Feb 29, 2020

Axios Future

Bryan Walsh

Welcome to issue #2 of the new Axios Future, where one thing I will not attempt to predict is the future of the markets.

🚨🚨 “Axios on HBO” returns with a bang: Roger Stone on his Christian salvation (clip); an in-depth interview with Weinstein prosecutor Cyrus Vance Jr. (clip); and Kerry, Dukakis, others on Super Tuesday drama - Sunday 6 p.m. ET/PT.

1 big thing: The next frontier for Big Science

Illustration: Eniola Odetunde/Axios

In 1945, engineer and science administrator Vannevar Bush laid out a framework for support of science in the U.S. that drove prosperity and American dominance. That model isn't enough anymore, experts said at an event this week in Washington, D.C., I write with managing editor Alison Snyder.

The big picture: With China threatening to overtake the U.S. in R&D spending even as research becomes more international, science must manage the tension between cooperation and competition.

Background: Bush served as FDR's scientific adviser during World War II. In July 1945, two weeks before Hiroshima, Bush authored a report titled "Science, the Endless Frontier," arguing that significant and centralized government funding of basic scientific research was vital for America's economic well-being and security.

  • Bush's report led to the establishment of the National Science Foundation in 1950, the chief federal agency for basic scientific research.
  • In 1940, the U.S. government and private industry spent the modern-day equivalent of $5.6 billion on scientific research. Today the U.S. as a whole spends $549 billion on R&D.

Yes, but: For years, the American proportion of total global R&D spending has been declining.

  • In a Jan. 29 congressional hearing, National Science Board chair Diane Souvaine testified that "in 2019 China may have surpassed the U.S. in total R&D expenditures," though the U.S. still spends much more on basic research.

What they're saying: Beyond increased international competition, the changing nature of the U.S. demands shifts in how basic science is done and what it should accomplish. The country is more diverse than it was at the end of World War II, life expectancy is now lower than other industrialized nations, and there are massive health disparities.

  • "Most exciting scientific advancements are creating moral quandaries that worry citizens, partially because they know they will bear any burdens and partially because they feel they have no voice over the direction of science and tech even in a democracy," said Shobita Parthasarathy, professor of public policy at the University of Michigan.
  • Parthasarathy says the country's political polarization has spilled over into science. Diversity, equity and inclusion efforts can't just be about building up the scientific enterprise but must allow it to be "more representative and ultimately more politically legitimate," she said.

What's next: "The Bush model alone is no longer enough," MIT president Rafael Reif told the audience at the event, which marked the 75th anniversary of Bush's report. While it remains credible, the world faces workforce changes from automation, climate change and other pressures Bush couldn't have envisioned, Reif said.

  • Reif called for focused investment in a short list of scientific fields including AI and quantum computing, and a "DARPA-like approach to fostering fundamental research in specific fields in pursuit of advances."
  • The U.S. also must remain open to foreign talent, Reif said.

But in a world where science is more international and cooperative than ever, is there still a place for the national science policy Bush advocated? Panelists said science as a global enterprise and as a national competitive advantage aren't incompatible and that the tension can be productive.

The bottom line: Bush's "endless frontier" laid the groundwork for postwar American prosperity. But if science is to help the U.S. and the world meet the challenges of the next 75 years, the colossus Bush helped create will need to grow more nimble.

2. Data centers use less energy than expected

Cables and LED lights in a computer server in Berlin. Photo: Thomas Koehler/Getty Images

Overall energy use from data centers has increased slightly over the past decade, but improved efficiency means that they're using less energy per operation, according to a new analysis.

Why it matters: Ultra-efficient server farms have kept energy consumption from growing as fast as data use. But with 5G and AI on the horizon, new innovations will be needed to prevent an explosion in energy use.

Background: As more of daily life has migrated to the digital cloud, many environmentalists have warned that the energy required to process all that data will create a climate catastrophe.

But the new analysis, the first major attempt in a decade to compile a global view of data center energy demands, found that between 2010 and 2018 energy use increased only 6%.

  • Over the same period of time, data center computing grew by 550%.

How it works: Three major trends have kept data center energy use in check, according to Eric Masanet, an adjunct professor of engineering at Northwestern University and the lead author of the report, published in Science on Feb. 28.

  • IT devices in data centers have grown more efficient. "Just like your laptop or your cellphone, you get more performance out of your IT devices for the same amount of energy," Masanet told me.
  • Big cloud computing companies like Google have consolidated servers in massive data centers that can be run more efficiently than smaller operations.
  • Virtualization allows cloud computing companies to run multiple applications on a single server, further enhancing efficiency.

