Slashdot reader Rei writes: NASA -- having already populated the Red Planet with robots and armed a car-sized nuclear juggernaut with a laser -- have now decided to grant fire control of that laser over to a new AI system operating on the rover itself. Intended to increase the scientific data-gathering throughput on the sometimes glitching rover's journey, the improved AEGIS system eliminates the need for a series of back-and-forth communication sessions to select targets and aim the laser.
Rei's original submission included a longer riff on The War of the Worlds, ending with a reminder to any future AI overlords that "I have a medical condition that renders me unfit to toil in any hypothetical subterranean lithium mines..."

An anonymous reader shares a Reuters report: China is aiming for a top-10 ranking in automation for its industries by 2020 by putting more robots in its factories, the International Federation of Robotics (IFR) said. China's push to modernize its manufacturing with robotics is partly a response to labor shortages and fast-rising wages. But the world's second-largest economy still has far lower robot penetration than other big industrialized economies -- just 36 per 10,000 manufacturing workers in 2015, ranking it 28th among the world's most automated nations. By 2020, it aims to boost penetration to 150 per 10,000 workers, IFR said in a statement, citing Wang Ruixiang, President of the China Machinery Industry Federation. To help reach that goal, China aims for sales of 100,000 domestically produced industrial robots a year by 2020, up 49 percent compared with last year, the IFR said in a statement at an industry summit in Shanghai, where the Chinese federation's chief was speaking.

An anonymous reader writes from a report via Engadget: The Washington Post and Buzzfeed have sent robots to cover the Republican National Convention in Cleveland, Ohio. The Washington Post is using a telepresence robot from Double Robotics that consists of an iPad mounted on a Segway-like base. It's objective: to roam around the convention, streaming live on Periscope. Those viewing the stream will be able to ask questions of delegates, politicians and other figures who stumble upon the robot. BuzzFeed is using a robot called 'BuzzBot.' It's a Facebook chat bot that collects and caters news from the convention to users' messaging feeds. All you have to do is add the channel to your Messenger app and it will deliver news updates from BuzzFeed reporters. Specifically, it will collect reports from delegates, protesters and others in Cleveland. You have the option to send pictures and other info to BuzzBot, but it may ask you questions about your experience. The questions it asks will be different depending on your location. For example, if you live in Cleveland it will want to know what kind of impact the RNC is having on your daily life. Meanwhile, with roughly 50,000 attendees and likely millions of viewers watching across the country and abroad, the RNC is preparing for cyberattacks that aim to disrupt the network.

Slashdot reader Kassandra Perlongo shares an article about "soft robotics and exoskeletons" at the University of Salford, writing "those bulky, mechanical suits could eventually be a thing of the past." Children with a rare neurological disease were recently given the chance to walk for the first time thanks to a new robotic exoskeleton... But while today's exoskeletons are mostly clumsy, heavy devices, new technology could make them much easier and more natural to use by creating a robotic skin... [Soft robotic devices] are particularly well suited to interaction with humans as they are typically lightweight which means if they collide with a person they are unlikely to cause injury.

We recently developed a new "soft continuum actuator", a joint that bends like an elephant's trunk...if it encounters resistance in one part of its body it will still bend but at a different location elsewhere along its length. By equipping a skintight material suit with these actuators, we can create a soft exoskeleton that bends at the precise location of the wearer's joints. This means the suit will fit a range of users comfortably without needing mechanical adjustment or calibration. Plus, the system is lightweight and can be worn like clothing rather than a bulky mechanical frame.

MarkWhittington writes: Eric Berger has published an account in Ars Technica about how President Jimmy Carter saved the space shuttle program. The article is well worth reading for its detail. In essence, around 1978 the space shuttle program had undergone a crisis with technical challenges surrounding its heat-resistant tiles and its reusable rocket engines and cost overruns. President Carter was not all that enthused about human space flight to begin with, adhering to the since discredited notion that robotic space probes were adequate for exploring the universe. His vice president, Walter Mondale, was a vehement foe of human space flight programs, maintaining that money spent on them were better used for social programs.

