The Debate About and Consequences of AI Regulation

0:00Today on the AI Daily Brief, a set of readings on the regulatory discourse surrounding AI. The AI Daily Brief is a daily podcast and video about the most important news and discussions in AI. To join the conversation, follow the Discord link in our show notes. Hello friends, happy weekend, and it of course being the weekend, that means it's time for an AI Long Reads episode. A big part of the discussion recently has been surrounding California's controversial AI legislation SB 1047. Today we read two pieces that somewhat relate to it, although one relates more to the EU AI Act, but both reflect the current state of discussion and discourse around AI regulation.

0:35I am going to turn to an 11 Labs version of myself to help read these, and we're going to start with an op-ed in The Economist by The Economist staff. Regulators are focusing on real AI risks over theoretical ones. Good. Rules on safety may one day be needed, but not yet. Here's that piece. I'm sorry, Dave. I'm afraid I can't do that. HAL 9000, The Murderous Computer in 2001, A Space Odyssey is one of many examples in science fiction of an artificial intelligence, AI, that outwits its human creators with deadly consequences. Recent progress in AI, notably the release of ChatGPT, has pushed the question of existential risk up the international agenda.

1:14In March 2023, a host of tech luminaries, including Elon Musk, called for a pause of at least six months in the development of AI over safety concerns. At an AI safety summit in Britain last autumn, politicians and boffins discussed how best to regulate this potentially dangerous technology. Fast forward to today, though, and the mood has changed. Fears that the technology was moving too quickly have been replaced by worries that AI may be less widely useful, in its current form than expected, and that tech firms may have overhyped it. At the same time, the process of drawing up rules has led policymakers to recognize the need to grapple with existing problems associated with AI, such as bias, discrimination, and violation of intellectual property rights.

1:54As the final chapter in our school's briefs on AI explains, the focus of regulation has shifted from vague hypothetical risks to specific and immediate ones. This is a good thing. AI-based systems that assess people for loans or mortgages and allocate benefits have been found to display racial bias, for instance. AI recruitment systems that sift resumes appear to favor men. Facial recognition systems used by law enforcement agencies are more likely to misidentify people of color. AI tools can be used to create deepfake videos, including pornographic ones, to harass people or misrepresent the views of politicians.

2:27Artists, musicians, and news organizations say their work has been used without permission to train AI models. And there is uncertainty over the legality of using personal data for training purposes without explicit consent. The result has been a flurry of new laws. The use of live facial recognition systems by law enforcement agencies will be banned under the European Union's AI Act, for example, along with the use of AI for predictive policing, emotion recognition, and subliminal advertising. Many countries have introduced rules requiring AI-generated videos to be labeled. South Korea has banned deepfake videos of politicians in the 90 days before an election.

3:00Singapore may follow suit. In some cases, existing rules will need to be clarified. Both Apple and Meta have said that they will not release some of their AI products in the EU because of ambiguity and rules on the use of personal data. In an online essay for The Economist Mark Zuckerberg, the chief executive of Meta, and Daniel Ek, the boss of Spotify, argue that this uncertainty means that European consumers are being denied access to the latest technology. And some things, such as whether the use of copyrighted material for training purposes is permitted under fair use rules, may be decided in the courts.

3:31Some of these efforts to deal with existing problems with AI will work better than others, but they reflect the way that legislators are choosing to focus on the real-life risks associated with existing AI systems. That is not to say that safety risks should be ignored. In time, specific safety regulations may be needed, but the nature and extent of future existential risk is difficult to quantify, which means it is hard to legislate against it now. To see that, look no further than SB 1047, a controversial law working its way through California's state legislature. Advocates say the bill would reduce the chance of a rogue AI causing a catastrophe, defined as mass casualties or more than$500 worth of damage, through the use of chemical, biological, radiological, or nuclear weapons or cyber attacks on critical infrastructure.

4:15It would require creators of large AI models to comply with safety protocols and build in a kill switch. Critics say its framing owes more to science fiction than reality, and its vague wording would hamstring companies and stifle academic freedom. Andrew Ning, an AI researcher, has warned that it would paralyze researchers because they would not be sure how to avoid breaking the law. After furious lobbying from its opponents, some aspects of the bill were watered down earlier this month. Bits of it do make sense, such as protections for whistleblowers at AI companies. But mostly it is founded on a quasi-religious belief that AI poses the risk of large-scale catastrophic harm, even though making nuclear or biological weapons requires access to tools and materials that are tightly controlled.

