Slashdot Mirror

In a startling demonstration, the machine drew on experimentation, data, and observation of humans to learn how simple implements could help it achieve a task. From a report: Learning to use tools played a crucial role in the evolution of human intelligence. It may yet prove vital to the emergence of smarter, more capable robots, too. New research shows that robots can figure out at least the rudiments of tool use, through a combination of experimenting and observing people. Chelsea Finn, a researcher at Google Brain, and Sergey Levine, an assistant professor at UC Berkeley, developed the robotic system together with several of Levine's students. The setup consists of an off-the-shelf robot arm that can be controlled by a person or a computer. It also includes a camera that sees the environment within reach of the arm -- and, most important, a computer running a very large neural network that lets the robot learn.

The robot worked out how to make use of simple implements, including a dustpan and broom and a duster, to move other objects around. The work hints at how robots might someday learn to perform sophisticated manipulations, and solve abstract problems, for themselves. "It's exciting because it means the robot can figure out what to do with a tool in situations it hasn't seen before," Finn says. "We really want to study that sort of generality, rather than a robot learning to use one tool."

Scientists in southern China report that they've created several transgenic macaque monkeys with extra copies of a human gene suspected of playing a role in shaping human intelligence. "According to their findings, the modified monkeys did better on a memory test involving colors and block pictures, and their brains also took longer to develop -- as those of human children do," reports MIT Technology Review. "There wasn't a difference in brain size." From the report: The experiments, described on March 27 in a Beijing journal, National Science Review, and first reported by Chinese media, remain far from pinpointing the secrets of the human mind or leading to an uprising of brainy primates. Bing Su, the geneticist at the Kunming Institute of Zoology who led the effort, specializes in searching for signs of "Darwinian selection" -- that is, genes that have been spreading because they're successful. His quest has spanned such topics as Himalayan yaks' adaptation to high altitude and the evolution of human skin color in response to cold winters. [Instead of the FOXP2 gene famous for its potential link to human speech] Su was fascinated by a different gene: MCPH1, or microcephalin. Not only did the gene's sequence differ between humans and apes, but babies with damage to microcephalin are born with tiny heads, providing a link to brain size. With his students, Su once used calipers and head spanners to the measure the heads of 867 Chinese men and women to see if the results could be explained by differences in the gene.

By 2010, though, Su saw a chance to carry out a potentially more definitive experiment -- adding the human microcephalin gene to a monkey. China by then had begun pairing its sizable breeding facilities for monkeys (the country exports more than 30,000 a year) with the newest genetic tools, an effort that has turned it into a mecca for foreign scientists who need monkeys to experiment on. To create the animals, Su and collaborators at the Yunnan Key Laboratory of Primate Biomedical Research exposed monkey embryos to a virus carrying the human version of microcephalin. They generated 11 monkeys, five of which survived to take part in a battery of brain measurements. Those monkeys each have between two and nine copies of the human gene in their bodies. After putting the monkeys inside MRI machines to measure their white matter, they gave them computerized memory tests. "According to their report, the transgenic monkeys didn't have larger brains, but they did better on a short-term memory quiz, a finding the team considers remarkable," reports MIT Technology Review.

An anonymous reader quotes a report from MIT Technology Review: Machine learning has been used to create basil plants that are extra-delicious. While we sadly cannot report firsthand on the herb's taste, the effort reflects a broader trend that involves using data science and machine learning to improve agriculture. The researchers behind the AI-optimized basil used machine learning to determine the growing conditions that would maximize the concentration of the volatile compounds responsible for basil's flavor. The study appears in the journal PLOS One today.

The basil was grown in hydroponic units within modified shipping containers in Middleton, Massachusetts. Temperature, light, humidity, and other environmental factors inside the containers could be controlled automatically. The researchers tested the taste of the plants by looking for certain compounds using gas chromatography and mass spectrometry. And they fed the resulting data into machine-learning algorithms developed at MIT and a company called Cognizant. The research showed, counterintuitively, that exposing plants to light 24 hours a day generated the best taste. The research group plans to study how the technology might improve the disease-fighting capabilities of plants as well as how different flora may respond to the effects of climate change.

An anonymous reader quotes a report from MIT Technology Review: Almost a thousand Google staff, academic researchers, and other tech industry figures have signed a letter protesting the makeup of an independent council that Google created to guide the ethics of its AI projects. The search giant announced the creation of the council last week at EmTech Digital, MIT Technology Review's event in San Francisco. Known as the Advanced Technology External Advisory Council (ATEAC), it has eight members including economists, philosophers, policymakers, and technologists with expertise in issues like algorithmic bias. It is meant to hold four meetings a year, starting this month, and write reports designed to provide feedback on projects at the company that use artificial intelligence.

But two of those members proved controversial. One, Dyan Gibbens, is CEO of Trumbull, a company that develops autonomous systems for the defense industry -- a contentious choice given that thousands of Google employees protested the company's decision to supply the US Air Force with AI for drone imaging. The greatest outrage, though, has come over the inclusion of Kay Coles James, president of the Heritage Foundation, a think tank that opposes regulating carbon emissions, takes a hard line on immigration, and has argued against the protection of LGBTQ rights. The creation of ATEAC -- and the inclusion of Gibbens and James -- may in fact have been designed to appease Google's right-wing critics. At roughly the same time the council was announced, Sundar Pichai, Google's CEO, was meeting with President Donald Trump. Trump later tweeted: "He stated strongly that he is totally committed to the U.S. Military, not the Chinese Military. [We] also discussed political fairness and various things that Google can do for our Country. Meeting ended very well!" "Not only are James' views counter to Google's stated values," the letter states, "but they are directly counter to the project of ensuring that the development and application of AI prioritizes justice over profit. Such a project should instead place representatives from vulnerable communities at the center of decision-making."

Since 2016, Microsoft has been working with the University of Washington to develop the first device to automatically encode digital information into DNA and back to bits again. "So far, DNA storage has been carried out by hand in the lab," reports MIT Technology Review. But now Microsoft and researchers at the University of Washington "say they created a machine that converts electronic bits to DNA and back without a person involved." From the report: The gadget, made from about $10,000 in parts, uses glass bottles of chemicals to build DNA strands, and a tiny sequencing machine from Oxford Nanopore to read them out again. According to a publication on March 21 in the journal Nature Scientific Reports, the team was able to store and retrieve just a single word -- "hello" -- or five bytes of data. What's more, the process took 21 hours, mostly because of the slow chemical reactions involved in writing DNA. While the team considered that a success for their prototype, a commercially useful DNA storage system would have to store data millions of times faster.

MIT Technology reports of how Huawei's technology road map, especially in the field of artificial intelligence, is progressing more rapidly than any other business in the world. "The [Chinese] government and private sector approach is to build companies that compete across the full tech stack," says Samm Sacks, who specializes in cybersecurity and China at New America, a Washington think tank. "That's what Huawei is doing." Huawei's AI strategy "will also raise a host of new security issues," the report notes. "The company's technological ubiquity, and the fact that Chinese companies are ultimately answerable to their government, are big reasons why the U.S. views Huawei as an unprecedented national security threat." From the report: In an exclusive interview with MIT Technology Review, Xu Wenwei, director of the Huawei board and the company's chief strategy and marketing officer, touted the scope of its AI plans. He also defended the company's record on security. And he promised that Huawei would seek to engage with the rest of the world to address emerging risks and threats posed by AI. Xu (who uses the Western name William Xu) said that Huawei plans to increase its investments in AI and integrate it throughout the company to "build a full-stack AI portfolio." Since Huawei is a private firm, it's tricky to quantify its technology investments. But officials from the company said last year that it planned to more than double annual R&D spending to between $15 billion and $20 billion. This could catapult the company to between fifth and second place in worldwide spending on R&D. According to its website, some 80,000 employees, or 45% of Huawei's workforce, are involved in R&D.

