YCLOAK

Facebook said a software bug led some users to post publicly by default regardless of their previous settings. The bug affected as many as 14 million users over several days in May.

The problem, which Facebook said it has fixed, is the latest privacy scandal for the world’s largest social media company.

It said the bug automatically suggested that users make new posts public, even if they had previously restricted posts to “friends only” or another private setting. If users did not notice the new default suggestion, they unwittingly sent their post to a broader audience than they had intended.

Erin Egan, Facebook’s chief privacy officer, said the bug did not affect past posts. Facebook is notifying users who were affected and posted publicly during the time the bug was active, advising them to review their posts.

The news follows recent furor over Facebook’s sharing of user data with device makers, including China’s Huawei. The company is also still recovering from the Cambridge Analytica scandal, in which a Trump-affiliated data-mining firm got access to the personal data of as many as 87 million Facebook users.

Jonathan Mayer, a professor of computer science and public affairs at Princeton University, said on Twitter that this latest privacy gaffe “looks like a viable Federal Trade Commission/state attorney general deception case.” That’s because the company had promised that the setting users set in their most recent privacy preferences would be maintained for future posts. In this case, this did not happen for several days.

Facebook’s 2011 consent decree with the FTC calls for the company to get “express consent” from users before sharing their information beyond what they established in their privacy settings. Even if the bug was an accident on Facebook’s part, Mayer said in an email that the FTC can bring enforcement action for privacy mistakes.

Facebook, which has 2.2 billion users, says the bug was active from May 18 until May 27. While the company says it stopped the error on May 22, it was not able to change all the posts back to their original privacy parameters until later.

The mistake happened when the company built a new way for people to share “featured items” on their profiles. These items, which include posts and photo albums, are automatically public. In the process of creating this feature, Facebook said it accidentally made the suggested audience for all new posts public.

When people post to Facebook, the service suggests a default distribution for their posts based on past privacy settings. If someone made all posts “friends only” in the past, it will set their next post to “friends only” as well. People can still manually change the privacy level of the posts — anywhere from “public” to “only me” — and this was the case while the bug was active as well.
…

High costs of extracting greenhouse gases from thin air could tumble with new technologies that can help to combat climate change, scientists said on Thursday.

Carbon Engineering, a Canadian-based clean energy company, outlined the design of a large industrial plant that it said could capture carbon dioxide from the atmosphere at a cost of between $94 and $232 a ton.

That is well below past estimates of about $600 a ton by the American Physical Society, said David Keith, a Harvard University physics professor and the founder of Carbon Engineering who led the research.

“I hope to show that this as a viable energy industrial technology, not something that is a magic bullet … but something that is completely doable,” he told Reuters of the peer-reviewed study published in the journal Joule.

Carbon Engineering, which has about 40 employees and produces about a ton of carbon dioxide a day from an experimental plant. The technology makes synthetic fuels using only air, water and renewable power.

Keith said an industrial-scale plant could make fuel at a dollar a liter. That would be competitive in California, where low carbon fuel standards to cut pollution from cars and trucks mean high prices.

He said some investors were interested. An industrial plant, costing hundreds of millions of dollars, could capture a million tons of carbon dioxide a year, equivalent to emissions by 250,000 cars.

Other experts welcomed the study as a step to clear up huge uncertainties about the costs of “direct air capture.”

U.N. reports indicate that governments may have to deploy such novel technologies this century to remove carbon from nature and bury it to limit global warming under the 2015 Paris climate agreement.

“Direct air capture is a politically promising route for carbon dioxide removal,” said Oliver Geden, of the German Institute for International and Security Affairs.

And an air capture plant could be built almost anywhere. It would not threaten farmland, unlike options of planting vast forests which soak up carbon dioxide, he said.

Climeworks, a Swiss company and Carbon Engineering’s main rival, also told Reuters it was hoping to cut its production price to $100 a ton in the next 5-10 years, from about $600 now. It sells carbon dioxide to greenhouses as an airborne fertilizer to grow tomatoes or cucumbers.

Jan Wurzbacher, a founder of Climeworks, said more and more governments are likely to jack up penalties on carbon emissions to limit floods, storms and rising seas in coming years, making the technology more viable.

The World Bank says Sweden has the highest carbon taxes, currently 1,150 Swedish crowns ($133) a ton.
…

Google won’t do artificial intelligence work for weapons, the company said Thursday.  

The company will not work on “technologies that cause or are likely to cause overall harm,” wrote Sundar Pichai, Google’s CEO, in a blog post.

Google has come under fire in recent months for its contract with the U.S. Department of Defense to use AI for sifting through drone footage. AI is a field of study whereby a computer or technology is able to do things typically associated with human behavior, such as make decisions, plan and learn.

