Science

Here is how I think about these issues. The Artificial in AI can sometimes mislead so let’s start by getting rid of the A and asking instead whether more NI, Natural Intelligence, will decimate the middle class. For example, will increasing education in China decimate the American middle class? I don’t think so.

As I said in my TED talk, the brainpower of China and India in the 20th century was essentially “offline”. Instead of contributing to the world technological frontier the people of China and India were just barely feeding themselves. China and India are now coming online and I see the increase in natural intelligence as one of the most hopeful facts for the future. It’s been estimated that a reduction in cancer mortality of just 10 percent would be worth $5 trillion to U.S. citizens (and even more taking into account the rest of the world). A reduction in cancer mortality is more likely to happen with a well-educated China than with a poorly educated China. So we have a huge amount to gain by greater NI.

In the case of low-skill labor the rise of China has hurt some US low-skill workers (although US workers as a whole are almost certainly better off due to lower prices). The US has historically had an abundance of highly-skilled labor and with greater education around the world we have less of a competitive advantage. In the case of high-skill labor, however, I think the opportunities for gains are much greater than with competition for low-skill labor. Ideas are what drives growth and ideas are non-rivalrous, they quickly spread around the world. The more idea creators the better for everyone. At the world level, for example, the standard of living and the growth rate of world GDP have both gotten larger as population has increased.

Greater foreign intelligence and wealth could be a threat if intelligence turns from production to destruction (this is also a potential problem with AI). We probably can’t keep China poor, even if we tried, and any attempt to try to do so would likely backfire in the worst possible way. Thus, if we want to keep high-skill Chinese workers working on medical rather than military breakthroughs, we must preserve a peaceful world of trade. Indeed, peace and trade become ever more important the richer the world gets.

Now let’s turn from NI to AI. For the foreseeable future I see AI as being very similar to additional NI. Smart people in China aren’t perfect substitutes for smart people in the United States and there are also plenty of opportunities for complementarity. Similarly AI is not a perfect substitute for NI and there are plenty of opportunities for complementarity. An AI that drives your car, for example, complements your NI because it leaves more time for more productive tasks.

(What happens when AI does become a perfect substitute for NI? We could easily be 100 years or more from that scenario but my foresighted colleague, Robin Hanson, has a new book The Age of Em that discusses the implications of uploads, human intelligence copied into software—Hanson’s book is the most complete and serious scenario analysis of the implications of a new technology ever written but most of us won’t live long enough to know whether he is right although Robin might.)

Thus, the analysis of AI and NI is similar except for one important fact. As Chinese workers become better educated a significant share of the gains will go to Chinese workers (although by no means all).  AI, however, is produced by capital. But in our world capital isn’t scarce. The world is awash in capital and computing power is getting ever-cheaper. AI isn’t like an oil field owned by a handful of people. AI will be cheap and ownership will be widespread. Just look at your cellphone—it’s faster and more powerful than a multi-million dollar Cray-2 supercomputer of 1990. Moreover, in 1990 there were only a handful of Cray-2s and today there are billions of cell-phone super-computers including hundreds of millions and soon billions in poor countries. The gains from AI, therefore, will flow not to capital but to consumers. So if anything the gains from more AI are even larger than the gains from more NI.

From my answer on Quora.

There is a new NBER paper on this topic, by Victoria Y. Fan, Dean T. Jamison, and Lawrence H. Summers, here is the abstract:

Estimates of the long-term annual cost of global warming lie in the range of 0.2-2% of global income. This high cost has generated widespread political concern and commitment as manifested in the Paris agreements of December, 2015. Analyses in this paper suggest that the expected annual cost of pandemic influenza falls in the same range as does that of climate change although toward the low end. In any given year a small likelihood exists that the world will again suffer a very severe flu pandemic akin to the one of 1918. Even a moderately severe pandemic, of which at least 6 have occurred since 1700, could lead to 2 million or more excess deaths. World Bank and other work has assessed the probable income loss from a severe pandemic at 4-5% of global GNI. The economics literature points to a very high intrinsic value of mortality risk, a value that GNI fails to capture. In this paper we use findings from that literature to generate an estimate of pandemic cost that is inclusive of both income loss and the cost of elevated mortality. We present results on an expected annual basis using reasonable (although highly uncertain) estimates of the annual probabilities of pandemics in two bands of severity. We find:

1. Expected pandemic deaths exceed 700,000 per year worldwide with an associated annual mortality cost of estimated at $490 billion. We use published figures to estimate expected income loss at $80 billion per year and hence the inclusive cost to be $570 billion per year or 0.7% of global income (range: 0.4-1.0%).