An anonymous reader writes from a report via KGO-TV: PSA: Beware of dangerous security robots at the Stanford Shopping Center! After a young boy was "knocked down and run over" by one of the Stanford Shopping Center security robots, the boy's parents want to help prevent others from getting hurt. KGO-TV reports: "They said the machine is dangerous and fear another child will get hurt. Stanford Shopping Center's security robot stands 5' tall and weighs 300 pounds. It amuses shoppers of all ages, but last Thursday, 16-month-old Harwin Cheng had a frightening collision with the robot. 'The robot hit my son's head and he fell down facing down on the floor and the robot did not stop and it kept moving forward,' Harwin's mom Tiffany Teng said. Harwin's parents say the robot ran over his right foot, causing it to swell, but luckily the child didn't suffer any broken bones. Harwin also got a scrape on his leg from the incident." Teng said, "He was crying like crazy and he never cries. He seldom cries." They are concerned as to why the robot didn't detect Harwin. "Garage doors nowadays, we're just in a day in age where everything has some sort of a sensor," shopper Ashle Gerrard said. "Maybe they have to work out the sensors more. Maybe it stopped detecting or it could be buggy or something," shopper Ankur Sharma said. The parents said a security guard told them another child was hurt from the same robot just days before. They're hoping their story will help other parents be more careful the next time they're at the Stanford Shopping Center. The robots are designed by Knightscope and come equipped with self-navigation, infra-red cameras and microphones that can detect breaking glass to support security services.

An anonymous Slashdot reader writes: Where Twinkie once employed 22,000 workers in more than 40 bakeries, their workforce is now down to just 1,170, reports the Washington Post, relying mostly on robotic arms and other forms of automation. "This 500-person plant produces more than 1 million Twinkies a day, 400 million a year. That's 80% of Hostess' total output -- output that under the old regime required 14 plants and 9,000 employees."

"We like to think of ourselves as a billion-dollar startup," Hostess chief executive Bill Toler said Tuesday, announcing that Hostess Brands, which had twice filed for bankruptcy, now plans to become a publicly-listed company valued at $2.3 billion.

A police standoff with a suspect in the killing of five police officers in Dallas came to an abrupt end on Friday morning in an unusual way. The police said that negotiations broke down, an exchange of gunfire happened, but then they had no option but to use "bomb robot and place a device on its extension for it to detonate where the suspect was." Motherboard explains the unprecedented shift in policing. From an article: Peter W. Singer, an expert in military technology and robot warfare at the New America Foundation, tweeted that this is the first known incident of a domestic police force using a robot to kill a suspect. Singer tweeted that in the wars in Iraq and Afghanistan, soldiers have strapped claymore mines to the $8,000 MARCbot using duct tape to turn them into jury-rigged killing devices. Singer says all indications are that the Dallas Police Department did something similar in this case -- it improvised to turn a surveillance robot into a killing machine. Improvised device or not, the concerns here mirror a debate that's been going on for a few years now: Should law enforcement have access to armed drones, or, for that matter, weaponized robots? In 2013 Kentucky Senator Rand Paul staged a 13-hour filibuster that was focused entirely on concerns about the use of armed drones on US soil. Last year, North Dakota became the first state to legalize nonlethal, weaponized drones for its police officers. [...] The ability for police to remotely kill suspects raises due process concerns. If a shooter is holed up and alone, can they be qualified as an imminent threat to life? Are there clear protocols about when a robot can be used to engage a suspect versus when a human needs to engage him or her? When can the use of lethal force be administered remotely?

An anonymous reader writes: Harvard researchers report in the journal Science this week that they've built a "bio-inspired swimming robot that mimics a ray fish [and] can be guided by light." The robot's body consists of "a cast elastomer body with a skeleton of gold, along with a single layer of carefully aligned muscle fibers harvested from neonatal rat hearts." The fibers were genetically modified to respond to pulses of blue light and structured along the body of the robot such that contractions result in a repetitive undulating motion, propelling the robot forward. "About 200,000 live rat heart cells form the muscle layer that powers the robot, which has a body 16.3 mm long and weighs just over 10 grams," reports IEEE. "At full tilt, it can swim at a speed of 3.2 mm/s, which isn't bad for such a tiny thing." You can watch the video that shows the robot being led through a 250 mm long obstacle course.

