Tesla Bots vs. Universal Basic Income: Which is Elon Musk overhyping more?
Also: Why a national 'venturesome' attitude may matter more than industrial policy
“People in 1900 did not consider the internet or nuclear power unlikely: they did not conceive of them at all. So our own future will be shaped by knowledge that we do not yet have. We cannot even predict most of the problems that we shall encounter, or most of the opportunities to solve them, let alone the solutions and attempted solutions and how they will affect events.” - David Deutsch, The Beginning of Infinity
In This Issue:
The Long Read: Tesla Bots vs. Universal Basic Income: Which is Elon Musk overhyping more?
The Short Read: The case for progress through a ‘venturesome’ economy
The Micro Reads: The risks from existing nuclear power; the miracle of the $1500 sandwich; what we know about innovation; and much more. . .
🤖 Tesla Bots vs. Universal Basic Income: Which is Elon Musk overhyping more?
Shares of Tesla have risen by roughly 25 percent since the middle of May. And if they stay on that tear, it probably won’t be because of the Tesla Bot. At the automaker’s AI Day last week, Tesla founder and CEO Elon Musk said his company would build a 5 feet-8 inch, 125-pound humanoid robot — conceptually demonstrated on stage by a dancing actor in a black-and-white Tesla Bot body suit — able to carry 45 pounds, lift 150 pounds, and walk-run 5 miles per hour. The automaton would use the chips and sensors currently employed on the Autopilot driver-assist feature in Tesla cars. “It’s intended to be friendly, of course, and navigate through a world of humans, and eliminate dangerous, repetitive and boring tasks,” Musk said.
News of the Tesla Bot made me immediately think of Uber. Specifically: how the superbullish case for Uber has long rested on its ability to develop autonomous technology and eventually remove human drivers from the profitability equation. Those carbon-based lifeforms represent an estimated 80 percent of Uber’s total per mile cost. If the technology should work, autonomous ride-hailing could be a multitrillion-dollar opportunity by decade’s end. As such, I thought that maybe Musk would present the Tesla Bot as a way to make it cheaper to build Tesla cars. Labor costs currently account for roughly $2,000 to $3,000 in the cost of every vehicle. As it is, the company’s lack of legacy labor costs, such as pensions and healthcare, give it a financial edge over other US automakers.
My mental hot-take wasn’t quite right. First, Musk himself said the Tesla Bot wasn’t intended to help with Tesla’s manufacturing. He instead described it as a consumer helper of sorts: “It should be able to, you know, ‘please go to the store and get me the following groceries,’ that kind of thing.”
Second, analysts at AI Day didn’t see it as gamechanger as much as an interesting kicker for the event. This summary from Goldman Sachs certainly lacks excitement:
Tesla said it is planning a robot in a humanoid form (5' 8" tall), and that this is a natural progression as self driving vehicles are a type of robot, and there is a lot of technology overlap with [full self driving] (as well as other capabilities at Tesla such as batteries, power electronics, and manufacturing). Tesla hasn't built this yet, but hopes to have a prototype next year.
Third, neither autonomous robots nor fully autonomous cars are why analysts currently recommend buying Tesla. Rather, the company simply sells really cool cars with some self-driving functionality. This from JPMorgan:
Its products are bold, distinctive, elegant, and highly entertaining to drive. The company is led by visionary leadership, backed by a management team with solid functional strength. Although both technology and execution risk seem substantially less than was once feared, expansion into higher volume segments with lower price points seems fraught with greater risk relative to demand, execution, and competition. Meanwhile, valuation appears to be pricing in upside related to expansion into mass-market segments well beyond our volume forecasts for the Model 3.