2. For moderately severe pandemics about 40% of inclusive cost results from income loss. For severe pandemics this fraction declines to 12%: the intrinsic cost of elevated mortality becomes completely dominant.

3. The estimates of mortality cost as a % of GNI range from around 1.6% in lower-middle income countries down to 0.3% in high-income countries, mostly as a result of much higher pandemic death rates in lower-income environments.

4. The distribution of pandemic severity has an exceptionally fat tail: about 95% of the expected cost results from pandemics that would be expected to kill over 7 million people worldwide.

In other words, in expected value terms an influenza pandemic is a big problem indeed.  But since, unlike global warming, it does not fit conveniently into the usual social status battles which define our politics, it receives far less attention.

I remember a very interesting debate that my father was involved in, where there was a water beetle that can’t travel very far and can’t fly. You have these in the north coast of Australia, and in millions of years, they haven’t been able to travel from one stream to another. And it came up that in the north coast of New Guinea, you have the same water beetle, with slight variations. The only way that could have happened was if New Guinea came off Australia and turned around, that the north coast of New Guinea used to be attached to the coast of Australia. It was very interesting seeing the reaction of the geologists to this argument, which was that ‘beetles can’t move continents.’ They refused to look at the evidence.

That is Geoffrey Hinton, being interviewed by Adrian Lee, mostly about AI and Go, interesting throughout.

For the pointer I thank Michelle Dawson.

Dousing every meal in salt might make food tastier, but all that extra sodium is eventually going to raise your blood pressure—giving you bigger problems than bland food. So researchers in Japan have built a prototype electric fork that uses electrical stimulation to simulate the taste of salt.

Designed and engineered using the research on electric flavoring at the University of Tokyo’s Rekimoto Lab, the battery-powered fork features a conductive handle that completes a circuit when the tines make contact with a diner’s tongue, electrically stimulating their taste buds.

The prototype fork, which was built from just $18 worth of electronics, creates the sensation of both salty and sour, and has adjustable levels of stimulation, given that everyone has unique taste buds. When pushed too far, though, the fork can produce an unpleasant metallic taste in the mouth. So if it’s ever commercialized, there will need to be an initial calibration procedure to ensure a pleasant and tasty dining experience, without going so far as to cause physical discomfort.

Take that, gdp deflator!

Here is the article, and for the pointer I thank Peter.

The paper title is Believing there is no free will corrupts intuitive cooperation, and the authors are John Protzko, Brett Ouimette, and Jonathan Schooler.  The abstract is this:

Regardless of whether free will exists, believing that it does affects one’s behavior. When an individual’s belief in free will is challenged, one can become more likely to act in an uncooperative manner. The mechanism behind the relationship between one’s belief in free will and behavior is still debated. The current study uses an economic contribution game under varying time constraints to elucidate whether reducing belief in free will allows one to justify negative behavior or if the effects occur at a more intuitive level of processing. Here we show that although people are intuitively cooperative, challenging their belief in free will corrupts this behavior, leading to impulsive selfishness. If given time to think, however, people are able to override the initial inclination toward self-interest induced by discouraging a belief in free will.

I would say that we need a large swathe of society to believe in ideals of free will and individual responsibility, even though such concepts are not entirely faultless from a metaphysical point of view.  For a given thinker, it is worth asking whether he or she adds to or takes away from that social belief.  For some writers, the concepts of individual blame and responsibility apply only to their intellectual adversaries!

For the pointer I thank Ben Southwood.

This one is transcript and podcast only, no video, and we will be doing some more in that format.  Jonathan was in top form, here are a few bits:

COWEN: If we get to a very fundamental question — left‑wing individuals and right‑wing individuals, and let’s take, for now, only America. As people, in other ways, how different do you think they are?

