Geoffrey Hinton, widely regarded as the godfather of artificial intelligence, has resigned from Google, expressing regret over his life's work and warning that AI chatbots could soon surpass human intelligence. The 75-year-old British-Canadian cognitive psychologist told the BBC that the dangers of the technology are 'quite scary'.
Hinton's pioneering research on neural networks and deep learning laid the foundation for modern AI systems like ChatGPT. He said that chatbots such as GPT-4 already hold more general knowledge than a person and are improving rapidly in reasoning. 'We need to worry about that,' he added.
Hinton warned of 'bad actors' using AI for malicious purposes, citing a hypothetical scenario where a leader like Vladimir Putin might give robots the ability to create their own sub-goals, potentially leading to a drive for more power. He contrasted biological intelligence with digital systems, noting that digital AIs can share knowledge instantly across countless copies, giving them a vast advantage.
Matt Clifford, chairman of the UK's Advanced Research and Invention Agency, stated that Hinton's departure 'underlines the rate at which AI capabilities are accelerating' and stressed the need for heavy investment in AI safety. Hinton joins a growing number of experts, including Elon Musk and Yoshua Bengio, who have called for a pause in advanced AI development to implement safety measures.
Despite his concerns, Hinton acknowledged that AI would deliver more benefits than risks in the short term and opposed halting development, citing international competition with China. He emphasised that it is the government's responsibility to ensure AI is developed safely. Hinton also praised Google as 'very responsible' but said his criticism would be more credible from outside the company.