4:54If the bill reaches the desk of California's governor, Gavin Newsom, he should veto it. As things stand, it is hard to see how a large AI model could cause death or physical destruction. But there are many ways in which AI systems already can and do cause non-physical forms of harm, so legislators are for now right to focus on those. Today's episode is brought to you by Venice. The leading AI companies store your entire conversation history and attach it to your identity forever. That's every question you ask, every answer you receive, every image you generate, every thought you share with the machine, it's all being spied on.

5:27If you trust all the company's hackers and NSA board members that will ever have access to your AI conversations, then rejoice, for you are well served. For the rest of us, Venice is an alternative. Venice is a powerful AI app for text, image, and code generation that respects you as a sovereign individual and believes privacy and free speech are not only human rights, but necessary for civilizational advancement. Private, permissionless, and uncensored, you can try it for free without an account. AI Daily Brief listeners receive a 20 % discount on Venice Pro. Visit venice.ai slash NLW and enter the discount code NLW Daily Brief.

5:59That's NLW Daily Brief. All one word. Today's episode is brought to you by super intelligent, which is of course our platform that helps you learn how to use AI tools and perhaps even more importantly, gives you ideas on the best use cases that are actually going to help you achieve whatever it is you want to achieve. To recognize the end of summer and back to school slash back to work, we are running our best promotion ever. When you sign up for super intelligent between now and the end of August using code so back your first month will be 100 % free. The platform features over 600 fun, highly practical AI tutorials that get you using AI fast and with an eye to actually transforming how you get things done.

6:44We've just launched Super for Teams, so if you have a group of people at your company that want to figure out how to use AI together, I highly suggest you check it out. But for those of you who are using Super Intelligent as an individual, once again, if you sign up for Super Intelligent between now and the end of the month using code SOBAC, you will get your first month 100 % free. Go to besuper.ai and check it out today. The next piece that we're going to read, also from The Economist, is Mark Zuckerberg and Daniel Eck on why Europe should embrace open source AI. Say the two tech CEOs, it risks falling behind because of incoherent and complex regulation.

7:20Once again, I turn to an 11 Labs version of myself for this reading. This is an important moment in technology. Artificial intelligence, AI, has the potential to transform the world, increasing human productivity, accelerating scientific progress, and adding trillions of dollars to the global economy. But as with every innovative leap forward, some are better positioned than others to benefit. The gaps between those with access to build with this extraordinary technology and those without are already beginning to appear. That is why a key opportunity for European organizations is through open-source AI, models whose weights are released publicly with a permissive license.

7:56This ensures power isn't concentrated among a few large players and, as with the internet before it, creates a level playing field. The internet largely runs on open-source technologies and so do most leading tech companies. We believe the next generation of ideas and startups will be built with open-source AI because it lets developers incorporate the latest innovations at low cost and gives institutions more control over their data. It is the best shot at harnessing AI to drive progress and create economic opportunity and security for everyone. Meta open-sources many of its AI technologies, including its state-of-the-art LAMA large language models, and public institutions and researchers are already using these models to speed up medical research and preserve languages.

8:34With more open-source developers than America has, Europe is particularly well-placed to make the most of this open-source AI wave. Yet its fragmented regulatory structure, riddled with inconsistent implementation, is hampering innovation and holding back developers. Instead of clear rules that inform and guide how companies do business across the continent, our industry faces overlapping regulations and inconsistent guidance on how to comply with them. Without urgent changes, European businesses, academics, and others risk missing out on the next wave of technology investment and economic growth opportunities.

9:04Spotify is proud to be held up as a European tech success, but we are also well aware that we remain one of only a few. Looking back, it's clear that our early investment in AI made the company what it is today. A personalized experience for every user that has led to billions of discoveries of artists and creators around the world. As we look to the future of streaming, we see tremendous potential to use open-source AI to benefit the industry. This is especially important when it comes to how AI can help more artists get discovered. A simplified regulatory structure would not only accelerate the growth of open-source AI, but also provide crucial support to European developers and the broader creator ecosystem that contributes to and thrives on these innovations.

