One of the main topics at this year’s Web Summit 2023 is of course this year’s hype topic of AI. Several experts in this field talk about current progress and experiences here. This includes Andrew McAfee, a well-known author and researcher at the Massachusetts Institute of Technology (MIT). In his keynote speech here, McAfee spoke specifically about the topic of AI regulation. This is currently particularly important in Europe due to the AI Act. But McAfee is against excessive regulation of AI; he is much more in favor of “permissionless innovation”.
“Regulation stands in the way of innovation”
“I believe that real innovation cannot take place in an over-regulated environment. Laws, of course, protect against harm and negative impacts, but they also stand in the way of innovation,” says McAfee. The expert particularly criticizes which areas are defined as high-risk sectors in the AI Act, for example, education. According to him, this classification should apply more to areas such as the environment or human health. In other areas, regulations should remain relaxed.
“You always have to remember that more laws mean that companies have to spend more money, be it on lawyers or on more developers. This in turn means that each process takes more time. This makes the current hype surrounding AI seem significantly less exciting,” says McAfee. The researcher sees two directions here: those who are for more “upstream governance”, i.e. more regulation from above, and the aforementioned “permissionless innovation”. McAffe falls squarely in the latter camp, saying there are some irreconcilable differences between the two.
Don’t make mistakes impossible
But there is an agreement between the two schools of thought: there is a lot at stake at the moment. “AI is fundamentally changing our world. We are seeing a democratization of access to very powerful technology. Of course, this can also have very negative effects that we need to protect ourselves from.” According to McAffe, permission-free innovation does not mean that there should be no regulation at all. What is more important is recognizing the context and setting the regulations at the right time and to the right extent.
McAfee is referring to Wikipedia founder Jimmy Wales, who also spoke at the Web Summit. “Jimmy Wales once said that it is more important to make correcting mistakes easy than to make mistakes impossible in the first place. I completely agree with this, because innovation only comes through trial and error. It is unpredictable and decentralized; you never know where innovators will come from. That’s why I think it’s important for the future of AI not to stand in their way unnecessarily,” says McAfee.