Hello, and welcome to this week’s installment of The Future In Five Questions. To cap off this year’s interview series I reached out to one of the most outspoken voices in tech policy: Matt Stoller, the antitrust advocate and director of research at the American Economic Liberties Project, who POLITICO Magazine called “Washington’s Angriest Progressive.” Stoller held forth via email on the blockchain (“a less efficient spreadsheet”), Microsoft (“can’t even fix Outlook search”) and how the conservative legal movement might view economic libertarianism in five to 10 years (“anachronistic and weird”). The following has been edited and condensed for clarity: What’s a technology that you think is overhyped? The blockchain. It’s just a less efficient spreadsheet. For some reason, there’s still chatter in Congress about regulating crypto, and CNBC and Bloomberg cover bitcoin as if it’s anything more than looking at the price of Beanie Babies. AI probably isn’t overhyped, but the policy discourse is. AI is a business method, not a product. It’s a bit like the wheel. There are wheels on trains, cars, toys and industrial machinery. We don’t have a wheel regulator or wheel-specific laws. We make laws and rules based on industrial sectors or basic legal concepts, like fraud or deception. So for instance, if you want to understand AI in search, think about the antitrust cases against Google, which will determine how machine learning is rolled out and who can innovate and reach customers. Or if you want to understand AI in health care, look at what UnitedHealth Group is doing with its unique data sets. If you want to understand building LLMs, look at Nvidia’s market power or cloud computing competition. And so forth. But there isn’t “AI policy,” that’s just a substitute for actually doing the grubby work of analyzing how markets and legal frameworks actually work in every sector. What’s one underrated big idea? Anti-monopoly laws. Big companies block innovation because they only deploy technology that doesn’t disrupt their business. It’s why Google invented but didn’t deploy AI, a much smaller firm — OpenAI did. Historically, the great technologies in Silicon Valley and America writ large were built by small companies where engineers and scientists had the freedom to invent, make, tinker with, and deploy what they thought would sell without being restrained by monopolists. What book most shaped your conception of the future? “Network Nation” by Richard John. [Asked to elaborate, Stoller referenced this 2020 blog post.] What could the government be doing regarding technology that it isn’t? Set a size cap on corporations so no firm can be larger than $100 billion in market capitalization. Talk to a lot of engineers at Google, Facebook, Amazon, Microsoft, Apple, you’ll find that they spend most of their time getting permission to do stuff or going to meetings. It’s why Microsoft — despite its vaunted pledge to lead in AI — can’t even fix Outlook search. Even doing simple break-ups, like separating YouTube and the rest of Google, would be great. If we unlocked all this knowledge and research capacity, the amount of genuine innovation would be amazing. What has surprised you the most this year? I would say two things. First, I’m surprised by how older policymakers simply cannot incorporate the shortages during Covid-19 into their framework of what matters. There’s just a sense of oh well, let’s move on to the next shiny object that doesn’t really affect the daily lives of most Americans, whether that’s Israel, Ukraine, AI, surveillance, whatever. But our drug shortage problem has worsened. We’re in the process of getting rid of our domestic baby formula industry. The U.S. became a net food importer this year, which is historic and extremely scary. Our national security posture is a disaster because people who know logistics don’t have decision-making power. Basically, the theme song from "Jaws" is starting but we’re just paddling along in the water. All of these are a pure function of a national strategy from the early 1980s that prioritized finance over making things, aka a ‘capital light’ strategy. Finance is now something like 7-9% of GDP instead of 2%, which is its historical share. The problems I focus on — monopolization of key industrial sectors — derive from that basic policy choice. Things are changing quickly, but this is a big country. So despite the chatter about tariffs and industrial policy, we haven’t built a new consensus to genuinely move away from our old strategy. And that’s because Wall Street — which organizes corporate America and has egregiously mismanaged our national resources — is offended whenever anyone in the public sector tries to change the direction of our social order and see things through a lens other than a profit and loss statement on a spreadsheet. And there aren’t yet enough policymakers confident enough to ignore bankers, economists, and CEOs, and see them as the fools they are. But there is a lot of rethinking going on. Second, I’m surprised by how quickly the conservative legal movement is rethinking its libertarian assumptions. I’d guess that within 5-10 years, there will be a new consensus about national strategy, and people will look back on a whole bunch of concepts, like Federal Reserve independence, the primacy of economics and finance, fear of a public administrative state, dismissal of private rights of action, respect for big companies, as anachronistic and weird.
| 
Comments
Post a Comment