Roundup of interesting internet articles, January 2018 edition

Career chemist sees no reason why chemical synthesis can’t be automated with current levels of technology. This may or may not
http://blogs.sciencemag.org/pipeline/archives/2018/01/30/automated-chemistry-a-vision

Alternate headline: “Humans can’t beat a simple computer program at predicting crimes.”
https://www.theatlantic.com/technology/archive/2018/01/equivant-compas-algorithm/550646/

Microsoft’s CEO says that quantum computing will be an indispensable component of AI.
http://www.bbc.com/news/business-42797846

The “Nuro” is a Smart Car-sized, remote-piloted delivery vehicle. I think something like this will prove cheaper and more practical than flying drones for autonomous deliveries of cargo.
https://www.theverge.com/2018/1/30/16936548/nuro-self-driving-delivery-last-mile-google

This is one of those future concepts that I think will 50% completely fail and 50% succeed–albeit taking a different form than the author imagines.
https://hackernoon.com/driverless-hotel-rooms-the-end-of-uber-airbnb-and-human-landlords-e39f92cf16e1

1997 NYT article: When a computer masters the game of Go, “it will be a sign that artificial intelligence is truly beginning to become as good as the real thing,” but don’t worry since “It may be a hundred years before a computer beats humans at Go — maybe even longer.” [It happened in 2016.]
http://www.nytimes.com/1997/07/29/science/to-test-a-powerful-computer-play-an-ancient-game.html?pagewanted=all

In the 20 years since Deep Blue beat Gary Kasparov, the Elo scores of the best human chess players have barely improved, whereas machines have gotten vastly better: today’s best player, Magnus Carlsen, has a score of 2834, and the best computer chess programs are in the 3400s.
https://en.wikipedia.org/wiki/Comparison_of_top_chess_players_throughout_history
http://www.computerchess.org.uk/ccrl/4040/rating_list_all.html

‘Once developed, [killer robots] will permit armed conflict to be fought at a scale greater than ever, and at timescales faster than humans can comprehend.’
https://mobile.nytimes.com/2018/01/29/opinion/killer-robots-weapons.html?rref=collection%2Fsectioncollection%2Ftechnology&referer

Just five years after Google’s neural network taught itself to recognize cat faces, the company thinks machines are ready to recognize 5,000 species of plants and animals. How many species could you correctly identify from photos?
https://qz.com/954530/five-years-ago-ai-was-struggling-to-identify-cats-now-its-trying-to-tackle-5000-species/

An article about how AIs will replace human workers slowly, and for the foreseeable future, will not destroy more human jobs than they create.
http://www.rationaloptimist.com/blog/artificial-intelligence-augments-human-skills/

One of the clearer descriptions of what a blockchain is.
https://www.nextbigfuture.com/2018/01/blockchain-technology-concepts-explained.html

A fascinating description of how a blockchain-based replacement for Uber might work:
https://www.nytimes.com/2018/01/16/magazine/beyond-the-bitcoin-bubble.html

Did someone say…”Predictions”?
https://qz.com/1171977/ten-2018-predictions-from-the-founder-of-the-blockchain-research-institute/

On the 50th anniversary of the Tet Offensive, it’s worth looking back at the weapons of the Vietnam War, and considering the handful of them that are still not obsolete (most notably, the AK-47 rifle).
https://en.wikipedia.org/wiki/Weapons_of_the_Vietnam_War

Forty-five years after the end of the Vietnam War, the U.S. is selling warships to its former enemy.
http://www.janes.com/article/76746/vietnam-s-ex-us-coast-guard-cutter-arrives-home

Iraq plans to slap on some upgrades to keep their junker T-55 and Type 69 tanks running. The Type 69 is a Chinese copy of the Soviet T-62, which was an evolutionary upgrade of the T-55, which was an evolutionary upgrade of the T-54. If countries like Iraq keep cheapskating their militaries, will we someday have robot crews driving around 100-year-old tanks?
http://warisboring.com/iraq-learned-tank-lessons-in-the-war-with-islamic-state/

This clunker was commissioned in 1970 and will probably stay in service until 2039.
https://www.stripes.com/news/navy-s-oldest-deployable-warship-comes-out-of-yokosuka-dry-dock-after-19-months-1.508005

The first Cash for Clunkers program
http://warisboring.com/49425-2/

Swarms of small, cheap, semi-intelligent drones will probably dominate future warfare. The WWII-era “Bat Bomb” concept could make a comeback.
http://www.thedrive.com/the-war-zone/17698/chinas-is-hard-at-work-developing-swarms-of-small-drones-on-multiple-levels

BAE is building microwave weapons that can fry the electronics of enemy planes, ships, and drones at close range. It might be an effective defense against swarms of small, cheap, semi-intelligent drones.
http://www.thedrive.com/the-war-zone/17796/bae-systems-wants-its-new-microwave-close-in-weapon-system-concept-on-us-navy-ships

India’s aircraft carriers are vanity projects that will have little use in any all-out war with the most likely enemy, Pakistan.
http://warisboring.com/indias-third-aircraft-carrier-is-most-likely-a-waste-of-money/

Nuclear depth charges were a thing: a 30 kiloton underwater detonation (this is a below-average yield tactical nuke) sank dummy subs out to a radius of 1 mile.
https://medium.com/war-is-boring/how-to-nuke-a-submarine-2f0bd50f39e

“The rule of thumb is that vertical takeoff and landing means a 50% reduction in payload.”
http://edition.cnn.com/2018/01/06/politics/us-navy-baby-carrier-deploys-to-pacific-intl/index.html

