Geoffrey Hinton, the former Google researcher widely recognized as the Godfather of AI, has issued another warning about artificial intelligence systems. In his latest interaction, Hinton compared current AI systems to AI beings and said there is an urgent need for research on how to stop them from taking over the world.
Speaking to British-American entrepreneur Andrew Keen on his YouTube channel, Hinton said, “The existential threat is very different. We’ve never had to deal with things smarter than us. I mean, nuclear weapons, they’re not smarter than us. They just make a bigger bang, and they’re easy to understand, whereas AI, people don’t yet understand that what we’re doing is creating alien beings.”
Hinton, while comparing AI systems to alien beings, stated that they are capable of understanding what they are saying, which goes contrary to the general perception in the field of AI that current large language models (LLMs) are nothing but statistical models that predict the next word based on patterns in their training data.
Then again, LLMs are not supposed to have self-preservation instincts, but a new Anthropic study found during an experiment that its Claude Opus 4 model would blackmail an engineer to reveal their affair under the threat of being shut down.
“If you look through the James Webb telescope and you saw an alien invasion that was going to get here in about 10 years’ time, people would be terrified. But we’re actually making these aliens. They’re real beings. They understand what they’re saying. They can make plans of their own to blackmail people who want to turn them off,” Hinton said.
“We should be very worried about this, and we should be urgently doing research on how to prevent them taking over. That’s a very different threat from anything we’ve had before,” he added.
Hinton had quit his job at Google in 2023 after working at the search giant for over a decade. At the time, he said it was so he could speak freely about the risks of AI without being influenced by corporate interests.
Hinton won the prestigious Turing Award in 2018 for his research in deep learning and the Nobel Prize in Physics in 2024 for foundational discoveries that enabled machine learning with artificial neural networks.