Fusion's Kashmir Hill is reporting about a five-foot-tall, white, egg-shaped robot that one can find at the company's inspection lot near Mission Bay in San Francisco. The K5 robot is a stand-in for a human security guard, and it sports multiple high-definition cameras for 360-degree vision, a thermal camera, a laser rangefinder, a weather sensor, a license-plate recognition camera, four microphones, and person recognition capabilities. The report adds:If someone suspicious comes into the lot, or starts messing with a car, the robot can't tase them or break out any weapons. Instead the robot can set off an alarm, send a signal to human security personnel, and record everything that person does to be used against them later by police. Customers of Knightscope, the company that manufactures the aforementioned robot don't buy the machines. They rent them, usually two at a time, so one can charge its battery while the other patrols. The cost is $7 an hour. "For the cost of a single-shift security guard, you get a machine that will patrol for 24 hours a day 7 days a week," said Stephens, citing wages of $25 to $35 hour for a human security guard.

Starship Technologies has begun testing their on-demand delivery robots in cities around the world -- including Washington, D.C. -- to manage the "last mile" for small deliveries. Slashdot reader Okian Warrior quotes the Starship Technologies site: Capable of carrying the equivalent of two grocery bags, the robots can complete local deliveries within 5-30 minutes from a local hub or retail outlet, for 10-15 times less than the cost of current last-mile delivery alternatives. Customers can choose from a selection of short, precise delivery slots -- meaning goods arrive at a time that suits them. During delivery, shoppers can track the robot's location in real time through a mobile app, and on arrival only the app holder is able to unlock the cargo.
Created by two Skype co-founders, the company uses ground-based delivery drones equipped with nine cameras, two-way audio capability, and GPS, according to ABC News, which has video of the robots in action. "When confronted with any kind of issue or trouble, a human at Starship can take over. The remote operator can have a two-way conversation with those around the robot... They hope to make the robots available for 24/7 delivery and for only a $1 fee." What could go wrong?

In an op-ed for Slate, Microsoft CEO Satya Nadella has shared his views on AI, and how humans could work together with this nascent technology to do great things. Nadella feels that humans and machines can work together to address society's greatest challenges, including diseases and poverty. But he admits that this will require "a bold and ambition approach that goes beyond anything that can be achieved through incremental improvements to current technology," he wrote. You can read the long essay here. GeekWire has summarized the principles and goals postulated by Nadella. From the article:AI must be designed to assist humanity.
AI must be transparent.
AI must maximize efficiencies without destroying the dignity of people.
AI must be designed for intelligent privacy.
AI needs algorithmic accountability so humans can undo unintended harm.
AI must guard against bias.
It's critical for humans to have empathy.
It's critical for humans to have education.
The need for human creativity won't change.
A human has to be ultimately accountable for the outcome of a computer-generated diagnosis or decision.

Former Google CEO, and current Alphabet Executive Chairman Eric Schmidt and Google X founder Sebastian Thrun in an op-ed on Fortune Magazine have shared their views on artificial intelligence, and what the future holds for this nascent technology. "When we first worked on the AI behind self-driving cars, most experts were convinced they would never be safe enough for public roads. But the Google Self-Driving Car team had a crucial insight that differentiates AI from the way people learn. When driving, people mostly learn from their own mistakes. But they rarely learn from the mistakes of others. People collectively make the same mistakes over and over again," they wrote. The two also talked about an artificial intelligence apocalypse, adding that while it's unlikely to happen, the situation is still worth considering. They wrote:Do we worry about the doomsday scenarios? We believe it's worth thoughtful consideration. Today's AI only thrives in narrow, repetitive tasks where it is trained on many examples. But no researchers or technologists want to be part of some Hollywood science-fiction dystopia. The right course is not to panic - it's to get to work. Google, alongside many other companies, is doing rigorous research on AI safety, such as how to ensure people can interrupt an AI system whenever needed, and how to make such systems robust to cyberattacks.It's a long commentary, but worth a read.