In spite of the recent saber-rattling and improvements to North Korea’s nuclear missiles, a second Korean War is highly unlikely.
https://www.washingtonpost.com/news/monkey-cage/wp/2018/01/11/should-you-worry-about-a-u-s-war-with-north-korea-not-really/
https://www.washingtonpost.com/news/monkey-cage/wp/2018/01/03/why-nuclear-war-with-north-korea-is-less-likely-than-you-think/

Bill Gates and Steven Pinker agree that the state of humankind is continuing to improve, but only seems to be getting worse thanks to information technology and to an increasing williness of long-marginalized people to speak out, making us hyper-aware of what problems remain.
https://Fmobile.nytimes.com/2018/01/27/business/mind-meld-bill-gates-steven-pinker.html

Another step towards fleets of cheap, autonomous ships for oceanic and climate monitoring. In time, they would have even more uses.
https://www.nextbigfuture.com/2018/01/141674.html

Someone has been photoshopping faces of female celebrities onto nude performers in porn videos. This is only the beginning: It’s only a matter of time before computers can easily make highly accurate, 3-D models of people just by looking at photos and video footage of them. As an intermediate step, the data could be used to find porn stars whose body types best matched each celebrity’s and to digitally graft the celebrity’s face onto each nude doppelganger, but ultimately, it will be possible to make fully lifelike, CGI porn movies starring any person.
https://motherboard.vice.com/en_us/article/bjye8a/reddit-fake-porn-app-daisy-ridley

“Average” is beautiful: “Facial averageness” is considered beautiful.
https://labs.la.utexas.edu/langloislab/face-perception/the-beauty-of-averageness/

Chinese geneticists have cloned monkeys, but the failure rate was extremely high: Out of 79 clone embryos they created and implanted into female monkeys, only two were born healthy. All the rest either spontaneously aborted or died shortly after being born. The technique or a slight modification of it could probably be used to clone humans.
http://www.cell.com/cell/fulltext/S0092-8674(18)30057-6

In the early 1800s, Iceland got its first non-white immigrant: a half black, half white man named “Hans Jonathan.” He had two kids with a white woman, and today, his DNA is found in 788 Icelanders. By sequencing the genomes of 182 of these descendants and cross-referencing them, geneticists were able to reconstruct 19% of Hans’ genome. The task was made easy by the fact that African DNA easily stood out, and by Iceland’s highly detailed genealogical records, but in principle, the same method could be used to reconstruct any ancestor’s DNA. With advanced enough technology, we could make “clones” of long-dead people.
https://www.nature.com/articles/s41588-017-0031-6

Genealogy is kind of meaningless since ultimately, we all share the same ancestors and are from the same place. Stopping your research at one point in the past is an arbitrary choice, and having an ancestor that was famous or royal doesn’t make you special.
http://nautil.us/issue/56/perspective/youre-descended-from-royalty-and-so-is-everybody-else
http://livingstingy.blogspot.com/2010/10/why-genealogy-is-bunk.html

Identical twins also share the same epigenetics.
https://cosmosmagazine.com/biology/not-just-genes-identical-twins-exhibit-supersimilarity

Using frozen human embryos for IVF is just as likely to succeed as using fresh embryos.
http://www.nejm.org/doi/full/10.1056/NEJMoa1705334

Baseline, subjective happiness level is also highly heritable, which raises the prospect of genetically engineering humans to be happier.
https://link.springer.com/article/10.1007/s10902-016-9781-6
https://www.thecut.com/2016/01/classic-study-on-happiness-and-the-lottery.html

More on that: Brain damage and surgical removals of parts of the brain can improve people’s personalities. I wonder if it will be common for people in the future to get brain surgeries that make them happier or nicer.
http://www.bbc.com/future/story/20180108-when-personality-changes-from-bad-to-good

Scientists have developed tiny needles that can inject drugs into the brain. Targeted injection sites can be as small as 1 sq mm.
http://news.mit.edu/2018/ultrathin-needle-can-deliver-drugs-directly-brain-0124

The amino acids and three-unit codons that form the basis of all genetics might be optimally evolved in form and function. Whenever we find organic alien life, its DNA will look and work almost identically to our own. There is, however, a small chance that a clean-sheet genome consisting of more types of amino acids and longer codons might be superior, but there’s no plausible way natural selection could bring it about (we have settled into a local optimum when the global optimum still lurks far off). Ultimately, optimizing biological life forms from the molecular level up is probably a task that only intelligent machines will be able to do in the distant future. All multicellular life forms are so finely-tuned to use the existing genetic alphabet that testing the hypothesis by changing things at the margins probably impossible.
https://www.theatlantic.com/science/archive/2018/01/expanding-the-genetic-alphabet/549620/

Is human wealth inequality just a manifestation of broad resource- and genetic-dominance inequality observed in any group of organisms?
http://www.pnas.org/content/pnas/114/50/13154.full.pdf

Reality check on a “revolutionary new cancer test” that the media massively overhyped. More health information isn’t always better thanks to the risk of false positives.
http://blogs.sciencemag.org/pipeline/archives/2018/01/22/a-hard-look-at-liquid-biopsies

Netflix, Hulu, and Amazon Prime provide extraordinarily cheap entertainment. Never in human history has it been easier for people to occupy themselves.
http://efficiencyiseverything.com/entertainment-per-dollar/

More on that:
http://empathyeducates.org/wp-content/uploads/2014/07/Costs-For-Americans.png

Fallout 3 is a very highly rated game, takes 23-130 hours to complete, and can be bought for only $5. That averages out to…basically free.
https://howlongtobeat.com/game.php?id=3340
http://www.gamelengths.com/games/playtimes/Fallout+3/

Stephen Spielberg thinks “Ready Player One” is an accurate representation of the future…everyone is unemployed and spends their lives playing virtual reality games.
https://youtu.be/aWz6d1Z6bnU