Google and other tech firms have been bringing the advances in AI to fields such as medicine, natural disaster planning, energy, transportation and manufacturing.

But these advances have also led to ethical concerns about the kinds of decisions being made without human input.

“How AI is developed and used will have a significant impact on society for many years to come,” Pichai wrote. “As a leader in AI, we feel a special responsibility to get this right.”

Sifting through drone footage

In recent months, more than 4,000 Google employees signed a petition calling for the cancellation of the company’s contract with the Department of Defense as part of the DoD’s Project Maven initiative. They joined other critics in raising alarms that the project could lead to the use of autonomous weapons.

Last week, a Google executive reportedly told employees that the company would not seek to renew its Project Maven contract with the military.  

Kirk Hanson, the executive director of the Markkula Center for Applied Ethics at Santa Clara University, which counts Google as a financial supporter, said Google’s contract highlights a larger debate about AI and military applications.

“Until we have trust that those systems will not make mistakes, we’re going to have a lot of doubts about the use of artificial intelligence for autonomous weapons,” Hanson said.

Google will continue some work for the military, Pichai said.

“We want to be clear that while we are not developing AI for use in weapons, we will continue our work with governments and the military in many other areas,” he said. “These include cybersecurity, training, military recruitment, veterans’ healthcare, and search and rescue.”
…

Google will not allow its artificial intelligence software to be used in weapons or unreasonable surveillance efforts under new standards for its business decisions in the nascent field, the Alphabet unit said Thursday.

The restriction could help Google management defuse months of protest by thousands of employees against the company’s work with the U.S. military to identify objects in drone video.

Google instead will seek government contracts in areas such as cybersecurity, military recruitment and search and rescue, Chief Executive Sundar Pichai said in a blog post Thursday.

“We want to be clear that while we are not developing AI for use in weapons, we will continue our work with governments and the military in many other areas,” he said.

Breakthroughs in the cost and performance of advanced computers have carried AI from research labs into industries such as defense and health in the last couple of years. Google and its big technology rivals have become leading sellers of AI tools, which enable computers to review large datasets to make predictions and identify patterns and anomalies faster than humans could.

But the potential of AI systems to pinpoint drone strikes better than military specialists or identify dissidents from mass collection of online communications has sparked concerns among academic ethicists and Google employees.

A Google official, requesting anonymity to discuss the sensitive issue, said the company would not have joined the drone project last year had the principles already been in place. The work comes too close to weaponry, even though the focus is on non-offensive tasks, the official said Thursday.

Google plans to honor its commitment to the project through next March, a person familiar with the matter said last week.

More than 4,600 employees petitioned Google to cancel the deal sooner, with at least 13 employees resigning in recent weeks in an expression of concern.

A nine-employee committee drafted the AI principles, according to an internal email seen by Reuters.

The Google official described the principles as a template that any software developer could put into immediate use. Though Microsoft and others released AI guidelines earlier, the AI community has followed Google’s efforts closely because of the internal pushback against the drone deal.

Google’s principles

Google’s principles say it will not pursue AI applications intended to cause physical injury, that tie into surveillance “violating internationally accepted norms of human rights,” or that present greater “material risk of harm” than countervailing benefits.

“The clear statement that they won’t facilitate violence or totalitarian surveillance is meaningful,” University of Washington technology law professor Ryan Calo tweeted Thursday.

Google also called on employees and customers developing AI “to avoid unjust impacts on people,” particularly around race, gender, sexual orientation, and political or religious belief.

The company recommended that developers avoid launching AI programs likely to cause significant damage if attacked by hackers because existing security mechanisms are unreliable.

Pichai said Google reserved the right to block applications that violated its principles. The Google official acknowledged that enforcement would be difficult because the company cannot track each use of its tools, some of which can be downloaded free of charge and used privately.

Google’s decision to restrict military work has inspired criticism from members of Congress. Representative Pete King, a New York Republican, tweeted Thursday that Google not seeking to extend the drone deal “is a defeat for U.S. national security.”
…

As the internet continues to revolutionize communications, the next world-changing technology may already be here. Blockchain, a way of recording data and automatically storing it on computers around the world, has the potential to change everything from collecting crime scene evidence to creating new digital currencies. VOA’s Jill Craig visited a blockchain hackathon in Memphis, Tennessee, to learn more.
…

Major changes could be ahead for the International Space Station but there will always be an American astronaut in orbit, NASA’s new boss said Wednesday.

The space agency is talking with private companies about potentially taking over the space lab after 2025, but no decision will made without the other 21 countries that are partners in the project, NASA Administrator James Bridenstine said in his first briefing with reporters.

President Donald Trump’s recent budget requests have put discussions about the station’s future “on steroids,” he said. Under Trump’s 2019 proposed budget, U.S. funding for the space station would end by 2025. The U.S. has spent more than $75 billion on the space station.