Machine-learning services are a new source of risk, since they can be exploited by hackers, and the data used to train such services may contain private information. The use of AI algorithms also makes systems more complex and opaque, which means security auditing is more challenging. As part of an effort to reassure doubters, Xu promised that Huawei would release a code of AI principles in April. This will amount to a promise that the company will seek to protect user data and ensure security. Xu also said Huawei wants to collaborate with its international competitors, which would include the likes of Google and Amazon, to ensure that the technology is developed responsibly. It is, however, unclear whether Huawei might allow its AI services to be audited by a third party, as it has done with its hardware. In other Huawei-related news, Huawei executive Meng Wanzhou is suing Canada for violating her constitutional rights when border officials detained and interrogated her for hours. "Meng, the chief financial officer of the Chinese telecom firm Huawei, was arrested by Canadian officials in December at the request of the United States," reports NPR. "The U.S. had sought Meng's arrest on charges of fraud, arguing Huawei had violated U.S. sanctions on Iran."

schwit1 shares a report from MIT Technology Review: Early last month, the security team at Coinbase noticed something strange going on in Ethereum Classic, one of the cryptocurrencies people can buy and sell using Coinbase's popular exchange platform. Its blockchain, the history of all its transactions, was under attack. An attacker had somehow gained control of more than half of the network's computing power and was using it to rewrite the transaction history. That made it possible to spend the same cryptocurrency more than once -- known as "double spends." The attacker was spotted pulling this off to the tune of $1.1 million. Coinbase claims that no currency was actually stolen from any of its accounts. But a second popular exchange, Gate.io, has admitted it wasn't so lucky, losing around $200,000 to the attacker (who, strangely, returned half of it days later).

Just a year ago, this nightmare scenario was mostly theoretical. But the so-called 51% attack against Ethereum Classic was just the latest in a series of recent attacks on blockchains that have heightened the stakes for the nascent industry. [...] In short, while blockchain technology has been long touted for its security, under certain conditions it can be quite vulnerable. Sometimes shoddy execution can be blamed, or unintentional software bugs. Other times it's more of a gray area -- the complicated result of interactions between the code, the economics of the blockchain, and human greed. That's been known in theory since the technology's beginning. Now that so many blockchains are out in the world, we are learning what it actually means -- often the hard way.

Sean Dorrance Kelly, a philosophy professor at Harvard, writes for MIT Technology Review: Human creative achievement, because of the way it is socially embedded, will not succumb to advances in artificial intelligence. To say otherwise is to misunderstand both what human beings are and what our creativity amounts to. This claim is not absolute: it depends on the norms that we allow to govern our culture and our expectations of technology. Human beings have, in the past, attributed great power and genius even to lifeless totems. It is entirely possible that we will come to treat artificially intelligent machines as so vastly superior to us that we will naturally attribute creativity to them. Should that happen, it will not be because machines have outstripped us. It will be because we will have denigrated ourselves.

[...] My argument is not that the creator's responsiveness to social necessity must be conscious for the work to meet the standards of genius. I am arguing instead that we must be able to interpret the work as responding that way. It would be a mistake to interpret a machine's composition as part of such a vision of the world. The argument for this is simple. Claims like Kurzweil's that machines can reach human-level intelligence assume that to have a human mind is just to have a human brain that follows some set of computational algorithms -- a view called computationalism. But though algorithms can have moral implications, they are not themselves moral agents. We can't count the monkey at a typewriter who accidentally types out Othello as a great creative playwright. If there is greatness in the product, it is only an accident. We may be able to see a machine's product as great, but if we know that the output is merely the result of some arbitrary act or algorithmic formalism, we cannot accept it as the expression of a vision for human good.

An anonymous reader quotes a report from MIT Technology Review: The brains of two genetically-edited girls born in China last year may have been changed in ways that enhance cognition and memory, scientists say. The twins, called Lulu and Nana, were modified using CRISPR, a new gene-editing tool, by a Chinese scientific team to make the girls immune to infection by HIV, the virus that causes AIDS. Now, new research shows the same genetic alteration introduced into the girls' DNA, to a gene called CCR5, not only makes mice smarter, but also improves human brain recovery after stroke, and could be linked to greater success in school.

"The answer is likely yes, it did affect their brains," says Alcino J. Silva, a neurobiologist at the University of California, Los Angeles. Silva's lab lab has been uncovering a major new role for the CCR5 gene in memory formation and the brain's ability to form new connections. "The simplest interpretation is that those mutations will probably have an impact on cognitive function in the twins," says Silva. He says the exact effect on the girls cognition is impossible to predict and "that is why it should not be done." The Chinese designer babies were created to be resistant to HIV. A team in Shenzhen, China, led by Southern University of Science and Technology He Jiankui used the gene-editing tool CRISPR to delete a single gene, called CCR5, from human embryos, some of which were later used to create pregnancies. The virus that causes AIDS requires the CCR5 gene to enter human blood cells. The scientist, He Jiankui of the Southern University of Science and Technology in Shenzhen, China, has been fired from the university as He is under investigation in China. There is no evidence that He actually set out to modify the twins' intelligence.

From a report: A new robot can navigate without GPS, using the same light-sensing abilities as desert ants. Desert ants survive in searingly hot conditions in the Sahara. They sometimes have just a few minutes to forage for food before they risk burning to death. As a result, they are very efficient navigators, using bands of polarized light, invisible to humans, to get around. They also carefully count their steps. These two tactics help to keep them alive. AntBot: The bot, described this week in Science Robotics, is fitted with UV light sensors that can detect polarized light from the sun. This is known as a "celestial compass" and is designed to mimic the way desert ants see the sun's light. This helps it work out the direction it's going in. The robot also counts its steps, much like its desert muse. In tests, it successfully managed to complete an outdoor homing task, where it was required to go to several checkpoints and then return to a fixed location within a range of 14 meters.

According to a new study published in the journal Geophysical Research Letters, NASA glaciologists found a crater thought to be more than 22 miles wide. "It is only 114 miles from the Hiawatha impact crater that was discovered in 2018," MIT Technology Review reports. "The identification of that first crater led NASA to dedicate additional resources for investigating the land under Greenland's ice." From the report: NASA glaciologists used topographical maps, satellite images, and radar scans to analyze the area. What they found was a flat, bowl-shaped depression in the bedrock. This was surrounded by an elevated edge and characteristic central peaks, which form on the crater floor after an impact. The crater has eroded significantly over time, causing the team to estimate it was created somewhere between a hundred thousand years and a hundred million years ago. That suggests it probably wasn't formed at the same time as the Hiawatha crater, which is younger. This would be the third pair of craters that sit close to one another that we've found on Earth. "We've surveyed the Earth in many different ways, from land, air, and space. It's exciting that discoveries like these are still possible," says Joe MacGregor, a glaciologist with NASA's Goddard Space Flight Center.

Researchers at the Allen Institute for AI (Ai2) believe that Pictionary could push machine intelligence beyond its current limits. To that end, they have devised an online version of the game that pairs a human player with an AI program. MIT Technology Review reports: In case you've never played it before, Pictionary involves trying to draw an image that conveys a written word or phrase for your teammates to guess. This tests a person's drawing skills but also the ability to convey complex meaning using simple concepts. Given the phrase "wedding ring," for example, a player might try to draw the object itself but also a bride and groom or a wedding ceremony.

That makes it the perfect vehicle to help teach machines. The team developed an online version of the game, called Iconary, that pairs a user with an AI bot called AllenAI. Both take turns as the artist and the guesser. Playing as artist, a user is given a phrase and then has to sketch things to convey it. The sketches are first turned into clip-art icons using computer vision; then the computer program tries to guess the phrase using a database of words and concepts and the relationship between them. If the program gets only part of the phrase, it will ask for another image to clarify. The AI program uses a combination of AI techniques to draw and guess. Over time, by playing against enough people, AllenAI should learn from their common-sense understanding of how concepts (like "books" and "pages") go together in everyday life, Fahadi says. It will also help the researchers explore ways for humans and machines to communicate and collaborate more effectively.