Rollable TV screens will become mainstream once rigid screens get too big to fit through standard-sized entry doors. I think rollable, 8K resolution TVs that cover entire walls should become common in the 2030s.
http://money.cnn.com/2018/01/08/technology/lg-rollable-display-ces-2018/index.html

Google Glass may have failed, but I’ve always thought augmented reality (AR) glasses were a worthy tech concept that would return in an improved form. The “Vuzix Blade,” which reviewers have described as being an improved version of Google Glass that also makes use of the Alexa voice-activated AI assistant, could herald that return.
https://www.theverge.com/2018/1/9/16869174/vuzix-blade-ar-glasses-augmented-reality-amazon-alexa-ai-ces-2018

I really like the prediction about the future “on-demand economy.” Someone puts out a gig announcement, and AIs instantly figure out which human can do the gig the best, and they notify the person. If robots are constantly telling you to do one gig after another, it will kind of add up to something like a job.
https://msblob.blob.core.windows.net/ncmedia/2018/01/The-Future-Computed.pdf

Related: https://www.theatlantic.com/education/archive/2018/01/filling-americas-6-million-job-vacancies/549752/

It’s been one month since the NYT front page article exposing the existence of a secret U.S. government program to study UFOs appeared, and the government has still made NO ATTEMPT to deny its claims.
https://www.nytimes.com/2017/12/16/us/politics/pentagon-program-ufo-harry-reid.html

Guillermo del Toro saw a UFO: ‘The UFO, says del Toro, “Went from 1,000 meters away [to much closer] in less than a second — and it was so crappy. It was a flying saucer, so clichéd, with lights [blinking]. It’s so sad: I wish I could reveal they’re not what you think they are. They are what you think they are. And the fear we felt was so primal. I have never been that scared in my life.’
https://www.hollywoodreporter.com/news/guillermo-del-toro-seeing-a-ufo-hearing-ghosts-shaping-water-1068754

A critic breaks down all of the reasons why the Star Wars First Order’s space ships make no design sense. I agree that advanced space ships won’t really have “top” and “bottom” parts, but they could still be asymmetrical.
https://www.popularmechanics.com/culture/movies/news/a28088/star-wars-the-last-jedi-dreadnought-star-destroyer/

The $30 million XPrize award for the first company that landed a private rover on the Moon will go unclaimed.
https://lunar.xprize.org/news/blog/important-update-google-lunar-xprize

It’s possible the recent SpaceX rocket launch was actually successful, but the military satellite they put in orbit is so secret that they had to claim it crashed and the mission failed.
http://www.thedrive.com/the-war-zone/17612/the-secret-zuma-spacecraft-could-be-alive-and-well-doing-exactly-what-it-was-intended-to

Diesel car engines have been unfairly villified by the Volkswagen emissions scandal.
http://www.bbc.com/news/science-environment-42666596

 

If New York’s government can sue oil companies for making global warming worse, than can someone else sue New York’s government for shutting down its nuclear power plants, which also made global warming worse?
https://www.nextbigfuture.com/2018/01/new-york-suing-oil-companies-but-how-about-suing-new-york-for-shutting-nuclear-reactors.html

An argument against “geoengineering” turned on its head: Whale feces contains high levels of iron, iron dumped in seawater causes plankton blooms, and plankton blooms sequester CO2. Whale populations are about 70% lower now thanks to human predation than they were in pre-Industrial times. If we dumped iron powder in the oceans, we’d be restoring the natural carbon sequestration mechanism of the ocean rather than doing something unnatural.
https://www.nextbigfuture.com/2018/01/iron-fertilization-of-the-ocean-is-as-natural-as-whale-poop-and-it-can-save-the-planet.html

 

 

Transhumanists are overwhelmingly likely to be the oldest children in their families.
https://slatestarcodex.com/2018/01/08/fight-me-psychologists-birth-order-effects-exist-and-are-very-strong/

My future predictions (2018 iteration)

Like any futurist worth his salt, I’m going to put my credibility on the line by publishing a list of my future predictions. I won’t modify or delete this particular blog entry once it is published, and if my thinking about anything on the list changes, I’ll instead create a new, revised blog entry. Furthermore, as the deadlines for my predictions pass, I’ll reexamine them.

I’ve broken down my predictions by the decade. Any prediction listed under a specific decade will happen by the end of that decade, unless I specify some other date (e.g. – “X will happen early in this decade.”).