Options include splitting the station into different segments or reducing its size by breaking it up and discarding one part.

Always a US astronaut in orbit

But no matter what happens, there won’t be any gap when Americans aren’t in space, Bridenstine vowed. It won’t be as it was after the Apollo moon program closed or even the retirement of the space shuttle fleet, which has forced NASA to pay Russia to ferry astronauts to the station.

“There are kids graduating from high school this month, that their entire lives, we’ve had an astronaut in space,” Bridenstine said. “And we want that to live on in perpetuity forever. No gaps.”

Companies are interested in running the station and “there’s a range of options” that are just now being examined, he said.

The first station piece was launched in 1998. The complex was essentially completed with the end of the shuttle program in 2011. It is about the size of a six-bedroom house, complete with two bathrooms, a gym and a 360-degree bay window. It usually has a crew of six.

Climate change

In wide-ranging remarks, the former Oklahoma Republican congressman said he generally supports NASA’s Earth science missions, including missions that monitor heat-trapping carbon dioxide. He said at least three climate science satellites that the Trump administration had tried to cancel earlier in budget proposals “could all end up in very good shape” and that he supported them in Congress, crossing party lines.

“We’re going forth with missions that are going to do carbon monitoring,” he said, ticking off a couple of projects. “We’re committed to that.”

When told that a Pew Research poll out Wednesday said that 63 percent of Americans said NASA’s top priority should be monitoring key parts of Earth’s climate, Bridenstine said “good” and reiterated his acceptance of human-caused climate change as a threat to national security and the globe.

Back to the moon

Bridenstine also said he hopes NASA will put some kind of small robotic landers on the moon next year, followed at some later date by humans. Astronauts should use the moon as a “proving ground” for future missions to Mars, especially checking out potential health issues for living far away from Earth for a long time. He said he worried about balance, vision, bone loss and heart issues that have been reported with space station astronauts.

“We do not want to go to Mars and have our astronauts to be marshmallows on the surface of Mars,” Bridenstine said. “The moon is our best opportunity to be successful when we go to Mars.”
…

Passenger jets of the future will be safer, lighter, faster, more fuel-efficient and … windowless.

So predicts Emirates Airlines chief Tim Clark. The Dubai-based airline has already introduced virtual windows in the first-class suites of its newest planes. 

Instead of being able to see out a conventional window, the passengers will be able to enjoy the view on a full display of windows that will project live camera feeds on a high-definition screen. 

Clark said the images are “so good, it’s better than with the natural eye.”

Clark told the BBC that the ultimate goal was to have a completely windowless plane. 

“Now you have a fuselage which has no structural weaknesses because of windows. The aircraft are lighter, the aircraft could fly faster, they’ll burn less fuel and fly higher,” he said.

But Emirates’ experiment has raised concerns that might not win it the votes of safety regulators. Some passengers have expressed concerns of possibly feeling claustrophobic on windowless planes. 
…

The former head of Cambridge Analytica admitted on Wednesday his firm had received data from the researcher at the center of a scandal over Facebook users’ personal details, contradicting previous testimony to

lawmakers.

Cambridge Analytica, which was hired by Donald Trump in 2016, has denied its work on the U.S. president’s successful election campaign made use of data allegedly improperly harvested from around 87 million Facebook users.

Former chief Alexander Nix, in earlier testimony to parliament’s media committee, also denied the political consultancy had ever been given data by Aleksandr Kogan, the researcher linked to the scandal.

On Wednesday he said it had received data from Kogan.

“Of course, the answer to this question should have been ‘yes,'” Nix said, adding that he thought he was being asked if Cambridge Analytica still held data from the researcher.

He denied deliberately misleading British lawmakers and said the company had deleted the data, which had been of no use.

The committee is investigating fake news, and focusing on the role of Cambridge Analytica and Facebook in the 2016 Brexit vote as well as the Trump election.

In lengthy, and often testy, questioning by lawmakers, Nix apologized for an undercover film in which he said Cambridge Analytica’s online campaign played a decisive role in Trump’s election win.

But he defended the now-defunct consultancy’s reputation and said he felt victimized.

Cambridge Analytica said after the film was broadcast by Channel 4 television in March that the comments did not “represent the values or operations of the firm.”

Lawmakers asked Nix to return to face questions about inconsistencies in his evidence.

Kogan had told lawmakers he did give Cambridge Analytica the data.

Facebook says Kogan harvested it by creating an app on the platform that was downloaded by 270,000 people, providing access not only to their own but also their friends’ personal data.

Facebook said Kogan then violated its policies by passing the data to Cambridge Analytica.

Embarrassed but vindicated

Nix apologized for his comments in the film, saying he had been foolish and had made exaggerated claims in order to attract what he thought was a potential client.