A research team from the University of California, Berkeley, and Lawrence Livermore National Laboratory has created a printer -- nicknamed the "replicator" -- that shines light onto specific spots in a rotating resin that solidifies when exposed to a certain light level. What this does is forms the entire item all in one go, rather that forming items by laying down one layer of material at a time, like most 3D printers. MIT Technology Review reports: "We've carried out a range of prints taking from 30 seconds to a few minutes," says senior author Hayden Taylor. He reckons that printing the same objects in the traditional way could take more than an hour. While the machine competes on speed, it still cannot match the details and size that other printers can achieve. The biggest item it can print right now is just four inches (10 centimeters) in diameter. Other printers can make things measured in meters. The sophistication of the machine lies in the software that creates intricate light patterns to accurate solidify the material.

The printer itself is fairly straightforward. It uses an off-the-shelf video projector plugged into a laptop that projects images of what you want to create, while a motor turns the cylinder of resin. Taylor thinks that because it's so relatively uncomplicated, both commercial and at-home versions of the printer are feasible. "The barrier to creating a very simple version of this tool is not that high," he says.

"China hasn't launched a new nuclear reactor build for over two years, but Chinese press reports that this nuclear hiatus has broken," writes Slashdot reader carbonnation. "Approvals have reportedly been made for four Hualong One reactors -- a domestic "Generation III" design -- instead of U.S.-designed AP1000s." From a report via MIT Technology Review: China's Jiemian News started the chatter on Tuesday with an exclusive interview with senior leadership of the Hualong One design's owner, Hualong International Nuclear Power Technology, a collaboration of nuclear heavyweights China General Nuclear Power (CGN) and China National Nuclear Corp. (CNNC). According to the news site, the joint venture's leaders said that two dual-reactor projects had received provisional permission to begin pouring concrete. Other publications also picked up the story yesterday, including First Financial Journal, which claimed to have confirmed the approvals through "relevant authoritative channels." CNNC and CGN have not responded to the media reports.

The reactors are slated for two new sites along China's coast: CNNC's Zhangzhou power project in Fujian and CGN's Huizhou Taipingling project in Guangdong. Both projects had been planned and approved by Chinese authorities with Westinghouse's AP1000 reactor design, which promises safety advances such as passive cooling. That means it stores water above the reactor, leveraging gravity to keep the plant cool should the pumps fail. But Westinghouse's flagship AP1000 projects have been plagued by cost overruns and delays. Those troubles may have helped the Hualong One to catch up. CNNC started building the first Hualong One reactor in 2015 at its Fuqing power plant and expects to have it operating later this year.

Facebook's new plan to integrate WhatsApp, Instagram and Facebook Messenger will lead to more data about users being shared between them, a new report warns. The effort to make it easier for people to participate in conversations across its various messaging platforms sounds harmless, but it raises issues about how data will be shared across the platforms, and with third parties. The good news is that the apps will all be required to use end-to-end encryption. MIT Technology Review reports: Facebook says it wants to make it easier for people to communicate across its "ecosystem" of apps. But the real driver here is a commercial one. By making it easier to swap messages, Facebook can mine even more data to target ads with, and come up with more money-spinning services. There's another potential benefit: by integrating its messaging apps more tightly, Facebook can argue it would be harder to spin one or more of them off, as some antitrust campaigners think it should be forced to do.

At the Data for Black Lives conference last weekend, technologists, legal experts, and community activists snapped the kind of impact AI has on our lives into perspective with a discussion of America's criminal justice system. There, an algorithm can determine the trajectory of your life. From a report: The US imprisons more people than any other country in the world. At the end of 2016, nearly 2.2 million adults were being held in prisons or jails, and an additional 4.5 million were in other correctional facilities. Put another way, 1 in 38 adult Americans was under some form of correctional supervision. The nightmarishness of this situation is one of the few issues that unite politicians on both sides of the aisle.

Under immense pressure to reduce prison numbers without risking a rise in crime, courtrooms across the US have turned to automated tools in attempts to shuffle defendants through the legal system as efficiently and safely as possible. This is where the AI part of our story begins. Police departments use predictive algorithms to strategize about where to send their ranks. Law enforcement agencies use face recognition systems to help identify suspects. These practices have garnered well-deserved scrutiny for whether they in fact improve safety or simply perpetuate existing inequities.

Researchers and civil rights advocates, for example, have repeatedly demonstrated that face recognition systems can fail spectacularly, particularly for dark-skinned individuals -- even mistaking members of Congress for convicted criminals. But the most controversial tool by far comes after police have made an arrest. Say hello to criminal risk assessment algorithms.

An anonymous reader quotes a report from MIT Technology Review: Algorithms are increasingly being used to make ethical decisions. They are built to pursue a single mathematical goal, such as maximizing the number of soldiers' lives saved or minimizing the number of civilian deaths. When you start dealing with multiple, often competing, objectives or try to account for intangibles like "freedom" and "well-being," a satisfactory mathematical solution doesn't always exist. "We as humans want multiple incompatible things," says Peter Eckersley, the director of research for the Partnership on AI, who recently released a paper that explores this issue. "There are many high-stakes situations where it's actually inappropriate -- perhaps dangerous -- to program in a single objective function that tries to describe your ethics." These solutionless dilemmas aren't specific to algorithms. Ethicists have studied them for decades and refer to them as impossibility theorems. So when Eckersley first recognized their applications to artificial intelligence, he borrowed an idea directly from the field of ethics to propose a solution: what if we built uncertainty into our algorithms?

Eckersley puts forth two possible techniques to express this idea mathematically. He begins with the premise that algorithms are typically programmed with clear rules about human preferences. We'd have to tell it, for example, that we definitely prefer friendly soldiers over friendly civilians, and friendly civilians over enemy soldiers -- even if we weren't actually sure or didn't think that should always be the case. The algorithm's design leaves little room for uncertainty. The first technique, known as partial ordering, begins to introduce just the slightest bit of uncertainty. You could program the algorithm to prefer friendly soldiers over enemy soldiers and friendly civilians over enemy soldiers, but you wouldn't specify a preference between friendly soldiers and friendly civilians. In the second technique, known as uncertain ordering, you have several lists of absolute preferences, but each one has a probability attached to it. Three-quarters of the time you might prefer friendly soldiers over friendly civilians over enemy soldiers. A quarter of the time you might prefer friendly civilians over friendly soldiers over enemy soldiers. The algorithm could handle this uncertainty by computing multiple solutions and then giving humans a menu of options with their associated trade-offs, Eckersley says.

In the era of AI superpowers, Finland is no match for the US and China. So the Scandinavian country is taking a different tack. From a report: It has embarked on an ambitious challenge to teach the basics of AI to 1% of its population, or 55,000 people. Once it reaches that goal, it plans to go further, increasing the share of the population with AI know-how. The scheme is all part of a greater effort to establish Finland as a leader in applying and using the technology.

Citizens take an online course that is specifically designed for non-technology experts with no programming experience. The government is now rolling it out nationally. As of mid-December, more than 10,500 people, including at least 4,000 outside of Finland's borders, had graduated from the course. More than 250 companies have also pledged to train part or all of their workforce.

An anonymous reader quotes a report from MIT Technology Review: As good as they are at causing mischief, researchers from the MIT-IBM Watson AI Lab realized GANs, or generative adversarial networks, are also a powerful tool: because they paint what they're "thinking," they could give humans insight into how neural networks learn and reason. [T]he researchers began probing a GAN's learning mechanics by feeding it various photos of scenery -- trees, grass, buildings, and sky. They wanted to see whether it would learn to organize the pixels into sensible groups without being explicitly told how. Stunningly, over time, it did. By turning "on" and "off" various "neurons" and asking the GAN to paint what it thought, the researchers found distinct neuron clusters that had learned to represent a tree, for example. Other clusters represented grass, while still others represented walls or doors. In other words, it had managed to group tree pixels with tree pixels and door pixels with door pixels regardless of how these objects changed color from photo to photo in the training set.