2020s

  • Better, cheaper solar panels and batteries (for grid power storage and cars) will make clean energy as cheap and as reliable as fossil fuel power for entire regions of the world, including some temperate zones. As cost “tipping points” are reached, it will make financial sense for tens of millions of private homeowners and electricity utility companies to install solar panels on their rooftops and on ground installations, respectively. This will be the case even after government clean energy subsidies are inevitably retracted. However, a 100% transition to clean energy won’t finish in rich countries until the middle of the century, and poor countries will use dirty energy well into the second half of the century.
  • Fracking and the exploitation of tar sands in the U.S. and Canada will together ensure growth in global oil production until around 2030, at which time the installed base of clean energy and batteries will be big enough to take up the slack. There will be no global energy crisis.
  • Vastly improved VR goggles with better graphics and no need to be plugged into desktop PCs will hit the market. Augmented reality (AR) glasses that are much cheaper and better than the original Google Glass will also make their market debuts.
  • “Full-immersion” audiovisual VR will be commercially available by the end of the decade. However, the tactile, olfactory, and physical movement/interaction aspects of the experience will remain underdeveloped.
  • LED light bulbs will become as cheap as CFL and even incandescent bulbs. It won’t make economic sense NOT to buy LEDs, and they will establish market dominance.
  • “Smart home”/”Wired home” technology will become mature and widespread in developed countries.
  • Video gaming will dispense with physical media, and games will be completely streamed from the internet or digitally downloaded. Business that exist just to sell game discs (Gamestop) will shut down.
  • Instead of a typical home entertainment system having a whole bunch of media discs, different media players and cable boxes, there will be one small, multipurpose box that, among other things, boosts WiFi to ensure the TV and all nearby devices can get signals at multi-Gb/s speeds.
  • Self-driving vehicles will start hitting the roads in large numbers in rich countries. The vehicles won’t drive as efficiently as humans (a lot of hesitation and slowing down for little or no reason), but they’ll be as safe as human drivers. Long-haul trucks that ply simple highway routes will be the first category of vehicles to be fully automated. The transition will be heralded by a big company like Wal-Mart buying 5,000 self-driving tractor trailers to move goods between its distribution centers and stores. Last-mile delivery–involving weaving through side streets, cities and neighborhoods, and physically carrying packages to peoples’ doors–won’t be automated until after this decade. Self-driving, privately owned passenger cars will stay few in number and will be owned by technophiles, rich people, and taxi cab companies.
  • A machine will pass the Turing Test by the end of this decade. The milestone will attract enormous amounts of attention and will lead to several retests, some of which the machine will fail, proving that it lacks the full range of human intelligence. It will lead to debate over the Turing Test’s validity as a measure of true intelligence (Ray Kurzweil actually talked about this phenomenon of “moving the goalposts” whenever we think about how smart computers are), and many AI experts will point out the existence of decades-long skepticism in the Turing Test in their community.
  • The best AIs circa 2029 won’t be able to understand and upgrade their own source codes. They will still be narrow AIs, albeit an order of magnitude better than the ones we have today.
  • Machines will become better than humans at the vast majority of computer, card, and board games. The only exceptions will be very obscure games or recently created games that no one has bothered to program an AI to play yet. But even for those games, there will be AIs with general intelligence and learning abilities that will be “good enough” to play as well as average humans by reading the instruction manuals and teaching themselves through simulated self-play.
  • The cost of getting your genome sequenced and expertly interpreted will drop below $1,000, and enough about the human genome will have been deciphered to make the cost worth the benefit. By the end of the decade, it will be common for newborns in rich countries to have their genomes sequenced.
  • At-home medical testing kits and diagnostic devices like swallowable camera-pills will become vastly better and more common.
  • China’s GDP will surpass America’s, India’s population will surpass China’s, and China will never claim the glorious title of being both the richest and most populous country.