Or, is it just there are these semi‑accidental triggers which have set off certain modules in the left‑wingers and different modules in the right‑wingers, but otherwise they’re going to dress the same, they’re going to treat their spouses the same way, or not? Are they fundamentally different?

HAIDT: Not fundamentally different, but different in predispositions. The most important finding in psychology in the last 50 to 100 years, I would say, is the finding that everything you can measure is heritable. The heritability coefficients vary between 0.3 and 0.6, or 30 to 60 percent of the variance, under some assumptions, can be explained by the genes. It’s the largest piece of variance we can explain.

If you and I were twins separated at birth and raised in different families, our families would pick which religions we were raised in and they would pick how often we go to church or synagogue, but once we’re out on our own, we’re going to both converge on our brain’s natural level of religiosity.

Same with politics, whether you’re on the right or left is not determined by your genes, but you’re predisposed.

And:

COWEN: If you’re in a swing state in, say, proverbial southern Ohio and in a natural setting you meet a person. With what probability do you think you can guess or forecast if they’re left‑wing or right‑wing? Even-up would be 0.5.

HAIDT: Probably 0.58, 0.57. People are incredibly variable.

And:

COWEN: Would it be a partial test of your theory if we looked at a lot of different cultures and asked, “Who are the people who dress neatly and who have a lot of calendars and stamps?” to measure whether those were typically the conservatives?

HAIDT: Yes, that would be a test.

And:

COWEN: For gay individuals, maybe not all minorities, but many minorities this is very much a positive thing. If morality is fundamentally so nonrational or arational in some key ways, is it not the case we’re always either undershooting or overshooting the target, that we can never hit it just right?

Maybe for America to be more tolerant, you need the norms to be quite crude and blunt, and overstated, and we get this political correctness. Yes it’s bad, but maybe it’s less bad then when we used to undershoot the target?

Jonathan had a good answer but it is too long to excerpt.

COWEN: Let’s say you’re Brown or Yale, and students set up a lacrosse team, and they call it the Brown Redskins, and they do some rituals which offend some people. No matter what the intent would be, should Brown or Yale step in and say, “You can’t do that?”

HAIDT: There’s a big, big line between saying, “Brown or Yale should step in and tell people what they can’t — .” In general I think no, in general the idea — .

COWEN: No they shouldn’t step in?

HAIDT: They should not step in. We should be extremely limited when we say that authorities can step in and change things. The very fact of doing that encourages microaggression culture, encourages students to orient themselves towards appealing to these authorities. The point of the microaggression article is young people these days have become moral dependents.

And:

COWEN: Let me try another analogy on you. You mentioned the army, but take private corporations, and Brown and Yale are in a sense private corporations. Harvard was originally. I wouldn’t call them restrictions on free speech, I think that’s the wrong phrase, but if one’s going to use the phrase that way, there are numerous restrictions on free speech within companies, at the work place.

If you went to the water cooler and said a number of offensive things, you would be asked to stop and eventually fired, and I don’t see anything wrong with that. So if we think of Brown, Yale, or Harvard as like a normal company, isn’t there still even with all the nonsense, a lot more free speech on campus than in actual companies?

HAIDT: Yes, and there should be. Again, a company is organized to be effective in the world. Just like the army where their priority is unit cohesion, in a company your goal isn’t to encourage everyone to express their values and criticize each other, your goal is to get them to work together.

There is much much more, including on LSD, Sigmund Freud (overrated or underrated), Cecil Rhodes, how Jonathan would change undergraduate admissions, whether behavioral economics is realizing its full potential, Adam Smith, antiparsimonialism, the replication crisis in psychology, and whether Jonathan enjoys eating insects.

Disorientation is always stressful, and before modern civilization, it was often a death sentence. Sometimes it still is. But recent studies have shown that people who use GPS, when given a pen and paper, draw less-precise maps of the areas they travel through and remember fewer details about the landmarks they pass; paradoxically, this seems to be because they make fewer mistakes getting to where they’re going. Being lost — assuming, of course, that you are eventually found — has one obvious benefit: the chance to learn about the wider world and reframe your perspective. From that standpoint, the greatest threat posed by GPS might be that we never do not know exactly where we are.