Not only that, but the GAN seemed to know what kind of door to paint depending on the type of wall pictured in an image. It would paint a Georgian-style door on a brick building with Georgian architecture, or a stone door on a Gothic building. It also refused to paint any doors on a piece of sky. Without being told, the GAN had somehow grasped certain unspoken truths about the world. Being able to identify which clusters correspond to which concepts makes it possible to control the neural network's output. The team has now released an app called GANpaint that turns this newfound ability into an artistic tool. It allows you to turn on specific neuron clusters to paint scenes of buildings in grassy fields with lots of doors. Beyond its silliness as a playful outlet, it also speaks to the greater potential of this research.

As America's government faces its longest-ever shutdown over the president's demands for border wall funding, House Speaker Nancy Pelosi has suggested "possible alternatives to a physical wall," according to one Silicon Valley newspaper: Among the president's justifications for a wall is to stop drugs from coming into the United States, so Pelosi proposed spending "hundreds of millions of dollars" for technology to scan cars for drugs, weapons and contraband at the border. "The positive, shall we say, almost technological wall that can be built is what we should be doing," Pelosi, D-San Francisco, said during her weekly press conference.

That didn't go over well with Fight for the Future, a digital rights advocacy group that on Friday started a petition asking Democrats to drop plans for a "technological wall" that it says could threaten Fourth Amendment rights that guard against unreasonable searches and seizures. "Current border surveillance programs subject people to invasive and unconstitutional searches of their cell phones and laptops, location tracking, drone surveillance, and problematic watchlists," the group's petition says...

In December, the Department of Homeland Security's Office of the Inspector General released a report that showed searches of electronic devices at the border were up nearly 50 percent in 2017. The report also found that border agents were not always following standard operating procedures for searches, including failing to properly document such searches. In addition, information copied by agents were not always deleted as required.
The article also notes that Anduril Industries -- founded by Oculus Rift designer Palmer Luckey (and funded by Peter Thiel) -- is one of several companies already working on "a virtual border wall."

CNN also reports on a GoFundMe campaign started by an Air Force veteran to simply crowdfund the construction of the wall. Though 340,747 people pledged over $20 million, it failed to reach its $1 billion goal, and is now pointing supporters to a newly-formed non-profit corporation -- named "We Build the Wall."

Meanwhile, another 7,121 GoFundMe members have pledged $160,985 to a rival campaign raising money for ladders to climb over Trump's wall.

An anonymous reader shares a report: In 2018, several high-profile controversies involving AI served as a wake-up call for technologists, policymakers, and the public. The technology may have brought us welcome advances in many fields, but it can also fail catastrophically when built shoddily or applied carelessly. It's hardly a surprise, then, that Americans have mixed support for the continued development of AI and overwhelmingly agree that it should be regulated, according to a new study from the Center for the Governance of AI and Oxford University's Future of Humanity Institute. These are important lessons for policymakers and technologists to consider in the discussion on how best to advance and regulate AI, says Allan Dafoe, director of the center and coauthor of the report. "There isn't currently a consensus in favor of developing advanced AI, or that it's going to be good for humanity," he says. "That kind of perception could lead to the development of AI being perceived as illegitimate or cause political backlashes against the development of AI."

Perhaps the most insightful piece that sums up why the U.S. and its allies are apprehensive of using Huawei's products. Six reasons, we are just highlighting the pointers, click on the source story to read the description:
1. There could be "kill switches" in Huawei equipment.
2. ... That even close inspections miss.
3. Back doors could be used for data snooping.
4. The rollout of 5G wireless networks will make everything worse.
5. Chinese firms will ship tech to countries in defiance of a US trade embargo.
6. Huawei isn't as immune to Chinese government influence as it claims to be.

An anonymous reader quotes MIT's Technology Review: Silicon Valley loves the idea of universal basic income. Many in the tech elites tout it as the answer to job losses caused by automation, if only people would give it a chance.... Getting people on board with basic income requires data, which is what numerous tests have been trying to obtain. But this year, a number of experiments were cut short, delayed, or ended after a short time. That also means the possible data supply got cut off.

Back in June we declared, "Basic income could work -- if you do it Canada style." We talked to the people on the ground getting the checks in Ontario's 4,000-person test and saw how it was changing the community. Then, just two months later, it was announced that the program is ending in the new year rather than running for three years. The last checks will be delivered to participants in March 2019.
The article complains that in addition, Finland's test program ended this year after its initial trial period, while Y Combinator's experiment "has also faced more delays, pushing the experiment into 2019," saying these programs illustrate the three basic issues faced by basic income tests. First, there's political disagreements. ("The Ontario program was shut down by the province's newly installed Conservative government.") Then there's also concerns about funding -- "As you might imagine, giving away free money is expensive" -- and also fears about disrupting existing benefits "To avoid that, they've had to work with municipal and state agencies to get waivers for pilot recipients. But getting those waivers takes a lot of time and bureaucracy....

"The only way the idea can ever be embraced on any sort of large-scale, meaningful level is with more data and bigger tests. Without that, no matter how much support it gets from Silicon Valley, it seems unlikely that the public, at least in the US, will ever come around."

An anonymous reader quotes MIT's Technology Review: Silicon Valley loves the idea of universal basic income. Many in the tech elites tout it as the answer to job losses caused by automation, if only people would give it a chance.... Getting people on board with basic income requires data, which is what numerous tests have been trying to obtain. But this year, a number of experiments were cut short, delayed, or ended after a short time. That also means the possible data supply got cut off.

Back in June we declared, "Basic income could work -- if you do it Canada style." We talked to the people on the ground getting the checks in Ontario's 4,000-person test and saw how it was changing the community. Then, just two months later, it was announced that the program is ending in the new year rather than running for three years. The last checks will be delivered to participants in March 2019.
The article complains that in addition, Finland's test program ended this year after its initial trial period, while Y Combinator's experiment "has also faced more delays, pushing the experiment into 2019," saying these programs illustrate the three basic issues faced by basic income tests. First, there's political disagreements. ("The Ontario program was shut down by the province's newly installed Conservative government.") Then there's also concerns about funding -- "As you might imagine, giving away free money is expensive" -- and also fears about disrupting existing benefits "To avoid that, they've had to work with municipal and state agencies to get waivers for pilot recipients. But getting those waivers takes a lot of time and bureaucracy....

"The only way the idea can ever be embraced on any sort of large-scale, meaningful level is with more data and bigger tests. Without that, no matter how much support it gets from Silicon Valley, it seems unlikely that the public, at least in the US, will ever come around."

Privately funded space startups are changing China's space industry, helping it become a space power on par with the United States. "2018 is shaping up to be the first year in which more rockets reach Earth orbit from China than from any other country," reports MIT Technology Review. "As of mid-December, China had made 35 successful launches, as against 30 for the U.S."

"As American and Russian space programs struggle with uncertain budgets, China is expanding its efforts on every front: communications and reconnaissance satellites; a navigation and positioning constellation to rival America's GPS; a human spaceflight program; and ambitious space-science and robotic exploration projects. All of these are enabled by a menagerie of new rockets with advanced capabilities." Here's an excerpt from the report summarizing some of China's space ambitions: In 2014, the Chinese government decided to allow private investment in space-related industry. Landspace began with a few dozen people. It now has over 200 employees at a manufacturing base in Huzhou in eastern China and at assembly and testing facilities in X'ian, a central Chinese city. The company plans to work incrementally, beginning with nano-satellites -- devices weighing between 1 and 10 kilograms (2 to 22 pounds) -- then moving to larger cargoes and, eventually, into human spaceflight. In September 2018, iSpace launched three nanosatellites on a brief suborbital flight, becoming the first Chinese space startup to successfully get beyond Earth's atmosphere. Another company, LinkSpace, plans to launch a vertical takeoff, vertical landing rocket in 2020. Landspace, OneSpace, iSpace, LinkSpace, and ExPace (which fashions itself as a startup though it's a subsidiary of a state-owned enterprise) are the leaders of a bevy of lesser-known Chinese launch startups.