2030s

  • VR and AR goggles will become refined technologies and probably merge into a single type of lightweight device. Like smartphones today, anyone who wants the glasses in 2030 will have them. Even poor people in Africa will be able to buy them.
  • Augmented reality contact lenses will be invented, though they won’t be as good as AR glasses and they might need remotely linked, body-worn hardware to provide them with power and data.
  • Wall-sized, thin, 8K or even 16K TVs will become common in homes in rich countries, and the TVs will be able to display 3D picture without the use of glasses. A sort of virtual reality chamber could be created at moderate cost by installing those TVs on all the walls of a room to create a single, wraparound screen.
  • The video game industry will be bigger than ever and considered high art.
  • Loneliness, social isolation, and other problems caused by overuse of technology and the atomized structure of modern life will be, ironically, cured to a large extent by technology. Chatbots that can hold friendly conversations with humans for extended periods, diagnose and treat mental illnesses as well as human therapists, and customize themselves to meet the needs of humans will become ubiquitous. The AIs will become adept at analyzing human personalities and matching lonely people with friends and lovers, and at recommending daily activities that will satisfy them, hour-by-hour. Machines will come to understand that constant technology use is antithetical to human nature, so in order to promote human wellness, they find ways to impel humans to get out of their houses, interact with other humans, and be in nature.
  • House robots will start becoming common in rich countries. They will be slower at doing household tasks than humans, but will still save people hours of labor per week. They may or may not be humanoid. For the sake of safety and minimizing annoyance, most robots will do their work when humans aren’t around. As in, you would come home from work every day and find the floors vacuumed, the lawn mowed, and your laundered clothes in your dresser, with nary a robot in sight since it will have gone back into its closet to recharge. You would never hear the commotion of a clothes washing machine, a vacuum cleaner or a lawnmower. All the work would get done when you were away, as if by magic.
  • Chatbots will steadily improve their “humanness” over the decade. The instances when AIs say or do something nonsensical will get less and less frequent. Dumber people, children, and people with some types of mental illness will be the first ones to start insisting their AIs are intelligent like humans. Later, average people will start claiming the same. By the end of the decade, a personal assistant AI like “Samantha” from the movie Her will be commercially available.
  • People will start having genuine personal relationships with AIs and robots. For example, people will resist upgrading to new personal assistant AIs because they will have emotional attachments to their old ones. The destruction of a helper robot or AI might be as emotionally traumatic to some people as the death of a human relative.
  • Thanks to improvements in battery energy density and cost, and in fast-charging technology, electric cars become cost-competitive with gas-powered cars this decade without government subsidies, leading to their rapid adoption. Electric cars are mechanically simpler and more reliable than gas-powered ones, which will hurt the car repair industry. Many gas stations will also go bankrupt or convert to fast charging stations.
  • Self-driving cars will become cheap enough and practical enough for average income people to buy, and their driving behavior will become as efficient as an average human. Over the course of this decade, there will be rapid adoption of self-driving cars in rich countries. Freed from driving, people will switch to doing things like watching movies/TV and eating. Car interiors will change accordingly. Road fatalities, and the concomitant demands for traffic police, paramedics, E.R. doctors, car mechanics, and lawyers will sharply decrease. The car insurance industry will shrivel, forcing consolidation. (Humans in those occupations will also face increasing levels of direct job competition from machines over the course of the decade.)
  • The “big box” business model will start taking over the transportation and car repair industry thanks to the rise of electric, self-driving vehicles and autonomous taxis in place of personal car ownership. The multitudes of small, scattered car repair shops will be replaced by large, centralized car repair facilities that themselves resemble factory assembly lines. Self-driving vehicles will drive to them to have their problems diagnosed and fixed, sparing their human owners from having to waste their time sitting in waiting rooms.
  • Car ownership won’t die out because it will still be a status symbol, and having a car ready in your driveway will always be more convenient than having to wait even just two minutes for an Uber cab to arrive at the curb. People are lazy.
  • The ad hoc car rental model exemplified by autonomous Uber cabs and private people renting out their autonomous cars when not in use faces a challenge since daily demand for cars peaks during morning rush hour and afternoon rush hour. In other words, everyone needs a car at the same time each day, so the ratio of cars : people can’t deviate much from, say, 1:2. Of course, if more people telecommuted (almost certain in the future thanks to better VR, faster broadband, and tech-savvy Millennials reaching middle age and taking over the workplace), and if flexible schedules became more widespread (also likely, but within certain limits since most offices can’t function efficiently unless they have “all hands on deck” for at least a few hours each day), the ratio could go even lower. However, there’s still a bottom limit to how few cars a country will need to provide adequate daily transportation for its people.
  • Automation will start having a major impact on the global economy. Machines will compensate for the shrinkage of the working-age human population in the developed world. Countries with “graying” populations like Japan and Germany will experience a new wave of economic growth. Demand for immigrant laborers will decrease across the world because of machines.
  • There will be a worldwide increase in the structural unemployment rate thanks to better and cheaper narrow AIs and robots. A plausible scenario would be for the U.S. unemployment rate to be 10%–which was last the case at the nadir of the Great Recession–but for every other economic indicator to be strong. The clear message would be that human labor is becoming decoupled from the economy.
  • Combining all the best AI and robotics technologies, it will be possible to create general-purpose androids that could function better in the real world (e.g. – perform in the workplace, learn new things, interact with humans, navigate public spaces, manage personal affairs) than the bottom 10% of humans (e.g. – elderly people, the disabled, criminals, the mentally ill, people with poor language abilities or low IQs), and in some narrow domains, the androids will be superhuman (e.g. – physical strength, memory, math abilities). Note that businesses will still find it better to employ task-specific, non-human-looking robots instead of general purpose androids.
  • By the end of this decade, only poor people, lazy people, and conspiracy theorists (like anti-vaxxers) won’t have their genomes sequenced. It will be trivially cheap, and in fact free for many people (some socialized health care systems will fully subsidize it), and enough will be known about the human genome to make it worthwhile to have the information.
  • Markets will become brutally competitive and efficient thanks to AIs. Companies will sharply grasp consumer demand through real-time surveillance, and consumers will be alerted to bargains by their personal AIs and devices (e.g. – your AR glasses will visually highlight good deals as you walk through the aisles of a store). Your personal assistant AIs and robots will look out for your self-interest by countering the efforts of other AIs to sway your spending habits in ways that benefit companies and not you.
  • “Digital immortality” will become possible for average people. Personal assistant AIs, robot servants, and other monitoring devices will be able, through observation alone, to create highly accurate personality profiles of individual humans, and to anticipate their behavior with high fidelity. Voices and mannerisms will be digitally reproducible without any hint of error. Digital simulacra of individual humans will be further refined by having them take voluntary personality tests, and by uploading their genomes, brain scans and other body scans. Even if all of the genetic and biological data couldn’t be made sense of at the moment it was uploaded to an individual’s digital profile, there will be value in saving it since it might be decipherable in the future.
  • Life expectancy will have increased by a few years thanks to pills and therapies that slightly extend human lifespan. Like, you take a $20 pill each day starting at age 20 and you end up dying at age 87 instead of age 84.
  • Global oil consumption will peak as people continue switching to other power sources.
  • Earliest possible date for the first manned Mars mission.

2040s

  • The world and peoples’ outlooks and priorities will be very different than they were in 2018. Cheap renewable energy will have become widespread and totally negated any worries about an “energy crisis” ever happening, except in exotic, hypothetical scenarios about the distant future. There will be little need for immigration thanks to machine labor and cross-border telecommuting. Moreover, there will be a strong sense in most Western countries that they’re already “diverse enough,” and that there are no further cultural benefits to letting in more foreigners since large communities of most foreign ethnic groups will already exist within their borders. There will be more need than ever for strong social safety nets and entitlement programs thanks to technological unemployment. AI will be a central political and social issue. It won’t be the borderline sci-fi, fringe issue it was in 2018.
  • Automation, mass unemployment, wealth inequalities between the owners of capital and everyone else, and differential access to expensive human augmentation technologies (such as genetic engineering) will produce overwhelming political pressure for some kind of wealth redistribution and social safety net expansion. Countries that have diligently made small, additive reforms as necessary over the preceding decades will be untroubled. However, countries that failed to adapt their political and economic systems will face upheaval.
  • 2045 will pass without the Technological Singularity happening. Ray Kurzweil will either celebrate his 97th birthday in a wheelchair, or as a popsicle frozen at the Alcor Foundation.
  • With robots running the economy, it will be common for businesses to operate 24/7: restaurants never close, online orders made at 3:00 am are packed in boxes by 3:10 am, and autonomous delivery trucks only stop to refuel.
  • Advanced energy technology, robot servants, 3D printers, telepresence, and other technologies will allow people to live largely “off-grid” if they choose, while still enjoying a level of comfort that 2018 people would envy.
  • It will be common for cities, towns and states to heavily restrict or ban human-driven vehicles within their boundaries. A sea change in thinking happens as autonomous cars become accepted as “the norm,” and human-driven cars are thought of as unusual and dangerous.