9:40Regulating against known harms is necessary, but preemptive regulation of theoretical harms for nascent technologies such as open-source AI will stifle innovation. Europe's risk-averse, complex regulation could prevent it from capitalizing on the big bets that can translate into big rewards. Take the uneven application of the EU's General Data Protection Regulation, GDPR. This landmark directive was meant to harmonize the use and flow of data, but instead EU privacy regulators are creating delays and uncertainty and are unable to agree among themselves on how the law should apply. For example, Meta has been told to delay training its models on content shared publicly by adults on Facebook and Instagram.

10:17Not because any law has been violated, but because regulators haven't agreed on how to proceed. In the short term, delaying the use of data that is routinely used in other regions means the most powerful AI models won't reflect the collective knowledge, culture, and languages of Europe, and Europeans won't get to use the latest AI products. These concerns aren't theoretical. Given the current regulatory uncertainty, Meta won't be able to release upcoming models like Llama Multimodal, which has the capability to understand images. That means European organizations won't be able to get access to the latest open-source technology, and European citizens will be left with AI built for someone else.

10:50The stark reality is that laws designed to increase European sovereignty and competitiveness are achieving the opposite. This isn't limited to our industry. Many European chief executives across a range of industries cite a complex and incoherent regulatory environment as one reason for the continent's lack of competitiveness. Europe should be simplifying and harmonizing regulations by leveraging the benefits of a single yet diverse market. Look no further than the growing gap between the number of homegrown European tech leaders and those from America and Asia, a gap that also extends to unicorns and other startups.

11:20Europe needs to make it easier to start great companies, and to do a better job of holding onto its talent. Many of its best and brightest minds in AI choose to work outside Europe. In short, Europe needs a new approach with clearer policies and more consistent enforcement. With the right regulatory environment combined with the right ambition and some of the world's top AI talent, the EU would have a real chance of leading the next generation of tech innovation. We believe that open-source AI can help European organizations make the most of this new technology by leveling the playing field, and we hope that the EU doesn't limit the possibilities that we are only starting to explore.

11:53Though Spotify and Meta use AI in different ways, we agree that thoughtful, clear, and consistent regulation can foster competition and innovation while also protecting people and giving them access to new technologies that empower them. While we can all hope that with time these laws become more refined, we also know that technology moves swiftly. On its current course, Europe will miss this once-in-a-generation opportunity, because the one thing Europe doesn't have, unless it wants to risk falling further behind, is time.

12:22All right, so back to non-AI me for a little bit of wrap-up. One of the things that's very positive about what's happening now is that the context of more regulation is giving us a chance to talk about how we think fundamentally and from a first principle standpoint about AI regulation. One of the things that has frustrated me, you can probably tell in the discourse around SB 1047, is the fact that it seems pretty clear to me that the central dividing point is that the people who wrote it and who are most for it believe that we are facing catastrophic risks, and the people who are most against it don't.

12:52And that is largely an irreconcilable position. Or at least it's irreconcilable for those two groups. The big old set of people in the middle who might take a, for example, probabilistic approach to handicapping the odds of how much that risk is and what it should say about regulation, well, there's probably a lot more room for discourse in there. I like that The Economist is pinning down exactly in that conversation. Not necessarily because I 100 % agree with their position, but more because by articulating that position, it gives people who agree with it a frame of reference, and it gives people who disagree with it something specifically to disagree upon.

13:26I would love to see a national conversation around how much we should think about big catastrophic risks and what we're willing to give up because of it. Interestingly, the two big issues that have surrounded SB 1047 are this question of catastrophic risk and on the other hand, the question of open source, which of course gets at what Zuckerberg and Eck wrote about. What's interesting about their piece is that Europe does give weight to the argument that regulation has real consequences. Apple and Meta are both not releasing certain aspects of their artificial intelligence into the European market because of the existing set of regulations.

13:59That is a loss to European citizens. Now, maybe it's a loss that those citizens and regulators are comfortable with, but it is a loss nonetheless. And in a world where honest assessment means recognizing trade-offs, we have to be actually able to talk about the real trade-offs that will come with any given decision. Anyways, like I said, I do think that the discourse is getting better, not worse when it comes to AI regulation, thanks to these bills that are emerging. And so even if you are frustrated about where they seem to be headed, maybe there is some consolation in that. For now though, that is going to do it for the AI Daily Brief.

14:31Appreciate you listening or watching as always, and until next time, peace.