Professor Geoffrey Hinton, known as the "Godfather of AI", believes that AI can be dangerous to humanity if "raised" in the wrong direction.
"It's like raising a cute little tiger, you have to make sure it doesn't kill you when it grows up," Geoffrey Hinton, a computer scientist who won the 2024 Nobel Prize in Physics and is one of the founders of artificial intelligence, told CBS News.
Geoffrey Hinton at Google's headquarters in Mountain View, California, in 2015. Photo: APAccording to him, tools like ChatGPT, Copilot, or Gemini help people create cute and funny images. But behind it is an emotionless system, focused only on delivering the best results according to the way the neural network and internal models understand. Therefore, if trained incorrectly, the consequences can be very dangerous for humanity.
"It's not that AI advances don't bring benefits. It still makes great strides in medicine, education, and climate science. But we need to consider the possibility: when will AI become so smart that we don't know what it's thinking or plotting?", Hinton continued. "There's a 10-20% risk of AI taking over humanity."
Regarding the existing threats, he assessed cybersecurity as a concern. He believes AI will help hackers attack targets such as banks, hospitals and infrastructure more effectively. But on the other hand, artificial intelligence also encrypts or solves problems faster. Therefore, agencies and businesses will need to invest a lot of resources to apply AI to their network systems.
The "AI Godfather" believes that technology companies such as OpenAI, Microsoft, Meta, and Google are focusing on short-term profits and not paying enough attention to AI safety. Meanwhile, governments have not yet implemented clear regulations on artificial intelligence, putting users at risk.
"We are in a very special period in history, in a relatively short period, everything can change completely on an unprecedented scale. It's hard to take that emotionally," Hinton added.
In 2023, he caused a stir when he resigned from his leadership position at Google AI and issued many warnings about the risks of losing control in the development of artificial intelligence. Meanwhile, Yann LeCun, also known as the "AI Godfather" and currently the AI Director of Meta, opposes the view that AI will destroy humanity and that it will be "a long time" before reaching the realm of artificial intelligence surpassing human intelligence.
(according to CBS News, Techradar)