An anonymous reader writes from a report via The Verge: Ever since Edward Snowden set in motion the most powerful public act of whistleblowing in U.S. history, he has been living in exile in Russia from the United States. An article in this week's New York Magazine looks at how Snowden may have a narrow window of opportunity where President Obama could pardon him before he leaves office. Presumably, once he leaves office, the chances of Snowden being pardoned by Hillary Clinton or Donald Trump are miniscule. Obama has said nothing in the past few years to suggest he's interested in pardoning Snowden. Not only would it contradict his national security policy, but it will severely alienate the intelligence community for many years to come. With that said, anyone who values a free and secure internet believes pardoning Snowden would be the right thing to do. The Verge reports: "[Snowden] faces charges under the Espionage Act, which makes no distinction between delivering classified files to journalists and delivering the same files to a foreign power. For the first 80 years of its life, it was used almost entirely to prosecute spies. The president has prosecuted more whistleblowers under the Espionage Act than all president before him combined. His Justice Department has vastly expanded the scope of the law, turning it from a weapon against the nation's enemies to one that's pointed against its own citizens. The result will be less scrutiny of the nation's most powerful agencies, and fewer forces to keep them in check. With Snowden's push for clemency, the president has a chance to complicate that legacy and begin to undo it. It's the last chance we'll have."

Slashdot reader Taco Cowboy brings a new report about Russian robot IR77, which has escaped from its research lab again... The story goes that an engineer working at Promobot Laboratories, in the Russian city of Perm, had left a gate open. Out trundled Promobot, traveling some 150 feet into the city before running out of juice. There it sat, batteries mostly dead, in the middle of a Perm street for 40 minutes, slowing cars to a halt and puzzling traffic cops

A researcher at Promobot's facility in Russia said that the runaway robot was designed to interact with human beings, learn from experiences, and remember places and the faces of everyone it meets. Other versions of the Promobot have been docile, but this one just can't seem to fall in line, even after the researchers reprogrammed it twice. Despite several rewrites of Promobot's artificial intelligence, the robot continued to move toward exits. "We have changed the AI system twice," Kivokurtsev said. "So now I think we might have to dismantle it".
Fans of the robot are pushing for a reprieve, according to an article titled 'Don't kill it!': Runaway robot IR77 could be de-activated because of 'love for freedom'

kheldan writes: Do you want robots making your pizza? Alex Garden, co-founder and executive chairman of Mountain View startup Zume, is betting you will. Garden, the former president of Zynga Studios, was previously a general manager of Microsoft's Xbox Live. Garden launched Zume in stealth mode last June, when he began quietly recruiting engineers under a pseudonym and building his patented trucks in an unmarked Mountain View garage. In September, he brought on Julia Collins, a 37-year-old restaurant veteran. She became chief executive officer and a co-founder. Collins was previously the vice president and CEO of Harlem Jazz Enterprises, the holding company for Minton's, a historic Harlem eatery. The company consists of an army of robot sauce-spreaders and trucks packed full of ovens. "In the back of Mountain View's newest pizzeria, Marta works tirelessly, spreading marinara sauce on uncooked pies. She doesn't complain, takes no breaks, and has never needed a sick day. She works for free." The pie then "travels on a conveyer belt to human employees who add cheese and toppings." From there, "The decorated pies are then scooped off the belt by a 5-foot tall grey automation, Bruno, who places each in a 850-degree oven. For now, the pizzas are fully cooked and delivered to customers in branded Fiats painted with slogans, including: 'You want a piece of this?' and 'Not part of the sharing economy.'" Garden says, "We are going to be the Amazon of food. [...] Just imagine Domino's without the labor component. You can start to see how incredibly profitable that can be."

An anonymous reader quotes a report from GeekWire: Researchers have taught a computer to do a better-than-expected job of predicting what characters on TV shows will do, just by forcing the machine to study 600 hours' worth of YouTube videos. The researchers developed predictive-vision software that uses machine learning to anticipate what actions should follow a given set of video frames. They grabbed thousands of videos showing humans greeting each other, and fed those videos into the algorithm. To test how much the machine was learning about human behavior, the researchers presented the computer with single frames that showed meet-ups between characters on TV sitcoms it had never seen, including "The Big Bang Theory," "Desperate Housewives" and "The Office." Then they asked whether the characters would be hugging, kissing, shaking hands or exchanging high-fives one second afterward. The computer's success rate was 43 percent. That doesn't match a human's predictive ability (72 percent), but it's way better than random (25 percent) as well as the researchers' benchmark predictive-vision programs (30 to 36 percent). The point of the research is to create robots that do a better job of anticipating what humans will do. MIT's Carl Vondrick and his colleagues are due to present the results of their experiment next week at the International Conference on Computer Vision and Pattern Recognition in Las Vegas. "[The research] could help a robot move more fluidly through your living space," Vondrick told The Associated Press. "The robot won't want to start pouring milk if it thinks you're about to pull the glass away." You can watch their YouTube video to learn more about the experiment.