2050s

  • This is the earliest possible time that AGI/SAI will be invented. It will not be able to instantly change everything in the world or to initiate a Singularity, but it will rapidly grow in intelligence, wealth, and power.
  • Humans will be heavily dependent upon their machines for almost everything (e.g. – friendship, planning the day, random questions to be answered, career advice, legal counseling, medical checkups, driving cars), and the dependency will be so ingrained that humans will reflexively assume that “The Machines are always right.” Consciously and unconsciously, people will yield more and more of their decision-making and opinion-forming to machines, and find that they and the world writ large are better off for it.
  • The doomsaying about Global Warming will start to quiet down as the world’s transition to clean energy hits full stride and predictions about catastrophes from people like Al Gore fail to pan out by their deadlines. Sadly, people will just switch to worrying about and arguing about some new set of doomsday prophecies about something else.
  • By almost all measures, standards of living will be better in 2050 than today. People will commonly have all types of wonderful consumer devices and appliances that we can’t even fathom. However, some narrow aspects of daily life are likely to worsen, such as overcrowding and further erosion of the human character. Just as people today have short memories and take too many things for granted, so shall people in the 2050s fail to appreciate how much the standard of living has risen since today, and they will ignore all the steady triumphs humanity has made over its problems, and by default, people will still believe the world is constantly on the verge of collapsing and that things are always getting worse.
  • Cities and their suburbs across the world will have experienced massive growth since 2018. Telepresence, relatively easy off-grid living, and technological unemployment will not, on balance, have driven more people out of metro areas than have migrated into them. Farming areas full of flat, boring land will have been depopulated, and many farms will be 100% automated. The people who choose to leave the metro areas for the “wilderness” will concentrate in rural areas (including national parks) where the climate is good, the natural scenery is nice, and there are opportunities for outdoor recreation.
  • Therapeutic cloning and stem cell therapies will become useful and will effectively extend human lifespan. For example, a 70-year-old with a failing heart will be able to have a new one grown in a lab using his own DNA, and then implanted into his chest to replace the failing original organ. The new heart will be equivalent to what he had when at age 18 years, so it will last another 52 years before it too fails. In a sense, this will represent age reversal to one part of his body.
  • Many factories, farms, and supply chains will be 100% automated, and it will be common for goods to not be touched by a human being’s hands until they reach their buyers. Robots will deliver Amazon packages to your doorstep or even carry them into your house.

2060s

  • China will effectively close the technological, military, and standard of living gaps with other developed countries. Aside from the unpleasantness of being a more crowded place, life in China won’t be worse overall than life in Japan or the average European country.
  • House robots and human-sized worker robots will be as strong, agile, and dexterous as most humans, and their batteries will be energy-dense enough to power them for most of the day. A typical American family might have multiple robot servants that physically follow around the humans each day to help with tasks.
  • If a manned Mars mission hasn’t happened yet, then there’s intense pressure to do so by the centennial of the first Moon landing (1969).

2100

  • Latest possible time that AGI/SAI will be invented. By this point, computer hardware will so powerful that we could do 1:1 digital simulations of human brains. If our AI still falls far short of human-like general intelligence and creativity, then it might be that only organic substrates have the necessary properties to support them.
  • Worst case scenario is that AGI/Strong AI hasn’t been invented yet, but thousands of different types of highly efficient, task-specific Narrow AIs have (often coupled to robot bodies), and they fill almost every labor niche better than human workers ever could (“Death by a Thousand Cuts” job automation scenario). Humans grow up in a world where no one has to work, and the notion of drudge work, suffering through a daily commute, and involuntarily waking up at 6:00 am five days a week is unfathomable. Every human will have machines that constantly monitor them or follow them around, and meet practically all their needs.
  • The world could in many ways resemble Ray Kurzweil’s predicted Post-Singularity world. However, the improvements and changes will have accrued thanks to decades of AGI/Strong AI steady effort. Everything will not instantly change on DD/MM/2045 as Kurzweil suggests it will.
  • Life expectancy escape velocity and perhaps medical immortality will be achieved. It will come not from magical, all-purpose nanomachines that fix all your body’s cells and DNA, but from a combination of technologies, including therapeutic cloning of human organs, cybernetic replacements for organs and limbs, and stem cell therapies that regenerate ageing tissues and organs inside the patient’s body. The treatments will be affordable in large part thanks to robot doctors and surgeons who work almost for free, and to medical patents expiring.
  • All other aspects of medicine and healthcare will have radically advanced. There will be vaccines and cures for almost all contagious diseases. We will be masters of human genetic engineering and know exactly how to produce people that today represent the top 1% of the human race (holistically combining IQ, genetic health, physical attractiveness, and likable/prosocial personality traits). However, the value of even a genius-IQ human will be questionable since intelligent machines will be so much smarter.
  • Augmentative cybernetics (including direct brain-to-computer links) will exist and be in common use.
  • FIVR exists wherein AI game masters constantly tailor environments, NPCs and events to suit each player’s needs and to keep them entertained. Every human has his own virtual game universe where he’s #1. With no jobs in the real world to occupy them, it’s quite possible that a large fraction of the human race will willingly choose to live in FIVR.
  • Unaugmented human beings will no longer be assets that can invent things and do useful work: they will be liabilities that do (almost) everything worse than intelligent machines and augmented humans. Ergo, the size of a nation’s human population will subtract from its economic and military power, and radical shifts in geopolitics are possible. Geographically large but sparsely populated countries like Russia, Australia and Canada might become very strong.
  • The transition to green energy sources will be complete, and humans will no longer be net emitters of greenhouse gases. The means will exist to start reducing global temperatures to restore the Earth to its pre-industrial state, but people will resist because they will have gotten used to the warmer climate. People living in Canada and Russia won’t want their countries to get cold again.
  • The means to radical alter human bodies, alter memories, and alter brain structures will be available.
  • Brain implants will make “telepathy” possible between humans, machines and animals.
  • Flying cars designed to carry humans could be common, but they will be flown by machines, not humans. Ground vehicles will retain many important advantages (fuel efficiency, cargo capacity, safety, noise level, and more) and won’t become obsolete.
  • Advanced nanomachines could exist.
  • Relatively cheap interplanetary travel (probably just to Mars and to space stations and moons that are about as far as Mars) will exist.
  • Androids that are outwardly indistinguishable from humans will exist, and humans will hold no advantages over them (e.g. – physical dexterity, fine motor control, appropriateness of facial expressions, capacity for creative thought).

