI and machine learning - and what that means for the future. We also get into the idea of AGI, artificial general intelligence - and its cousin, artificial superintelligence, which may already exist in certain areas. With many researchers concerned about AI overthrowing humanity, is it even worth worrying about? We dig into whether AGI is even possible and who would want to develop it. This discussion has to include some mention of the human and environmental costs of these technologies, too. Energy demands are expected to skyrocket over the next few years - can the planet keep up with the demand? And alongside that, there’s a lot of human exploitation going on to help fuel these machines - a little-known fact that has to be tackled. Finally, is superintelligent AI a threat to the existence of humankind - will they want to wipe us out when they get smart enough? Or is the threat more insidious, one where we watch the slow erosion of truth and democracy? Chapters: (02:49) How chatbots and LLMs came to dominate (15:50) Superintelligent AI (18:18) What does $500 billion buy? (19:30) The high energy demand of AI (20:56) The murky ethics of the AI race (25:15) How AI is being thrust upon us (26:48) The existential threat of AI (29:57) Is AI a bubble waiting to burst? Hosted by Rowan Hooper and Sophie Bushwick, with guests Alex Wilkins, Adam Becker and Emily Bender.To read more about these stories, visit https://www.newscientist.com/ Learn more about your ad choices. Visit megaphone.fm/adchoices
更多

单集文稿 ...

  • From the creators of the popular science show with millions of YouTube subscribers comes the MinuteEarth podcast.

  • Every episode of the show dives deep into a science question you might not even know you had.

  • But once you hear the answer, you'll want to share it with everyone you know.

  • Why do rivers curve?

  • Why did the T-Rex have such tiny arms?

  • And why do so many more kids need glasses now than they used to?

  • Spoiler alert, it isn't screen time.

  • Our team of scientists digs into the research and breaks it down into a short,

  • entertaining explanation, jam-packed with science facts and terrible puns.

  • Subscribe to MinuteEarth wherever you like to listen.

  • Welcome to the World, the Universe and Us, the weekly news podcast from New Scientist.

  • I'm Rowan Hooper.

  • And I'm Sophie Bushwick.

  • This week,

  • we're focusing on what is maybe the hottest topic in science right now and certainly the best funded.

  • Yep, it's an AI special.

  • This is your guide to the technology that's changing the world and paving the way for the future.

  • We don't need to tell you that artificial intelligence has flooded into our lives in the last couple of years,

  • really.

  • ChatGPT made by the AI company OpenAI.