These launch companies are operating hand in hand with a number of new, privately funded Chinese companies that are focused on doing things in space, rather than on getting there. Spacety and Commsat, among others, are planning large constellations of small imagery and communication satellites. Such constellations -- whether Chinese or American -- are transforming aspects of the way space is used. By making low-resolution satellite imagery much cheaper to gather (among other novel applications for small satellites), they are catalyzing an era of more nimble commercial, scientific, and military experimentation.

The blockchain system has daunting technical problems to fix. But first, its disciples need to figure out how to govern themselves. From a report: The handful of idealistic researchers, developers, and administrators in charge of maintaining its software are under increasing pressure to overcome technical limitations that stymie the network's growth. At the same time, well-funded competitors have emerged, claiming that their blockchains perform better. Crackdowns by regulators, and a growing understanding of how far most blockchain applications are from ready for prime time, have scared many cryptocurrency investors away: Ethereum's market value in dollars has fallen more than 90% since its peak last January.

The reason Devcon (the annual "family reunion" organized by the Ethereum Foundation; this year's edition was held in October) feels so upbeat despite these storm clouds is that the people building Ethereum have something bigger in mind -- something world-changing, in fact. Yet to achieve its goal, this ragtag community needs to crack a problem as complicated as any of the toe-curling technical challenges it faces: how to govern itself. It must find a way to organize a scattered global network of contributors and stakeholders without sacrificing "decentralization" -- the principle, which any cryptocurrency community strives for, that no one entity or group should be in control.

Chemists have found a material that can display superconducting behavior at a temperature warmer than it currently is at the North Pole. The work brings room-temperature superconductivity tantalizingly close.

From a report: The work comes from the lab of Mikhail Eremets and colleagues at the Max Planck Institute for Chemistry in Mainz, Germany. Eremets and his colleagues say they have observed lanthanum hydride (LaH10) superconducting at the sweltering temperature of 250 K, or -23C. That's warmer than the current temperature at the North Pole.

"Our study makes a leap forward on the road to the room-temperature superconductivity," say the team. (The caveat is that the sample has to be under huge pressure: 170 gigapascals, or about half the pressure at the center of the Earth.)

Like Google's Duplex, Chinese internet giant Alibaba has its own humanlike voice assistant capable of making restaurant reservations and salon appointments. But unlike Google, which has rolled out the feature to select audience, Alibaba's offering already has a wider reach. And it's smart, too. From a report: On December 2 at the 2018 Neural Information Processing Systems conference, one of the largest annual gatherings for AI research, Alibaba demoed the AI customer service agent for its logistics company Cainiao. Jin Rong, the dean of Alibaba's Machine Intelligence and Technology Lab, said the agent is already servicing millions of customer requests a day. The pre-recorded demo call involved the agent asking a customer where he wanted his package delivered. In the back-and-forth exchange, the agent successfully navigated several conversational elements that demonstrated the breadth of its natural-language capabilities.

Take this exchange at the beginning of the call, translated from Mandarin: Agent: Hello, I am Cainiao's voice assistant. I am --
Customer: Hello.
A: Yes, hi, you have package scheduled for morning delivery to 588 Culture West Road. Is it convenient for you to receive?
C: Who are you?
A: I am Cainiao's voice assistant. I'd like to confirm your morning delivery to 588 Culture West Road. Does that work for you?
C: I'm not home in the morning.
A: Then do you have another address that you'd like to use?

Within 30 seconds, the agent has smoothly handled three common, and tricky, conversational ingredients: interruption, nonlinear conversation, and implicit intent. Interruption is self-explanatory: the agent can respond to the customer's interruption and continue relaying relevant information without starting over or skipping a beat.

A new report (PDF) from the AINow Institute calls for the U.S. government to take general steps to improve the regulation of facial recognition technology amid much debate over the privacy implications. "The implementation of AI systems is expanding rapidly, without adequate governance, oversight, or accountability regimes," it says. The report suggests, for instance, extending the power of existing government bodies in order to regulate AI issues, including use of facial recognition: "Domains like health, education, criminal justice, and welfare all have their own histories, regulatory frameworks, and hazards." MIT Technology Review reports: It also calls for stronger consumer protections against misleading claims regarding AI; urges companies to waive trade-secret claims when the accountability of AI systems is at stake (when algorithms are being used to make critical decisions, for example); and asks that they govern themselves more responsibly when it comes to the use of AI. And the document suggests that the public should be warned when facial-recognition systems are being used to track them, and that they should have the right to reject the use of such technology.

The report also warns about the use of emotion tracking in face-scanning and voice detection systems. Tracking emotion this way is relatively unproven, yet it is being used in potentially discriminatory ways -- for example, to track the attention of students. "It's time to regulate facial recognition and affect recognition," says Kate Crawford, cofounder of AINow and one of the lead authors of the report. "Claiming to 'see' into people's interior states is neither scientific nor ethical."

An anonymous reader quotes a report from MIT Technology Review: The world relies on encryption to protect everything from credit card transactions to databases holding health records and other sensitive information. A new report from the U.S. National Academies of Sciences, Engineering, and Medicine says we need to speed up preparations for the time when super-powerful quantum computers can crack conventional cryptographic defenses. The experts who produced the report, which was released today, say widespread adoption of quantum-resistant cryptography "will be a long and difficult process" that "probably cannot be completed in less than 20 years." It's possible that highly capable quantum machines will appear before then, and if hackers get their hands on them, the result could be a security and privacy nightmare.

Today's cyberdefenses rely heavily on the fact that it would take even the most powerful classical supercomputers almost unimaginable amounts of time to unravel the cryptographic algorithms that protect our data, computer networks, and other digital systems. But computers that harness quantum bits, or qubits, promise to deliver exponential leaps in processing power that could break today's best encryption. The report cites an example of encryption that protects the process of swapping identical digital keys between two parties, who use them to decrypt secure messages sent to one another. A powerful quantum computer could crack RSA-1024, a popular algorithmic defense for this process, in less than a day. The U.S., Israel and others are working to develop standards for quantum-proof cryptographic algorithms, but they may not be ready or widely adopted by the time quantum computers arrive.

"[I]t will take at least a couple of decades to get quantum-safe cryptography broadly in place," the report says in closing. "If that holds, we're going have to hope it somehow takes even longer before a powerful quantum computer ends up in a malicious hacker's hands."

Even as a furious debate broke out in China over gene-edited babies, some scientists in the US are also hoping to improve tomorrow's children. From a report: [...] Amid the condemnation, though, it was easy to lose track of what the key experts were saying. Technology to alter heredity is for real. It is improving very quickly, it has features that will make it safe, and much wider exploratory use to create children could be justified soon. That was the message delivered at a gene-editing summit in Hong Kong on Wednesday, by Harvard Medical School dean George Daley, just ahead of He's own dramatic appearance on the stage (see video starting at 1:15:30).

Astounding some listeners, the Harvard doctor and stem-cell researcher didn't condemn He but instead characterized the Chinese actions as a wrong turn on the right path (see video). "The fact that it is possible that the first instance of human germ-line editing came forward as a misstep should in no way lead us to stick our heads in the sand," Daley said. "It's time to ... start outlining what an actual pathway for clinical translation would be."

An estimated 4 billion people in the world lack a physical address. Researchers at the MIT Media Lab and Facebook are now proposing a new way to address the unaddressed: with machine learning. From a report: The team first trained a deep-learning algorithm to extract the road pixels from satellite images. Another algorithm connected the pixels together into a road network. The system analyzed the density and shape of the roads to segment the network into different communities, and the densest cluster was labeled as the city center. The regions around the city center were divided into north, south, east, and west quadrants, and streets were numbered and lettered according to their orientation and distance from the center.