Roundup of interesting internet articles, December 2017 edition

The U.S. working-age population would be shrinking right now if not for immigrants and the children of immigrants. This will continue until at least 2035.
http://www.pewresearch.org/fact-tank/2017/03/08/immigration-projected-to-drive-growth-in-u-s-working-age-population-through-at-least-2035/

East Asia will have to import 275m people between the ages of 15 and 64 by 2030 to keep its working age population stable.
http://econ.st/2jT5FYx

But will demand for immigrant workers slacken once we have robot workers? Ben Goertzel thinks “toddler-level AGI” will be invented by 2030, followed by the Singularity.
https://www.nextbigfuture.com/2017/12/ai-researcher-ben-goertzel-launches-singularitynet-marketplace-and-agi-coin-cryptocurrency.html

Skip the first 15 minutes. Greg Brockman believes deep learning isn’t close to hitting the limits of what it can do, and its capabilities will continue radically improving thanks to better, faster hardware. By the end of 2018, he thinks machines will be able to generate artificial audio and video (like imitations of human voices and totally fake video footage) that humans won’t be able to distinguish from reality. Within five years, he thinks a breakthrough will happen in robotics, making them much more capable and practical for use.
https://twimlai.com/twiml-talk-74-towards-artificial-general-intelligence-greg-brockman/

Google claims it can already convincingly fake human voices.
https://qz.com/1165775/googles-voice-generating-ai-is-now-indistinguishable-from-humans/

A timely counterpoint to bellicose declarations that “machines will never replace humans” and “human judgement will always be needed,” etc.
https://www.theatlantic.com/politics/archive/2017/12/washington-wreck-positive-train-control/548744/

Self-driving cars might offer people free taxi rides in the future, so long as passengers are willing to endure sales pitches from various corporate sponsors.
https://www.theatlantic.com/technology/archive/2017/12/self-driving-cars-free-future/548945/

The typical American thinks his personal odds of losing work to machines are 30%, while everyone else’s odds average out to 49%.
http://www.pewinternet.org/2017/10/04/automation-in-everyday-life/#many-americans-expect-a-number-of-professions-to-be-dominated-by-machines-within-their-lifetimes-but-relatively-few-expect-their-own-jobs-or-professions-to-be-impacted

When asked individually, Americans say that they would not feel threatened by the discovery of non-intelligent aliens and it would not shake their religious beliefs, but they assume that would not be true for everyone else. ‘That may be because “most Americans tend to think, on any desirable trait or ability, that they’re better than the average person.”‘
https://www.washingtonpost.com/news/speaking-of-science/wp/2017/12/04/how-will-humanity-react-to-alien-life-psychologists-have-some-predictions/

An article appeared on the front page of the NYT exposing a secret Pentagon program devoted to studying UFOs. It has evidence of UFOs doing impossible aerial maneuvers and anomalous physical materials recovered at UFO land/crash sites. As of the date of me writing this blog post, the Pentagon has not denied anything in the story.
https://www.nytimes.com/2017/12/16/us/politics/pentagon-program-ufo-harry-reid.html

Videos leaked from that UFO program’s trove, showing a 2004 encounter between U.S. fighter planes and a strange object off the coast of California. The pilots could see it with their eyes and it also showed up on their visioning sensors.
https://www.yahoo.com/news/tom-delonge-takes-alien-research-215651746.html\

Humans are genetically programmed to believe bad news over good news, and are likelier to remember bad things. In opinion polls, this expresses itself as overestimation of metrics like the crime rate,  incidence of terrorism, and incidence of poor health.
http://www.bbc.com/news/uk-42329014

Of course, sometimes the conventional wisdom that the world is getting worse is true: People with lower IQs and people who are overweight are breeding faster than everyone else.
https://www.economist.com/news/science-and-technology/21732803-it-does-however-no-longer-seem-favour-braininess-data-half-million

‘This may be the most powerful gene-manipulation toolkit that has yet been described, and you can expect to see a lot of work on it in the coming months as other groups give it a shakedown. ‘
http://blogs.sciencemag.org/pipeline/archives/2017/12/08/crispr-the-latest-edition

The first two “gay genes” have been identified. By themselves, they don’t automatically make men gay, but they’re more common in gays.
https://www.newscientist.com/article/2155810-what-do-the-new-gay-genes-tell-us-about-sexual-orientation/

The FDA will make it easier for companies to sell DNA testing kits directly to Americans.
https://gizmodo.com/the-fda-just-made-it-a-lot-easier-for-dna-health-tests-1820216695

The FDA has approved a new wearable medical device: The Kardiaband EKG, which can be attached to an Apple Watch to detect abnormal heart activity with high accuracy.
http://www.dailymail.co.uk/health/article-5138101/FDA-approves-Apple-Watchs-medical-device-accessory.html

‘..So perhaps we’re finally heading for that era of personalized medicine that everyone keeps talking about…as sequencing gets relentlessly cheaper and more widespread.’
http://blogs.sciencemag.org/pipeline/archives/2017/12/18/genetic-variation-gets-more-real-all-the-time