Masanet compares the changes to a city shifting commuters from individual cars to buses.

  • "A single bus uses more energy than a single car, but it can carry so many more people that the energy per person is greatly reduced," he said. "Similarly, you can minimize the energy use of a server by maximizing how much they're utilized."

What's next: Data demand is expected to double within the next few years, and "continuing those energy efficiency gains is not a guarantee," Masanet said.

  • He and his colleagues call for more stringent IT energy standards, further research into next-generation computing, and the incentivization of renewable energy sourcing.

The bottom line: Technology and policy have kept a lid on energy consumption even as data demand has exploded — which means that the only thing you're wasting when you binge-watch Love Is Blind on Netflix is your time.

3. The Vatican lays out ethics for AI

Illustration: Eniola Odetunde/Axios

Catholic leaders presented Pope Francis with a broad proposal for AI ethics, education and rights on Friday as part of an AI conference at the Vatican in Rome, Alison reports.

Why it matters: Algorithms are already starting to replace human decision-making, but ethicists and activists say now is the time to speak up about the values those algorithms should embody.

Driving the news: Members of the Pontifical Academy for Life, a group of scholars who study bioethics, are calling for AI to be developed in a way that protects the planet and safeguards "the rights and the freedom of individuals so they are not discriminated against by algorithms."

  • IBM executive vice president John Kelly and Microsoft president Brad Smith signed the "Rome Call for AI Ethics" on behalf of the two tech companies.

Between the lines: AI underpins technology that could be used to make autonomous weapons and is being used to automate jobs, putting the lives and livelihood of many at risk.

  • In attempting to mimic human intelligence, humans are building technology that is challenging our understanding of ourselves.
  • "How can we as human beings maintain our moral capacities and become better at them while automating a lot of decision-making power?" said Brian Green of the Markkula Center for Applied Ethics at Santa Clara University.

Last year, the Vatican hosted tech leaders, and Pope Francis warned them of the potential dangers of misguided use of AI.

Tech companies need binding, detailed policies that hold them accountable in addressing the many ethical concerns surrounding AI, says Meredith Whittaker, co-founder of the AI Now Institute at New York University.

The bottom line: "All of these things come down to our human choices, and that’s where the great religions of the world and people who think about ethics constantly can guide us because in the end it will come back to us," said Kelly of IBM, which has called for "targeted regulation" of AI.

Go deeper.

4. Teladocs's coronavirus bump

Fears about the coronavirus haven't shattered every stock. Look at the telehealth firm Teladoc, Axios' Bob Herman writes.

Driving the news: Teladoc's stock price has soared 12.6% this week and is now valued at more than $9 billion, because apparently Wall Street believes we will only see doctors on our iPads or on the phone as we avoid the outside world.

Reality check: Teladoc is getting more people to use digital checkups, but the company is not remotely close to turning a profit.

• Teladoc lost about $100 million in 2019, which was roughly the same loss as 2018.

• Teladoc is still spending 20% of its revenue on advertising and marketing.

• Teladoc has had major accounting problems.

• And there are still concerns telehealth visits don't save money and instead are precursors to in-person clinic visits.

5. Worthy of your time

Activists march at a climate protest in London on February 22. Photo: Tolga Akmen/Getty Images

Welcome to the climate crisis newsletter (Bill McKibben — The New Yorker)

How to build a genome (Michael Eisenstein — Nature)

The war on food waste is a waste of time (Austin Bryniarski — The Outline)

Slouching towards dystopia: the rise of surveillance capitalism (John Naughton — New Statesman)

6. 1 memorable life: Freeman Dyson, 1923-2020

Freeman Dyson at an event in 2016. Photo: Bryan Bedder/Getty Images

The renowned physicist and polymath Freeman Dyson passed away near Princeton, N.J., on Feb. 28, at the age of 96.

Why he matters: Dyson was a dazzling scientist, but his true genius lay in his astonishing imagination, which reached for the furthest edges of the cosmos.

  • In 1949, Dyson tackled one of the trickiest problems in the field: how to describe the behavior of electrons and photons.
  • Dyson's insight, which came to him while riding a Greyhound bus through Nebraska, proved key to quantum electrodynamics, which the physicist Richard Feynman called the "jewel of physics."

Dyson never won a Nobel Prize for his work. He never even bothered to earn a PhD.

The bottom line: Few scientists can be said to have played as important a role in the making of our present than Dyson — and even fewer could so brilliantly envision the future.

Bryan Walsh