When they compared their final results with a random sample of unmapped regions whose streets had been labeled manually, their approach successfully addressed more than 80% of the populated areas, improving coverage compared with Google Maps or OpenStreetMaps. This isn't the only way to automate the creation of addresses. The organization what3words generates a unique three-word combination for every 3-by-3-meter square on a global grid. The scheme has already been adopted in regions of South Africa, Turkey, and Mongolia by national package delivery services, local hospitals, and regional security teams. But Ilke Demir, a researcher at Facebook and one of the creators of the new system, says its main advantage is that it follows existing road topology and helps residents understand how two addresses relate to one another.

At its annual re:Invent conference in Seattle this week, Amazon unveiled a service that lets owners of satellites rent time on Amazon-managed ground stations to send and receive data from orbit. "The service, called AWS Ground Station, works in much the same way as Amazon's well-established business for tapping computing capacity via the cloud," reports MIT Technology Review. From the report: According to an AWS blog post, big businesses with a large number of satellites typically build and operate their own ground stations at a cost of a million dollars or more for each one. Smaller companies that can't afford their own often end up signing inflexible, long-term contracts with third parties that own and run such stations. The new service will let satellite operators get access to a ground station at short notice on a pay-as-you-go basis. Those who know how much capacity they will need well in advance can book ahead and pay less for downlink time. AWS is kicking off with a pair of ground stations and says it will have a total of a dozen up and running by the middle of next year. It will monitor how demand develops before deciding how many more stations to add.

An algorithm that remembers previous explorations in Montezuma's Revenge and Pitfall! could make computers and robots better at learning how to succeed in the real world. From a report: A new kind of machine-learning algorithm just mastered a couple of throwback video games that have proved to be a big headache for AI. Those following along will know that AI algorithms have bested the world's top human players at the ancient, elegant strategy game Go, one of the most difficult games imaginable. But two pixelated classics from the era of 8-bit computer games -- Montezuma's Revenge and Pitfall! -- have stymied AI researchers. There's a reason for this seeming contradiction. Although deceptively simple, both Montezuma's Revenge and Pitfall! have been immune to mastery via reinforcement learning, a technique that's otherwise adept at learning to conquer video games.

DeepMind, a subsidiary of Alphabet focused on artificial intelligence, famously used it to develop algorithms capable of learning how to play several classic video games at an expert level. Reinforcement-learning algorithms mesh well with most games, because they tweak their behavior in response to positive feedback -- the score going up. The success of the approach has generated hope that AI algorithms could teach themselves to do all sorts of useful things that are currently impossible for machines. The problem with both Montezuma's Revenge and Pitfall! is that there are few reliable reward signals. Both titles involve typical scenarios: protagonists explore blockish worlds filled with deadly creatures and traps. But in each case, lots of behaviors that are necessary to advance within the game do not help increase the score until much later. Ordinary reinforcement-learning algorithms usually fail to get out of the first room in Montezuma's Revenge, and in Pitfall! they score exactly zero.

A diet supplemented with red algae could lessen the huge amounts of greenhouse gases emitted by cows and sheep, if we can just figure out how to grow enough. From a report: In a wooden barn on the edge of campus at the University of California, Davis, cattle line up at their assigned feed slots to snatch mouthfuls of alfalfa hay. This past spring, several of these Holstein dairy cows participated in a study to test a promising path to reducing methane emissions from livestock, a huge source of the greenhouse gases driving climate change. By adding a small amount of seaweed to the animals' feed, researchers found, they could cut the cows' methane production by nearly 60%. Each year, livestock production pumps out greenhouse gases with the equivalent warming effect of more than 7 gigatons of carbon dioxide, roughly the same global impact as the transportation industry. Nearly 40% of that is produced during digestion: cattle, goats, and sheep belch and pass methane, a highly potent, albeit relatively short-lived, greenhouse gas.

If the reductions achieved in the UC Davis study could be applied across the worldwide livestock industry, it would eliminate nearly 2 gigatons of those emissions annually -- about a quarter of United States' total climate pollution each year. Ermias Kebreab, an animal science professor at UC Davis who leads the work, is preparing to undertake a more ambitious study in the months ahead, evaluating whether smaller amounts of a more potent form of seaweed can cut methane emissions even further. Meanwhile, some businesses have begun to explore what could be the harder challenge: Growing it on a massive scale.

If a snowstorm hits Denver, it can delay thousands of packages that travel through the city before reaching their final destinations on the other side of the country. But if UPS knows a storm is coming, what is the most efficient way to divert all those online orders and holiday gifts around the bad weather? UPS grapples with this question every winter. From a report: To help, UPS recently built an online platform that combines machine learning and advanced analytics. The app -- called Network Planning Tools, or NPT for short -- lets the company's engineers view activity at UPS facilities around the world and route shipments to the ones with the most capacity. They can also see details about the packages in transit, including their weight, volume, and delivery deadlines. While UPS already has a system called ORION that maps out last-mile delivery routes, and a program called EDGE focused on upgrading UPS facilities, NPT gives its engineers a bird's-eye view of package volume and distribution across all its pickup and delivery operations.

The app gets some of its smarts from AI, which it uses to create forecasts about package volume and weight based on analysis of historical data. Rob Papetti, who leads NPT development for UPS, says the machine-learning algorithms also analyze decisions the company's engineers made and assess how they affected customer satisfaction and internal costs. "[The app] starts to learn from itself and suggest this option versus that option, based on what enabled us to give our customers better service," he says.

That kind of insight is crucial during the frenetic holiday season. This year, UPS expects to deliver more parcels during that period than ever before --nearly 800 million, up 5% from 2017. In preparation, the company has used the NPT app to identify and eliminate bottlenecks, such as an Illinois facility that was struggling to process packages quickly. "Within a few minutes, we were able to determine how to get around and relieve [the backlog in] that building and still make our service commitments to customers," says Papetti. "Before NPT, that would have taken at least a week." UPS expects the program to save it $100 million to $200 million a year.

The UK government should encourage companies like Apple, Google, and Uber to publish more map data to help the development of technologies like driverless cars and drones, according to a new report by the Open Data Institute. From a report: This sort of data, which includes addresses and city boundaries, fuels tons of everyday services, from parcel and food deliveries to apps like Google Maps and Uber. Internet giants are sitting on top of vast amounts of geospatial data, but it is largely inaccessible to others. The ODI argues it should be as open as possible as a part of "national infrastructure." Analyzing map data can help communities and organizations make decisions across a vast range of sectors -- for example, how to improve access to a school or hospital.

Yoshua Bengio is a grand master of modern artificial intelligence. Alongside Geoff Hinton and Yan LeCun, Bengio is famous for championing a technique known as deep learning that in recent years has gone from an academic curiosity to one of the most powerful technologies on the planet. Here's an excerpt from an interview he gave to MIT Technology Review: MIT TR: What do you make of the idea that there's an AI race between different countries?
Bengio: I don't like it. I don't think it's the right way to do it. We could collectively participate in a race, but as a scientist and somebody who wants to think about the common good, I think we're better off thinking about how to both build smarter machines and make sure AI is used for the well-being of as many people as possible.

MIT TR: Are you worried about just a few AI companies, in the West and perhaps China, dominating the field of AI?
Bengio: Yes, it's another reason why we need to have more democracy in AI research. It's that AI research by itself will tend to lead to concentrations of power, money, and researchers. The best students want to go to the best companies. They have much more money, they have much more data. And this is not healthy. Even in a democracy, it's dangerous to have too much power concentrated in a few hands.

