Geoffrey Hinton

The government should insist that the big companies do lots of safety experiments, spend considerable resources like a third of their compute resources on doing safety experiments while these things are still not as intelligent on us to see how they might evade control and what we could do about it. And I think that's a lot of the debate at OpenAI. The people interested in safety like Ilya Sutskever wanted significant resources to be spent on safety. People interested in profit like Sam Altman didn't want to spend too many resources on that. (source)

As AI systems become increasingly sophisticated and integrated into everyday life, the need for thoughtful regulation becomes imperative. This isn't about stifling innovation; it's about ensuring safety, fairness, and accountability. We must guard against unintended consequences and ensure that AI technologies are used to benefit society as a whole.

I believe third-party audits should be mandatory for major AI systems. They can help ensure transparency and accountability, and mitigate potential risks associated with AI deployment. As the technology continues to evolve, maintaining public trust is crucial.

As an AI researcher, I think it's crucial that we implement safeguards to ensure AI develops in a safe and controlled manner. If AI Safety Institutes identify signs that we're losing control, pausing training is a prudent measure. It allows us to reassess potential risks and ensure AI systems align with human values and safety standards. We must prioritize safety to protect the broader public interest.