Geoffrey Everest Hinton is a British computer scientist who, as of today, has left its location a Google to speak freely about the risks derived from the development ofartificial intelligence. What would be strange? Just that Hinton is one of the pioneers of AI, a kind of "godfather of artificial intelligence“, known precisely for his contributions to the development of machine learning. In fact, his pioneering work on neural networks shaped the AI systems that power many of today's products. “I left to be able to talk about its dangers”, he said in a tweet after the New York Times broke the news yesterday.
Hinton: AI will soon be smarter than us
According to the scientist, there is a future danger that artificial intelligence could change and become more intelligent than man "at this moment, they are not smarter than us, as far as I know. But I think soon they could be”Hinton explained.
In a BBC interview he then further explained his fears and how the chatbot could soon surpass the information level of a human brain: “Right now, what we are seeing is that things like GPT-4 overshadow a person in the amount of general knowledge that it has and it far obscures it. In terms of reasoning, it's not that good, but it already makes simple reasoning. And given the pace of progress, we expect things to improve quite quickly. So we have to worry about it." Hinton's fears stem from what's called deep learning: In artificial intelligence, neural networks are systems similar to the human brain in how they learn and process information. This allows the AI to learn from experience, as an ordinary person would.
THEabandonment of Google, is not a criticism of the technological giant which, on the contrary, Hinton underlines in his tweet “He acted very responsibly”. “I actually want to say a few good things about Google. And they are more credible if I don't work for Google” explained the scientist. Google indeed, according to a statement from the chief scientist Jeff Dean would be committed “to a responsible approach to AI. We continue to learn to understand emerging risks and innovate boldly.”
Hinton did not, however, regret his research: “I console myself with the usual excuse: if I hadn't done it, someone else would have done it”.