MIT TR:There has been a lot of controversy over military uses of AI. Where do you stand on that?
Bengio: I stand very firmly against.
MIT TR: Even non-lethal uses of AI?
Bengio: Well, I don't want to prevent that. I think we need to make it immoral to have killer robots. We need to change the culture, and that includes changing laws and treaties. That can go a long way. Of course, you'll never completely prevent it, and people say, "Some rogue country will develop these things." My answer is that one, we want to make them feel guilty for doing it, and two, there's nothing to stop us from building defensive technology. There's a big difference between defensive weapons that will kill off drones, and offensive weapons that are targeting humans. Both can use AI.
MIT TR: Shouldn't AI experts work with the military to ensure this happens?
Bengio: If they had the right moral values, fine. But I don't completely trust military organizations, because they tend to put duty before morality. I wish it was different.

California's devastating wildfires are causing unhealthy air conditions for locals breathing in harmful fumes. From a report: In San Francisco, which currently has the second-worst-rated air quality out of any city in the world, one driver was spotted selling N95 respirator masks for $5 apiece. That's significantly above market rate. Right now you can buy a 10-pack of similar masks for about $13 on Amazon. But considering the masks are sold out at many local stores, riders may be willing to dish out the cash for immediate access to the protective gear. Further reading: California needs to reinvent its fire policies, or the death and destruction will go on.

A sensor that can spot the wind direction from miles away will let DARPA's surveillance balloons hover at the very edge of space in one spot indefinitely. An anonymous reader writes: DARPA, the US military's research arm, is currently testing a wind sensor that could allow devices in its Adaptable Lighter-Than-Air (ALTA) balloon program to spot wind speed and direction from a great distance and then make the necessary adjustments to stay in one spot. DARPA has been working on ALTA for some time, but its existence was only revealed in September. "By flying higher we hope to take advantage of a larger range of winds," says ALTA project manager Alex Walan. ALTA will operate even higher than Loon at 75,000 to 90,000 feet (22,900 to 27,400 meters or 14 to 17 miles), where the winds are less predictable. That shouldn't be a problem if the balloon can see exactly where the favorable winds are.

The wind sensor, called Strat-OAWL (short for "stratospheric optical autocovariance wind lidar"), is a new version of one originally designed for NASA satellites. Made by Ball Aerospace, OAWL shines pulses of laser light into the air. A small fraction of the beam is reflected back, and the reflected laser light is gathered by a telescope. The wavelength of the reflected light is changed slightly depending on how fast the air it bounced back from is moving, a change known as doppler shift. By analyzing this shift, OAWL can determine the speed and direction of the wind. Unlike other wind sensors, OAWL looks in two directions at once, giving a better indication of wind speed and direction.

Bill Gates wants to end malaria, and so he's particularly "energized" about gene drives, a technology that could wipe out the mosquitoes that spread the disease. Gates calls the new approach a "breakthrough," but some environmental groups say gene drives are too dangerous to ever use. From a report: Now the sides are headed for a showdown. In a letter circulated this week, scientists funded by the Bill & Melinda Gates Foundation and others are raising the alarm over what they say is an attempt to use a United Nations biodiversity meeting this week in Sharm El-Sheikh, Egypt, to introduce a global ban on field tests of the technology. At issue is a draft resolution by diplomats updating the UN Convention on Biological Diversity, which -- if adopted -- would call on governments to "refrain from" any release of organisms containing engineered gene drives, even as part of experiments. The proposal for a global gene-drive moratorium has been pushed by environmental groups that are also opposed to genetically modified soybeans and corn. They have likened the gene-drive technique to the atom bomb.

In response, the Gates Foundation, based in Seattle, has been funding a counter-campaign, hiring public relations agencies to preempt restrictive legislation and to distribute today's letter. Many of its signatories are directly funded by the foundation. "This is a lobbying game on both sides, to put it bluntly," says Todd Kuiken, who studies gene-drive policy at North Carolina State University. (He says he was asked to sign the Gates letter but declined because he is a technical advisor to the UN.) New technology The gene-drive technique involves modifying a mosquito's DNA so that, when the insect breeds, it spreads a specific genetic change -- one that's bad for its survival.

The British Army is testing out over 70 new technologies, including unmanned vehicles and surveillance drones, in a four-week experiment on one of its biggest training grounds. From a report: What sort of stuff? The department isn't giving out specifics but said the focus will be on "surveillance, long-range, and precision targeting, enhanced mobility and the re-supply of forces, urban warfare and enhanced situational awareness." The development is part of a $1030 million "innovation fund" launched in 2016. The aim? Primarily it's about reducing the danger to troops during combat, according to the UK's Ministry of Defense (MoD). One of the main areas it'll test is "last mile" supply of fuel, food, and ammunition. The exercise will culminate in a simulated battle involving over 200 soldiers to test out the ideas and products.

Weather forecasting is impressively accurate given how changeable and chaotic Earth's climate can be. It's not unusual to get 10-day forecasts with a reasonable level of accuracy. But there is still much to be done. One challenge for meteorologists is to improve their "nowcasting," the ability to forecast weather in the next six hours or so at a spatial resolution of a square kilometer or less. From a report: In areas where the weather can change rapidly, that is difficult. And there is much at stake. Agricultural activity is increasingly dependent on nowcasting, and the safety of many sporting events depends on it too. Then there is the risk that sudden rainfall could lead to flash flooding, a growing problem in many areas because of climate change and urbanization. That has implications for infrastructure, such as sewage management, and for safety, since this kind of flooding can kill. So meteorologists would dearly love to have a better way to make their nowcasts. Enter Blandine Bianchi from EPFL in Lausanne, Switzerland, and a few colleagues, who have developed a method for combining meteorological data from several sources to produce nowcasts with improved accuracy.

Their work has the potential to change the utility of this kind of forecasting for everyone from farmers and gardeners to emergency services and sewage engineers. Current forecasting is limited by the data and the scale on which it is gathered and processed. For example, satellite data has a spatial resolution of 50 to 100 km and allows the tracking and forecasting of large cloud cells over a time scale of six to nine hours. By contrast, radar data is updated every five minutes, with a spatial resolution of about a kilometer, and leads to predictions on the time scale of one to three hours. Another source of data is the microwave links used by telecommunications companies, which are degraded by rainfall.

An anonymous reader quotes a report from MIT Technology Review: In a laboratory that overlooks a busy shopping street in Cambridge, Massachusetts, a robot is attempting to create new materials. A robot arm dips a pipette into a dish and transfers a tiny amount of bright liquid into one of many receptacles sitting in front of another machine. When all the samples are ready, the second machine tests their optical properties, and the results are fed to a computer that controls the arm. Software analyzes the results of these experiments, formulates a few hypotheses, and then starts the process over again. Humans are barely required.

The setup, developed by a startup called Kebotix, hints at how machine learning and robotic automation may be poised to revolutionize materials science in coming years. The company believes it may find new compounds that could, among other things, absorb pollution, combat drug-resistant fungal infections, and serve as more efficient optoelectronic components. The company's software learns from 3-D models of molecules with known properties. Kebotix uses several machine-learning methods to design novel chemical compounds. The company feeds molecular models of compounds with desirable properties into a type of neural network that learns a statistical representation of those properties. This algorithm can then come up with new examples that fit the same model. To strain out potentially useless materials, Kebotix uses another neural network and "then the company's robotic system tests the remaining chemical structures," reports MIT Technology Review. "The results of those experiments can be fed back into the machine-learning pipeline, helping it get closer to the desired chemical properties. The company dubs the overall system a 'self-driving lab.'"

An anonymous reader shares an excerpt from a report via MIT Technology Review: A lumpy disc of dark-gray steel covers a bench in the lab space of Boston Metal, an MIT spinout located a half-hour north of its namesake city. It's the company's first batch of the high-strength alloy, created using a novel approach to metal processing. Instead of the blast furnace employed in steelmaking for centuries, Boston Metal has developed something closer to a battery. Specifically, it's what's known as an electrolytic cell, which uses electricity -- rather than carbon -- to process raw iron ore.