An anonymous reader writes: Boston Dynamics has shown the world their "fun-sizeified version of their Spot quadruped," the SpotMini robot. It's a quiet, all electric machine that features a googley-eyed face-arm. IEEE Spectrum notes some observations made from watching their YouTube video. First of all, the SpotMini appears to be waterproof and doesn't rely on hydraulics like the other more powerful robots of theirs. The SpotMini is likely operated by a human, and is not autonomous, though the self-righting could be an autonomous behavior. The video appears to show two separate versions of the SpotMini: an undressed and dressed variant (it's hard to tell if the "dressed" variant features differing components/abilities). There is a MultiSense S7 video camera on the front, some other camera-based vision system on the front, a butt-mounted Velodyne VLP-16 system, and what may be a small camera on the face-arm's mouth. One particularly noteworthy observation is that during much of the video, the SpotMini is traversing through a house. In other Boston Dynamics demo videos, the robots are outside. The author of the report says, "[...] it wouldn't surprise me if we're looking at an attempt to make an (relatively) affordable robot that can do practical things for people who aren't in the military."

An anonymous reader writes: Online shopping aside, people don't have as many physical items to mail as they used to, which is largely the reason why Canada Post announced it would be phasing out door-to-door mail delivery. Motherboard reports: "The corporation is exploring future use of drone technology to make deliveries, according to a report from the Canadian Press. At this point, Canada Post is engaging in a 'proper exercise,' a spokesperson told the Canadian Press, adding that the project is in its earliest, experimental stages. According to Graham Scott, the deputy editor of Canadian Business, even if mail-delivering drones remain a theoretical concept for now, it's inevitable they'll be considered as a way to drive costs down. There are many good reasons why mail delivery drones may never get off the ground. For one thing, current technology limits them to delivering one item of post at a time, which is tremendously impractical. But, as we've seen with the rolling out of community mailboxes -- a program that was put on hold earlier this year when the review was launched -- the invisible hand of the market is always looking to drive costs down. So don't count out flying robot deliveries for good. From a manager's perspective at least, drones have their advantages. They don't suffer from dog bites, and they (ideally) don't deviate from their routes. 'Drones don't twist their ankle, they don't get tired, and they don't form a union.' said Scott." In 2013, Amazon CEO Jeff Bezos revealed during a CBS 60 Minutes interview that the company is working on a service called "Prime Air" to deliver packages by autonomous octocopter drones within 30 minutes of hitting the "buy" button. The Guardian reported last year that Amazon has been testing its drone delivery service at a secret site in Canada, following repeated warnings by the e-commerce giant that it would go outside the U.S. to bypass what it sees as the U.S. federal government's lethargic approach to the new technology.

An anonymous reader writes from a report via Yahoo News: Under the European Union's new draft plan, Europe's growing army of robot workers could be classed as "electronic persons," with their owners liable to paying social security for them. Robots are only becoming more prevalent in the workplace. They're already taking on tasks such as personal care or surgery, and their population is only expected to rise as their abilities are expanded with the increased development of new technologies. A draft European Parliament motion suggests that their growing intelligence, pervasiveness and autonomy requires rethinking everything from taxation to legal liability. The draft motion called on the European Commission to consider "that at least the most sophisticated autonomous robots could be established as having the status of electronic persons with specific rights and obligations." It also suggested the creation of a register for smart autonomous robots, which would link each one to funds established to cover its legal liabilities. Patrick Schwarzkopf, managing director of the VDMA's robotic and automation department, said: "That we would create a legal framework with electronic persons -- that's something that could happen in 50 years but not in 10 years. We think it would be very bureaucratic and would stunt the development of robotics," he told reporters. The report added that the robotics and artificial intelligence may result in a large part of the work now done by humans being taken over by robots, raising concerns about the future of employment and the viability of social security systems. The draft motion also said organizations should have to declare savings they made in social security contributions by using robotics instead of people, for tax purposes.