Woman, 26, gives birth to baby who spent 24 years as frozen embryo

Should society pay for uterus transplants so that infertile women can have “the experience of pregnancy”? The experience of raising a child seems to be what really counts, and it can be had much more cheaply and at lower risk through hiring a surrogate mother.
https://www.npr.org/sections/health-shots/2017/12/05/568453168/first-baby-born-to-u-s-uterus-transplant-patient-raises-ethics-questions

‘Public Health England says there is a large amount of evidence that shows e-cigarettes are much less harmful than smoking – at least 95%.’
http://www.bbc.com/news/health-42328236

Metformin is one of the best candidates for a human anti-aging pill.
http://www.bbc.com/news/health-42273362

Within your lifetime, the means to make yourself digitally immortal will probably be invented. Here is its nascent form.
https://www.npr.org/2017/12/19/572068474/illinois-holocaust-museum-preserves-survivors-stories-as-holograms

What bad futurism looks like:
Article title: ‘2018 is when something finally gives on North Korea’
At the end of article: ‘It’s possible that a year from now not much will have changed: no war, no talks, no significant results from sanctions.’
https://qz.com/1157919/2018-is-when-something-finally-gives-on-north-korea/

‘“It is beyond me why we think an enemy [like North Korea] would waste a perfectly good nuclear weapon to experiment with a hypothetical EMP when they could destroy an actual city…EMP is a loony idea. Once an enemy uses a nuclear weapon — for any reason — it crosses the nuclear threshold and invites a nuclear response.”’
http://nationalinterest.org/blog/the-buzz/expert-emp-weapons-are-loony-idea-23695

The first atomic bomb–“Little Boy”–was surprisingly simple, and one man was able to build detailed blueprints for it using open source data. https://www.npr.org/2017/12/26/570806064/north-korea-designed-a-nuke-so-did-this-truck-driver

The first stealth aircraft, the F-117, is 40 years old. The U.S. built it thanks to insights gleaned from a Soviet paper on radar reflectivity, which was published 51 years ago.
https://arstechnica.com/tech-policy/2017/12/stealth-turns-40-looking-back-at-the-first-flight-of-have-blue/

Almost 20 years after the fall of Communism, most of NATO’s Eastern European members are still using Cold War-era weapons whose technology is not compatible with the West’s. A partial, affordable solution might be program to modify the Eastern European weapons.
http://www.janes.com/article/76473/kharkov-morozov-design-bureau-unveils-new-t-72-upgrade

Bulgaria has to send its fighter planes to Russia for maintenance, even though the country is in NATO and would have to use those same aircraft to fight Russia someday. The underfunding and failure to get rid of Soviet-era hardware could be a disaster in a war.
http://www.janes.com/article/76391/bulgaria-turns-to-russia-for-mig-29-logistics

In the U.S., we see problems from the opposite extreme, where the military is overfunded and has the luxury of falling for the siren song of advanced, unproven weapons tech that never get off the ground.
http://www.thedrive.com/the-war-zone/16695/the-navy-is-changing-its-plans-for-its-dumbed-down-zumwalts-and-their-ammoless-guns

Upgrading old clunkers like UH-1 helicopters with autonomous capabilities could keep them in service for decades to come.
http://www.janes.com/article/76439/usmc-onr-conduct-final-autonomously-operated-uh-1-demonstration

Flying drones that are indistinguishable from birds would have great reconnaissance value to militaries. ‘Robirds use flapping wing flight as a means of propulsion, with a flight performance comparable to real birds.’
https://youtu.be/-gc8kBmzOOI

Fleets of cheap, autonomous mini-subs could map the seafloor. Someday, every shipwreck will be known and every chest-o-pirate-gold recovered.
http://www.bbc.com/news/science-environment-42335230

Google Maps has so comprehensively mapped the Earth’s surface that it’s moving on to cataloging the exact locations of exterior building doors and mailboxes.
https://www.justinobeirne.com/google-maps-moat

By the end of 2018, the U.S. might be a bigger oil producer than Saudi Arabia or Russia. (U.S.-Canada fossil fuels production won’t peak until around 2030.)
https://www.nextbigfuture.com/2017/12/us-crude-oil-production-could-be-more-than-saudi-arabia-in-january.html

The DoD’s experimental “safe” alternative to cluster bombs was basically a giant nail bomb. (It failed, and we’re just bringing back cluster bombs.)
http://www.janes.com/article/76101/pentagon-reverses-cluster-munition-ban

‘If you asked experts a few years ago when they expected this to happen, they’d have been likely to say in one or two decades. Earlier this year, some experts I polled had revised their forecast to within two to five years. But Martinis’s team at Google recently announced that they hope to achieve quantum supremacy by the end of this year.’
https://www.newstatesman.com/science-tech/technology/2017/12/how-quantum-computing-will-change-world

Throwing cold water on AlphaGo Zero’s recent gaming milestones:
https://medium.com/@josecamachocollados/is-alphazero-really-a-scientific-breakthrough-in-ai-bf66ae1c84f2

Two years ago, Elon Musk said we’d have autonomous cars in two years.
https://electrek.co/2015/12/21/tesla-ceo-elon-musk-drops-prediction-full-autonomous-driving-from-3-years-to-2/

This year, Elon Musk said he’d build a massive battery farm in Australia within 100 days. He did it in 60.
http://www.bbc.com/news/world-australia-42190358

“Space blindness”: The latest monkey wrench thrown into our big plans to go to Mars. “ The Mission has been “Only 20 or 30 years away” since I was a little kid.
At least our killer robots will get there someday.
https://lasvegassun.com/news/2017/dec/02/space-blindness-must-be-solved-before-mission-to-m/