That is from a long and very interesting Kim Tingley NYT piece on the former navigation secrets of the Marshall Islanders.  Here is one bit on how it works:

The Marshalls provide a crucible for navigation: 70 square miles of land, total, comprising five islands and 29 atolls, rings of coral islets that grew up around the rims of underwater volcanoes millions of years ago and now encircle gentle lagoons. These green dots and doughnuts make up two parallel north-south chains, separated from their nearest neighbors by a hundred miles on average. Swells generated by distant storms near Alaska, Antarctica, California and Indonesia travel thousands of miles to these low-lying spits of sand. When they hit, part of their energy is reflected back out to sea in arcs, like sound waves emanating from a speaker; another part curls around the atoll or island and creates a confused chop in its lee. Wave-piloting is the art of reading — by feel and by sight — these and other patterns. Detecting the minute differences in what, to an untutored eye, looks no more meaningful than a washing-machine cycle allows a ri-meto, a person of the sea in Marshallese, to determine where the nearest solid ground is — and how far off it lies — long before it is visible.

Recommended.

The Summer Institute on Field Experiments (SIFE) is a highly selective and innovative program at the University of Chicago that brings together the brightest young economists in the world and companies interested in using rigorous field experiment methods and behavioral economics to revolutionize the way they do business. The Institute seeks organization partners who are willing to be “matched” to the most promising of the Institute’s participants to work together and design solutions to problems they face. Organization partners will share their business challenges, and the Institute’s academics help them to scientifically test new ideas and solutions.

Run by John List, here is full information.

However, we find that economic conservatives are as or more scientifically literate and optimistic about science than economic leftists. Our results highlight the importance of separating different sub-dimensions of political orientation when studying the relationships between political beliefs, scientific literacy and optimism about science.

That is from Carl, Cofnis, and Woodley, via Ben Southwood.

No experiment can ever be replicated so each attempted replication must assume that the things which differ don’t matter. The more and the more important the things that we can plausibly assume don’t matter, the stronger is the original study. Chemistry students have done the same experiments for hundreds of years and that’s useful because we can plausibly assume that who and when the experiment is conducted doesn’t matter. The recent brouhaha between Nosek et al. and Gilbert et al. illustrates a weaker case.

In their critique of Nosek et al., Gilbert et al. say that some of their replications failed because things were different.

An original study that asked Israelis to imagine the consequences of military service was replicated by asking Americans to imagine the consequences of a honeymoon

Now that sounds like two very different studies but Nosek provides important context. The original study in question wasn’t about military service or honeymoons it was about the conditions that promote reconciliation between victims and injurers after an injustice has been committed. The original study asked Israelis what they would do and how they would feel about a specific injustice. Namely you and a co-worker have been working on a project for a long-time but just before submission you are called away for reserve duty [male]/maternity leave [female]. Your co-worker takes credit for all the work and gets promoted while you later get demoted. The study then went on to ask questions about the conditions necessary for reconciliation. The reserve duty/maternity leave bit was just the story element needed to explain the situation not the focus of the study.

Nosek et al. tried to replicate the study in the United States where being called up for reserve duty is less common than in Israel and where being demoted for taking maternity leave could raise legal issues so they substituted ‘had to leave for honeymoon’. Everything else was the same. One of the original authors approved the new design.

Nosek et al. were not able to replicate the original findings. Is this because they didn’t replicate the study or because the study failed to replicate? Gilbert et al. say Nosek et al. failed to replicate the study.

In my view, Gilbert et al. are caught on the horns of a dilemma. If the studies don’t replicate they aren’t interesting and if the studies replicate but only under extremely precise conditions they also aren’t interesting. We are interested in general features of the human condition not in descriptions of the choices that 75 female and 19 male Israeli students made at a particular point in time. Moreover, if changes in wording matter then surely so does the fact that the original study was on Israeli’s in 2008 and the replication used Americans in 2013 (a lot has changed over these years!) and so must also a hundred other differences. But if so, what’s the point?

Hat tip: Andrew Gelman who has more to say.

I will be doing a Conversations with Tyler with Jonathan Haidt, but with no public event and no video, transcript and podcast only.

What should I ask him?

