\n<\/p>\n
As projects like Pacing the Frontier look to major labs as a way to keep AI research safe, open-source models have become a sore spot for the industry<\/a>. With free distribution and little control over how they\u2019re used, open-weight models aren\u2019t easily controlled, leading some labs<\/a> to treat them as downright scary. <\/p>\n But at the Ai4 conference in Las Vegas last week, three of the world\u2019s most respected AI researchers \u2014 Nobel Prize winner Geoffrey Hinton, World Labs CEO and co-founder Fei-Fei Li, and Coursera co-founder Andrew Ng \u2014 spoke out on the issue. And while they disagreed on particular tactics, all three made a powerful case for keeping AI open.<\/p>\n For the three speakers, the core concern was allowing a handful of major AI companies to control the pace of progress. When a few companies control access to a technology, as Apple and Google do with mobile operating systems, innovation can slow and the companies that control the platforms can influence what gets built on them.<\/p>\n Andrew Ng said that he worried about a similar dynamic emerging in AI. \u201cI don\u2019t want there to be gatekeepers,\u201d Ng said. \u201cThat limits how all of us can access AI.\u201d<\/p>\n Companies have an incentive to protect their competitive advantages, including by influencing the rules that govern the industry. That could create a dynamic where only the largest, best-capitalized firms with the resources to build the most advanced AI systems. <\/p>\n Ng\u2019s solution was to maintain multiple providers, with models and companies competing rather than allowing a handful of of players to dominate the field. \u201cIf I were to try to give one prescription, it would be to promote openness,\u201d Ng said, \u201cbecause AI is amazing technology and I want it to be in everyone\u2019s hands.\u201d<\/p>\n But not everyone agreed that open-weight models would help preserve that state of play. Hinton, in particular, drew a distinction between open-source software, which makes the underlying code available for inspection and modification, and open-weight models, which release the parameters of a trained AI model to the public.<\/p>\n \u201cOpen source is great. You show people the code, and lots of people look at the lines of code and say, \u2018Oh, there\u2019s a bug.\u2019 Open weights means you train a big model and then you give people the weights. That\u2019s very different,\u201d Hinton said. \u201cI was against open [weights] because it makes it so easy for people to take these big foundation models, which are very expensive to train, and for much less money train them to do bad things like cyber attacks.\u201d<\/p>\n But whatever his reservations, Hinton acknowledged that open-weight models are already a permanent fixture of AI. \u201cI think that battle\u2019s been lost. We now have open-weight models, so the barrier to lots of people getting these big models, which was the cost of training foundation models, that barrier has disappeared. It\u2019s too late.\u201d<\/p>\n Yet accepting reality didn\u2019t mean ignoring the risks. Hinton\u2019s position was clear: AI would continue to advance, and he thought that was largely a good thing. He said it would boost productivity and improve education and healthcare. \u201cWorrying about the possible bad effects of AI and the things that intelligent beings might do when they\u2019re smarter than us. I don\u2019t think that\u2019s unfair. I think it is unfair to label anybody who thinks like that as a fear-monger,\u201d Hinton added.<\/p>\n Ng took a different view. The question, he argued, wasn\u2019t whether open models were risky, but who controlled access and who would win the market. Whoever built the cheaper model would have the advantage. If China\u2019s open-weight models gained widespread adoption across Asia, Africa, and\/or the developing world, he warned, they could influence how billions of people encountered ideas about democracy, freedom, and human rights.<\/p>\n \u201cOne thing I hope we do is encourage American competitiveness and open-source AI. It turns out that AI is a tremendous source of soft power. You can see the way China\u2019s model has tremendous accomplishment with Africa, for example,\u201d Ng said. \u201cBut my worry is because of all the lobbying in the U.S. and the fear-mongering, building open-source AI in America is struggling to compete with open-weight models coming out of China, and my worry is that if China figures out a fundamentally more cost-efficient way to build AI, then things that are more cost-efficient have a fundamental business adoption advantage.\u201d<\/p>\n Li pushed back on that framing. \u201cIt\u2019s very dangerous to make this a dichotomy between complete openness all the way to complete closedness,\u201d she said. \u201cIn complex software systems as well as scientific systems it\u2019s much more nuanced.\u201d<\/p>\n Li used nuclear physics as an example: scientific papers are published openly, but uranium is regulated, while laboratory work falls somewhere in between. The lesson, she explained, was that openness doesn\u2019t have to be an all-or-nothing choice. Different layers of the ecosystem can operate at different levels of openness.<\/p>\n She also highlighted collaborations between public and private institutions, such as the Human Genome Project. The resulting knowledge became a platform that others could build on, she said, allowing pharmaceutical companies to profit, scientists to advance their work and society to benefit.<\/p>\n \u201cSo I think we have to use [AI] as that kind of infrastructure,\u201d Li said. \u201cWe need some levels of openness, both in scientific discovery, in education, in global partnership, as well as lucrative business models for entrepreneurs. But we also will accept closed-source systems. This debate, especially at the sweeping level of \u2018we can only tolerate one,\u2019 is a false debate. We need to get to a level of nuance.\u201d<\/p>\n But everyone agreed that some level of regulation would be necessary to keep AI on the right track. \u201cWhat we want to do is develop AI in a direction that helps people, and regulation will help us do that,\u201d Hinton said. \u201cYou can\u2019t leave it to people like Elon Musk and Mark Zuckerberg to decide how AI should be done.\u201d<\/p>\n<\/div>\n When you purchase through links in our articles, we may earn a small commission<\/a>. This doesn\u2019t affect our editorial independence.<\/em><\/p>\n