It doesn’t surprise me that analysts are blasé over the Tesla Bot, especially given Musk’s role as corporate hypeman. It’s not just that, however. If there’s anything the pace of AV development has taught us, it’s that autonomy is, you know, really hard. In 2019, Musk somewhat infamously proclaimed that “a year from now” there would be “over a million cars with full self-driving, software, everything.” There are not. And last month he tweeted:
Roboticist Rodney Brooks has become a go-to source for somewhat curmudgeonly takes on the future of AI and robotics. But readers of his analysis would certainly have gained strong immunity against tech hype. Back in 2018, Brooks made several predictions about those and other emerging technologies. For example: No earlier than 2027 does he think a “major city bans parking and cars with drivers from a non-trivial portion of a city so that driverless cars have free reign in that area.” And no earlier than 2035 will there be an affordable robot that “can navigate around just about any US home, with its steps, its clutter, its narrow pathways between furniture, etc.”
Of course, if you think Brooks is wrong and Musk, for all his salesmanship and showmanship, will end up closer to the truth, then you also might think the latter is correct with his hyping of a basic income for all. "Essentially, in the future, physical work will be a choice," Musk also said during the AI Day presentation. "This is why I think long term there will need to be a universal basic income," he added.
Ah, yes, UBI — the economically elegant policy response to robots taking all our jobs. My curmudgeon-ish baseline take on UBI: With the help of smart public policy, advanced economies like America’s can continue to generate high levels of employment — even assuming continued technological progress results in higher productivity. While we can imagine AI/robot-driven jobless scenarios for tomorrow, doing so is unhelpful for policymakers today. In research from last March looking at skill requirements in US job vacancies, economists Daron Acemoğlu, David Autor, Jonathon Hazell, Pascual Restrepo conclude that “despite the notable surge in AI adoption, the impact of AI is still too small relative to the scale of the US labor market to have had first-order impacts on employment patterns — outside of AI hiring itself.”
Likewise, a 2020 analysis from Stanford University researcher Yong Suk Lee examined the impact of AI and robotics on three sectors: manufacturing, retail banking, and nursing homes. Although the introduction of those technologies initially replaced human workers, new jobs often emerged that at least partially offset the losses. In other cases, automation eventually increased the number of jobs as higher productivity allowed business expansion.
It’s notable — and odd — that the recent wave of technological automation fears percolating in the years before the pandemic coincided with a steady decline in the jobless rate to a 50-year low. As David Autor commented in a different piece, “Journalists and expert commentators overstate the extent of machine substitution for human labor and ignore the strong complementarities that increase productivity, raise earnings, and augment demand for skilled labor.”
Not a surprising view given the dystopian take that many in the media have about modern capitalism. UBI also has a sexy simplicity that appeals to both commentators and tech entrepreneurs. My boring response is that the best path for now is to make sure more workers can do more of what robots can’t and that we have a dynamic economy that produces a never-ending supply of new jobs for the robots to try to steal. That said, I would love a Tesla Bot that could take a late-night walk to the local grocery store and grab a gallon of milk for the morning.
🎈 The case for progress through a ‘venturesome’ economy
Concern about geopolitics — China, to be specific — rather than chronically weak productivity growth is more likely the driver of Washington’s newfound interest in industrial policy, broadly considered. For example: Congress is currently considering legislation that would expand Washington R&D efforts beyond basic research with a heavier than previous focus on particular technologies and particular missions, such as climate change. Clearly, one inspiration is Beijing’s “Made in China: 2025” plan (announced in 2015 and since rebranded due to a Western backlash) to create "national champions” in 10 high-tech manufacturing sectors. Global supply-chain issues, from PPE to semiconductors, have surely exacerbated Washington’s national security and economic reasons worries.