Here is my new column for The Upshot, asking whether an undervalued tech sector might explain a significant part of the productivity slowdown.  The answer is not really, here are a few excerpts:

Chad Syverson, a professor of economics at the University of Chicago Booth School of Business, has looked more scientifically at the evidence and concluded that the productivity slowdown is all too real. These results are outlined in his recent National Bureau of Economic Research working paper “Challenges to Mismeasurement Explanations for the U.S. Productivity Slowdown.”

…the countries with smaller tech sectors still have comparably sized productivity slowdowns, and that is not what we would expect if a lot of unmeasured productivity were hiding in the tech industry.

…The productivity slowdown is too big in scale, relative to the size of the tech sector, to be plausibly compensated for by tech progress. Basically, under a conservative estimate, as outlined by Professor Syverson, the productivity slowdown has led to a cumulative loss of $2.7 trillion in gross domestic product since the end of 2004; that is how much more output would have been produced had the earlier rate of productivity growth been maintained. To make up for this difference, Professor Syverson estimates, consumer surplus (consumer benefits in excess of market price) would have to be five times as high as measured in the industries that produce and service information and communications technology. That seems implausibly large as a measurement gap.

Don’t forget that most of the outputs of the tech sector are paid for, or they induce a higher demand for broadband and smart phones, which do enter into gdp.  And this:

Or look at it this way: The tech economy just isn’t big enough to account for the productivity gap. That gap has caused measured G.D.P. to be about 15 percent lower than it would have been otherwise, yet digital technology industries were only about 7.7 percent of G.D.P. in 2004. Even if the free component of the Internet has become more important since 2004, it’s hard to imagine that it is so much better now that it accounts for such a big proportion of G.D.P.

As my column was published, this new piece by Byrne, Fernald, and Reinsdorf (pdf) came out from Brookings.  They reach similar conclusions as does Syverson, albeit using some different arguments and paths.  I’ll be giving it some independent coverage soon.

From The Economist:

In a paper just published in Science, Colin Camerer of the California Institute of Technology and a group of colleagues from universities around the world decided to check. They repeated 18 laboratory experiments in economics whose results had been published in the American Economic Review and the Quarterly Journal of Economics between 2011 and 2014.

For 11 of the 18 papers (ie, 61% of them) Dr Camerer and his colleagues found a broadly similar effect to whatever the original authors had reported. That is below the 92% replication rate they would have expected had all the original studies been as statistically robust as the authors claimed—but by the standards of medicine, psychology and genetics it is still impressive.

One theory put forward by Dr Camerer and his colleagues to explain this superior hit rate is that economics may still benefit from the zeal of the newly converted. They point out that, when the field was in its infancy, experimental economists were keen that others should adopt their methods. To that end, they persuaded economics journals to devote far more space to printing information about methods, including explicit instructions and raw data sets, than sciences journals normally would.

For the pointer I thank John B. Chilton.

The new pitcher, called the Brita Infinity pitcher, will be able to track how much water is flowing through the pitcher. When approximately 40 gallons of water have passed through the pitcher’s purification filter, the pitcher will then send a signal to the Dash Replenishment Service to reorder more filters.

The new Brita Infinity pitcher will sell on Amazon for $44.95. A three-pack of replacement filters costs between $15 and $20. Brita says the pitcher’s two lithium metal (non-rechargeable) batteries should last nearly five years, even if stored in a cold environment. You know, like your fridge. The pitcher holds up to eight cups of water, and is BPA-free.

Here is more, with a photo, via the excellent Samir Varma.

From their new report (pdf, pp.25-26):

…insurance policies are customarily written for one year and repriced annually to reflect changing exposures. Increased possibilities of loss translate promptly into increased premiums.

Up to now, climate change has not produced more frequent nor more costly hurricanes nor other weather-related events covered by insurance. As a consequence, U.S. super-cat rates have fallen steadily in recent years, which is why we have backed away from that business. If super-cats become costlier and more frequent, the likely – though far from certain – effect on Berkshire’s insurance business would be to make it larger and more profitable.

As a citizen, you may understandably find climate change keeping you up nights. As a homeowner in a low-lying area, you may wish to consider moving. But when you are thinking only as a shareholder of a major insurer, climate change should not be on your list of worries.

The pointer is from Joseph Weisenthal.