“It’s not only deeply embarrassing, but it’s something I regret enormously,” he said.

Nix said that Channel 4 had heavily edited the footage to portray him in a worse light. “All Mr Nix’s comments carried in our reports were used in context, including any caveats,” the broadcaster said in a statement.

On other matters, Nix was less apologetic.

He said that he was vindicated in saying Cambridge Analytica had not been involved in the Brexit campaign by a report by the Electoral Commission, and that whistleblower Christopher Wylie had lied about the consultancy’s involvement in Brexit.

Wylie had told the committee that Cambridge Analytica played a pivotal role in the campaign.

On Wednesday he told Channel 4: “I actually backed up everything I said with documents. I am quite comfortable standing by the statements that I made.”

Nix denied a story in the Financial Times that he had withdrawn $8 million from Cambridge Analytica before its collapse last month.

Asked about a Guardian report that a Cambridge Analytica employee visited Wikileaks founder Julian Assange in the Ecuadorian embassy in 2017, he said he had been unaware of the meeting.
…

A surge in participation by startup companies this week, at a highlight of Asia’s biggest annual tech event, shows an increased reliance on young entrepreneurs to come with the IT industry’s strongest ideas for connected devices and artificial intelligence.

The InnoVEX segment of Taipei Computex 2018 brought together 388 startups, a term usually defined as founder-owned firms of three to five years old. That number is a jump from 272 at the same event a year ago. Venture capitalists, including at least one with half a billion dollars in investment funds, evaluated them one-on-one and at formal pitching events.

Startups are catching attention as inventors of Internet-of-things technology because there’s no market leader yet, said Jamie Lin, founding partner of AppWorks Ventures, a startup accelerator in Taipei. That technology refers to software and hardware that let computers or phones communicate with everyday devices such as cameras and alarm systems.

Some connections run on artificial intelligence, which means computerized processing of the data collected from those devices. That can mean making human-like decisions.

“Computers continue to morph and there are no dominant players in IoT,” Lin said. “That’s why they need startups and that’s what makes the show relevant.”

In software, by contrast, Google and Microsoft dominate markets worldwide. Apple and Samsung, among others, lead in smartphones.

Coinciding with the tech show this week, Lin’s accelerator, another like it and a Japanese venture capital firm are all holding their own events in Taipei this week for startups.

Expanding market

More than 20 billion things will be connected to the internet by 2020, up from 8.4 billion connected last year, market research firm Gartner forecasts. The number will pick up especially as 5G wireless services speed up connections.

By next year, Gartner anticipates, startup firms working with artificial intelligence will overtake Amazon, Google, Microsoft and IBM in “driving the artificial intelligence economy” for businesses.

Artificial intelligence, also known by its abbreviation AI, will reach a market value of $1.2 trillion per year by 2020 as investment triples between now and then, Forrester Research said.

“There’s a process, which is experimental — error and trial, error and trial – so there’s no one with a ready solution, and AI is so broad that one that can do it all,” said Tracy Tsai, a Gartner research VP in Taipei.

“With AI startups, they say ‘I’m focused, I just do some part of it and I do it well, and I do it attentively,’” she said. “For companies looking for a full solution, if you can show your part works, then they use it.”

Venture capitalists watching

Venture capital firms at the three-day InnoVEX show Wednesday watched a spread of mostly Asian startups with software and hardware ideas focused largely on connected devices. Healthcare and the management of drones were among the fields that companies said they could help with AI.

The show offered chances for startups to pitch their ideas to venture capital firms and accelerators, which are programs that show young firms how to improve their businesses.

Startup promotion authorities from 13 countries, including France and the Netherlands, also scanned the exhibition hall for Asian firms that might complement their own.

“What we care about the most is whether these startups or smaller firms have technology, so if it’s a just a business model only, they aren’t suitable for us,” said Amanda Liu, CEO of the Taiwan government-backed business accelerator StarFab. Her accelerator takes 10 to 15 of every 100 applicants. “They need to have products and their core competence must come from technology.”

Taiwanese firms are good at altering hardware specs, Liu said, and for technology ideal for businesses rather than individual consumers, Liu said. Taiwan positioned itself decades ago as a high-tech hardware manufacturing hub for much of the world.

Qara was one AI-dependent startup at InnoVEX. The 4-year-old South Korean developer with $1 million in venture capital funding uses an AI algorithm to predict the movement of stock and cryptocurrency markets. It has earned revenues of $1.5 million and reports a profit.

“Anyone can see the predictions powered by AI,” said Qara’s global CEO Katie Bomi Son. In terms of accuracy, she said, “Some are from 70, or between 70 to 90. Most of our information [comes] from the machine.”

Qara counts mostly companies as clients but it’s looking for a way to monetize the free app for common users.
…