If the technology works at scale as cheaply as the founders hope, it could offer a clear path to cutting greenhouse-gas emissions from one of the hardest-to-clean sectors of the global economy, and the single biggest industrial source of climate pollution. After working on the idea for the last six years, the nine-person company is shifting into its next phase. If it closes a pending funding round, the startup plans to build a large demonstration facility and develop an industrial-scale cell for steel production. The process to produce steel results in around 1.7 gigatons of carbon dioxide being pumped into the atmosphere annually, "adding up to around 5 percent of global carbon dioxide emissions, according to a recent paper in Science," MIT Technology Review reports.

The electrolytic cell that Boston Metal developed was realized after it was proposed to be used to extract oxygen from the moon's surface. "The by-product was molten metal," the report says. "But producing something like steel would require an anode made from cheap materials that wouldn't corrode under high temperatures or readily react with iron oxide. In 2013, [MIT chemist] Sadoway and MIT metallurgy researcher Antoine Allanore published a paper in Nature concluding that anodes made from chromium-based alloys might check all those boxes."

An anonymous reader quotes a report from Ars Technica: If you talk to experts on election security (I studied with several of them in graduate school) they'll tell you that we're nowhere close to being ready for online voting. "Mobile voting is a horrific idea," said election security expert Joe Hall when I asked him about a West Virginia experiment with blockchain-based mobile voting back in August. But on Tuesday, The New York Times published an opinion piece claiming the opposite. "Building a workable, scalable, and inclusive online voting system is now possible, thanks to blockchain technologies," writes Alex Tapscott, whom the Times describes as co-founder of the Blockchain Research Institute. Tapscott is wrong -- and dangerously so. Online voting would be a huge threat to the integrity of our elections -- and to public faith in election outcomes.

Tapscott focuses on the idea that blockchain technology would allow people to vote anonymously while still being able to verify that their vote was included in the final total. Even assuming this is mathematically possible -- and I think it probably is -- this idea ignores the many, many ways that foreign governments could compromise an online vote without breaking the core cryptographic algorithms. For example, foreign governments could hack into the computer systems that governments use to generate and distribute cryptographic credentials to voters. They could bribe election officials to supply them with copies of voters' credentials. They could hack into the PCs or smartphones voters use to cast their votes. They could send voters phishing emails to trick them into revealing their voting credentials -- or simply trick them into thinking they've cast a vote when they haven't.

Google wants NASA to help it prove quantum supremacy within a matter of months, MIT Technology Review reported Monday, citing the Space Act Agreement. From the report: Quantum supremacy is the idea, so far undemonstrated, that a sufficiently powerful quantum computer will be able to complete certain mathematical calculations that classical supercomputers cannot. Proving it would be a big deal because it could kick-start a market for devices that might one day crack previously unbreakable codes, boost AI, improve weather forecasts, or model molecular interactions and financial systems in exquisite detail. The agreement, signed in July, calls on NASA to "analyze results from quantum circuits run on Google quantum processors, and ... provide comparisons with classical simulation to both support Google in validating its hardware and establish a baseline for quantum supremacy." Google confirmed to MIT Technology Review that the agreement covered its latest 72-qubit quantum chip, called Bristlecone. Where classical computers store information in binary bits that definitely represent either 1 or 0, quantum computers use qubits that exist in an undefined state between 1 and 0. For some problems, using qubits should quickly provide solutions that could take classical computers much longer to compute.

Can researchers built a new kind of battery powerful enough to fuel an electric airplane? MIT's Technology Review profiles a company co-founded by MIT materials science professor Yet-Ming Chiang: He and his colleague, Venkat Viswanathan, are taking a different approach to reach their next goal, altering not the composition of the batteries but the alignment of the compounds within them. By applying magnetic forces to straighten the tortuous path that lithium ions navigate through the electrodes, the scientists believe, they could significantly boost the rate at which the device discharges electricity. That shot of power could open up a use that has long eluded batteries: meeting the huge demands of a passenger aircraft at liftoff. If it works as hoped, it would enable regional commuter flights that don't burn fuel or produce direct climate emissions...

The initial plan is to develop a battery that could power a 12-person plane with 400 miles (644 kilometers) of range -- enough to make trips from, say, San Francisco to Los Angeles, or New York to Washington. In a second phase, they hope to enable an electric plane capable of carrying 50 people the same distance.... Last year, the company announced plans to deliver a line of "hybrid to electric" aircraft with room for 12 passengers in 2022. At launch, the company intends to offer a hybrid plane with a gas turbine and two battery packs capable of flying around 700 miles (1,127 kilometers), as well as an all-electric version with three battery packs and a range of less than 200 miles....But crucially, the plane itself is expected to feature an open architecture that allows owners to switch out these modules over time, enabling them to upgrade to better batteries developed in the future or shift from hybrid to all-electric operation.
About 2% of the world's CO2 emissions come from air travel, and it's one of the fastest-growing sources of greenhouse-gas pollution. "More than a dozen companies, including Uber, Airbus, and Boeing, are already exploring the potential to electrify small aircraft," the article points out, "creating the equivalent of flying taxis that can cover around 100 miles (161 kilometers) on a charge. The hope is that these one- or two-passenger vehicles -- in most cases envisioned as autonomous vertical takeoff and landing aircraft -- could shorten commutes, ease congestion, and reduce vehicle emissions."

But with less ambitious batteries, "these would largely replace car rides for the rich, not displace air travel."

An anonymous reader shares a report: When you ask him about the challenges facing the blockchain system he founded five years ago, Vitalik Buterin often launches into a rapid-fire lecture full of Ethereum-specific jargon. But ask him about his own role in the technology's future, and he quickly becomes more circumspect. There's no doubt that Buterin, 24, is still the guiding light for the idealistic community that has sprung up around his creation. That's been front and center this week in Prague, where thousands of developers, investors, and entrepreneurs have gathered for Devcon, the annual "family reunion" organized by the nonprofit Ethereum Foundation. The most popular discussion topic is Buterin's vision for "Ethereum 2.0," a future iteration meant to be capable of operating efficiently at a much larger scale -- and appeal to a much broader user base -- than it can today. But in a conversation with MIT Technology Review on the sidelines of Devcon, Buterin said it's time for him to start fading into the background as "a necessary part of the growth of the community."

In what is referred to as the "Moral Machine Experiment", a survey of more than two million people from nearly every country on the planet, people preferred to save humans over animals, young over old, and more people over fewer. From a report: Since 2016, scientists have posed this scenario to folks around the world through the "Moral Machine," an online platform hosted by the Massachusetts Institute of Technology that gauges how humans respond to ethical decisions made by artificial intelligence. On Wednesday, the team behind the Moral Machine released responses from more than two million people spanning 233 countries, dependencies and territories. They found a few universal decisions -- for instance, respondents preferred to save a person over an animal, and young people over older people -- but other responses differed by regional cultures and economic status.

The study's findings offer clues on how to ethically program driverless vehicles based on regional preferences, but the study also highlights underlying diversity issues in the tech industry -- namely that it leaves out voices in the developing world. The Moral Machine uses a quiz to give participants randomly generated sets of 13 questions. Each scenario has two choices: You save the car's passengers or you save the pedestrians. However, the characteristics of the passengers and pedestrians varied randomly -- including by gender, age, social status and physical fitness. What they found: The researchers identified three relatively universal preferences. On average, people wanted: to spare human lives over animals, save more lives over fewer, prioritize young people over old ones. When respondents' preferences did differ, they were highly correlated to cultural and economic differences between countries. For instance, people who were more tolerant of illegal jaywalking tended to be from countries with weaker governance, nations who had a large cultural distance from the U.S. and places that do not value individualism as highly. These distinct cultural preferences could dictate whether a jaywalking pedestrian deserves the same protection as pedestrians crossing the road legally in the event they're hit by a self-driving car. Further reading: The study; and MIT Technology Review.