5 questions for Chris Miller
With Professor Chris Miller, Associate Professor of International History at The Fletcher School
Welcome back to our regular Friday feature, The Future in Five Questions. This week we interview Chris Miller, the author of “Chip War: the Fight For the World’s Most Critical Technology.” Miller teaches international history at Tufts University’s Fletcher School, and is a fellow at the American Enterprise Institute. Responses have been edited for length and clarity.
What’s one underrated big idea?
Moore’s Law — the prediction that the amount of computing power on each chip will double every other year. From the early 1960s to today, Moore’s Law has produced a billionfold increase in computing power. The exponential growth is almost impossible to get your head around. Imagine if, every other year, airplanes flew twice as fast or houses were twice as big. We’ve come to take it for granted in computing.
Today the most advanced chips have billions of electrical circuits with each component smaller than a coronavirus. Making them even smaller so that more can be fit on each chip is mind-numbingly difficult and brutally expensive. So there’s a risk that over the coming decade, Moore’s Law simply becomes too expensive to keep going.
What’s a technology you think is overhyped?
Artificial intelligence and big data. “Data is the new oil,” some analysts say, but the dramatic increase in the application of artificial intelligence across the economy hasn’t been driven by an increase in the quantity of data in the world. Nor is it because computer scientists have become more capable of writing smart programs. The key driver of AI has been a vast expansion in the availability of processing power and memory capability, enabled by an exponential growth rate in the quality of semiconductors. Today Nvidia’s H100 GPU has 80 billion transistors on it. An advanced chip from a decade ago had only one fortieth the processing power.
What book most shaped your conception of the future?
The best way to understand the future is to read history. Paul Kennedy’s Rise and Fall of the Great Powers is a magisterial account of how great powers compete for influence, and how this clash of influences is shaped by economic and technological factors. Every generation of humans likes to think their era is unique. When we talk about tech we usually focus on what is new. But so much is not new. Geopolitical competition is a constant. Technology changes, but the competitive dynamics driving it persist.
What could government be doing regarding tech that it isn’t?
When most people think about “tech,” they envision social media firms or the internet. But the hardest part about technology is the hardware on which it all relies. For too long, government’s focused predominantly on software and on the internet. This is obviously important but it’s only part of the story. We’ve spent our time debating how to regulate Facebook but we’ve taken for granted the improvements in computing power, memory capacity and signals processing that all Big Tech firms rely on. One of the main points of “Chip War” is that Americans don’t think about “tech” the right way. We’re too focused on software and the internet.
Other countries have done a lot more to nurture the companies building the chips on which all computing depends. In key East Asian countries like Taiwan, Korea, and especially China, tax treatment, regulation, and permitting are easier. Building manufacturing facilities there is just more straightforward, which is why so much manufacturing of high tech equipment has moved to East Asia.
What has surprised you most this year?
How tough the Biden administration has been when it comes to restricting tech exports to China. The ramifications will be huge. It’s not only China’s military that faces new restrictions. The implications will be felt by China’s big, consumer-focused tech firms like Alibaba and Tencent. Tech decoupling is going deeper than most people expected.
More bad press for facial recognition tech: A new study from the University of Cambridge’s Minderoo Center shows that three high-profile uses of the technology in the U.K. have failed to meet even the most superficial ethical recommendations.
The report’s author, Minderoo fellow Evani Radiya-Dixit, points out a handful of key areas where police in London and South Wales fell short of the nascent industry best practices for the technology: Privacy, as the recognition tools were applied widely and indiscriminately; bias, which the tools “were not transparently evaluated for”’; accountability, as there was no human user to check or intervene in the system and redress potential complaints; and a lack of regular oversight from non-police.
But the problem with the technology isn’t just limited to police work. As Radiya-Dixit writes, “The line between the public and private sector is becoming increasingly blurred, as police and private companies often collaborate in the development and deployment of facial recognition.”
The report concludes by calling for researchers to continue scrutinizing the tech’s use along the same lines — which are, notably, similar to those made in the Biden administration’s “AI Bill of Rights,” and the EU’s nascent AI Act — and for a ban to police use of the technology in public spaces. — Derek Robertson
An esoteric little news blip you might have missed: Elon Musk is, at long last, the owner of Twitter.
And not only that, he’s bringing a decidedly personal touch to the platform: He’s also the company’s new CEO, having fired a cadre of longtime executives including his now-predecessor Parag Agrawal.
When I wrote about Musk’s initial bid for the platform way back in April I made the case that although it might seem odd for someone nominally obsessed with Mars colonies, giant networks of underground tunnels, brain-computer interfaces — you know, actual future stuff — to be obsessed with owning a decidedly old-school piece of 2D software, it actually fits perfectly into his futurist agenda in terms of the raw power it gives him.
Now the world will see how he decides to actually wield it. (Read POLITICO’s Rebecca Kern on the white knuckles in Washington.) As that vision comes into focus, it’ll be an impossible-to-ignore, ongoing case study in how a simple piece of text-and-image software can continue to shape public life. — Derek Robertson
This piece is republished from Politico.