With techno-nationalism on the rise — and considering the problematic history of industrial policy — it’s probably good to think about the counterargument and potential downsides. In 2009’s The Venturesome Economy: How Innovation Sustains Prosperity in a More Connected World, Harvard University economist Amar Bhidé argues for deep skepticism toward the notion of expanded government subsidies for R&D and the idea that America should be worried about the expanded R&D capabilities of other nations such as India and China. He explains that high-level know-how produced abroad is highly mobile and cheap and available for global use:
A nation's "venturesome consumption" — the willingness and ability of intermediate producers and individual consumers to take a chance on and effectively use new know-how and products — is at least as important as, if not more important than, its capacity to undertake high-level research. Maryland has a higher per capita income than Mississippi, Norway has a higher per capita income than Nigeria, and Bosnia has a higher per capita income than Bangladesh; the richer places are not ahead because they are (or once were) significant developers of breakthrough through technologies. Rather, they are wealthier because of their capacity to benefit from innovations that originated elsewhere. . . . The United States has more than just great scientists and research labs: it also hosts an innovation game with many players who can exploit high-level breakthroughs regardless of where they originate. Therefore, the erosion of the U.S. lead in cutting-edge research isn't just harmless: an increase in the world's supply of high-level know-how provides more raw material for mid- and ground-level innovations that increase living standards in the United States.
Apple’s success, for instance, isn’t just because of its excellent core technology — much of which came from other domestic and foreign firms — but also the design and marketing genius of Steve Jobs, as well as the company’s legion of enthusiasts and early adopters. Bhidé notes the wisdom of British economists, C. F. Carter and B. R. Williams, who cautioned four decades ago that “it is easy to impede growth by excessive research, by having too high a percentage of scientific manpower engaged in adding to the stock of knowledge and too small a percentage engaged in using it. This is the position in Britain today.”
America’s ability to incorporate, commercialize, diffuse, and use innovations from elsewhere is a tremendous national strength — one influenced by tax policy, immigration, trade regulation, and a culture of curiosity and openness.
All that said, I think there is good reason to encourage immigration of all sorts — immigrants are by nature rather venturesome — and more federal science investment. But we should also be aware of the potential trade-offs, especially when government starts picking which technological paths to follow. Bhidé: “Entrepreneurial ‘leaps into the dark’ are therefore best sustained by great caution in expanding the scope of government intervention.”
Germany Flirts With Power Crunch in Nuclear and Coal Exit - Bloomberg | Actually, it’s worse. Not only has the exit from nuclear power raised the prospects of blackouts, Germany has actually made it harder to build renewables: “The Norwegian utility Statkraft SF says it takes twice as long to build a wind park in Germany compared with the U.S. Complaints from locals, a lack of space, stricter environmental standards and a longer permitting process are just some of the reasons growth is slowing.”
$1,500 Sandwich Illustrates How Exchange Raises Living Standards - Human Progress
Innovation: Market Failures and Public Policies - Kevin A. Bryan and Heidi L. Williams, NBER Working Paper | “Innovation is at the core of many fundamental economic problems. The sustained invention and diffusion of new technologies during the Industrial Revolution brought us from the Hobbesian life of man — ‘solitary, poor, nasty, brutish, and short’ — to the prosperity of the modern era. Innovation that replaces dirty technologies will be a key tool for mitigating climate change. The spread of ‘Green Revolution’ crops allowed poor countries to grow without fear of famine. The world’s seven largest private companies as of December 2020 all produce products that had not been invented fifty years earlier.”
The radical potential of nuclear fusion exposes the folly of our net zero deadline - Matt Ridley, The Telegraph
NASA Wants To Return To The Moon By 2024, But The Spacesuits Won't Be Ready - NPR | “NASA has been developing new spacesuits for the last 14 years, which have so far cost $420 million. The agency is expected to spend another $625 million by the time just two flight-ready suits are ready in the 2025 fiscal year.”
Can start-ups fast-track fusion energy? - Nature | “Developing and integrating the technologies needed to form a working and economical fusion power plant is beyond the current scope of one company or public lab. The next phase could be like the public–private partnerships between NASA and companies SpaceX and Orbital Sciences to develop commercial transportation for the International Space Station, in which cost and risk were shared.”
Is Geoengineering Research Objectionable? - Timothy Taylor | “I’ll just add that if those concerned about climate change want to use a slogan of ‘follow the science’ — which is I think is perhaps their strongest argument — then it’s a bad look to start arguing that certain kinds of science shouldn’t be followed.”
Tweet of the Issue: