Welcome to the January 8, 2020 edition of ACM TechNews, providing timely information for IT professionals three times a week.
ACM TechNews mobile apps are available for Android phones and tablets (click here) and for iPhones (click here) and iPads (click here).
To view "Headlines At A Glance," hit the link labeled "Click here to view this online" found at the top of the page in the html version.
The online version now has a button at the top labeled "Show Headlines."
|
|
|
The Case of the Elusive Majorana: So-Called 'Angel Particle' Still a Mystery
Penn State News
Sam Sholtis
January 3, 2020
A study by researchers at Pennsylvania (Penn) State University and Germany's University of Wurzburg casts doubt on the reported discovery of the chiral Majorana fermion in 2017. Analog versions of Majorana fermions are considered a potential pathway for building a topological quantum computer, with quantum bits shielded from environmental decoherence. The researchers analyzed dozens of devices similar to the one used to generate the so-called angel particle in the 2017 report. They found that the feature claimed to be the manifestation of the Majorana fermion was unlikely to be triggered by its existence. Said Penn State's Cui-Zu Chang, "An important first step toward this distant dream of creating a topological quantum computer is to demonstrate definitive experimental evidence for the existence of Majorana fermions in condensed matter. Over the past seven or so years, several experiments have claimed to show such evidence, but the interpretation of these experiments is still debated."
|
White House Proposes Guidelines for Regulating Use of AI
Associated Press
Matt O'Brien
January 7, 2020
The Trump administration is proposing new federal rules for regulating the use of artificial intelligence (AI) in the private sector, but those regulations do not address concerns about a dearth of accountability as computers increasingly assume human responsibilities in high-risk settings. A White House document said agencies must weigh fairness, lack of bias, openness, transparency, safety, and security in deciding regulatory action. However, the rules will not apply to how federal agencies like law enforcement use AI. U.S. chief technology officer Michael Kratsios said he hopes these guidelines can be a model for other Western democratic institutions to protect shared values without excessive regulation.
|
Robotic Device Helps Spine Injury Patients Sit More Comfortably
Business Standard
January 6, 2020
Columbia University researchers have built a robotic device to assist and train people with spinal cord injuries (SCIs) to sit with greater stability, which could help with recovery. The Trunk-Support Trainer (TruST) is a motorized cable-driven belt placed on the user's torso to assess postural and workspace-area control limits while sitting; the tool applies force to the torso when the user moves their upper body beyond postural stability limits. Columbia's Sunil Agrawal said TruST prevents patients from falling, while maximizing their trunk movements beyond their balance limits. Columbia's Victor Santamaria added that TruST's ability to deliver continuous force-feedback tailored for users' individual postural limits creates opportunities to deploy motor learning-based paradigms to retrain functional sitting in SCI patients.
|
Delta Wants to Use Tech to Improve Flying Experience
Reuters
David Shepardson
January 7, 2020
Delta Air Lines CEO Ed Bastian announced plans at CES 2020 to improve the flying experience with technology upgrades resulting from investments the airline has made over the last five years. As part of the effort, Delta said it will introduce virtual queuing via an app to notify passengers when their seat is boarding. The airline said it is using artificial intelligence and machine learning to try to shorten delays, by analyzing millions of operational data points to generate hypothetical outcomes. Bastian said the airline's goal is to get “Delta people from out behind the counter so they can assist you, solve your problems in real time.”
|
Vaping Lung Injury Symptoms Reported Online for at Least Seven Years
UC Riverside News (CA)
Iqbal Pittalwala
January 6, 2020
A study by University of California, Riverside (UCR) researchers sifted through an online discussion forum for electronic-cigarette users, and determined this group reported vaping-associated pulmonary illness symptoms for at least seven years. The researchers compiled data posted on the forum from January 2008 to July 2015, using a Java-based Web crawler to mine information on symptoms and disorders from some 41,000 posts. Sentiment was found to be negative in 45% of the posts, neutral in 38%, and positive in 17%. Said UCR's Prue Talbot, "Our data, which shows many of the symptoms characterizing the current patients have been reported online for at least seven years, suggests cases similar to those in the current VAPI epidemic have existed previously and been unreported or simply not linked to vaping."
|
Office ID Badge About to Become Unrecognizable
The Wall Street Journal
Catherine Stupp
January 6, 2020
Biometric identification systems, microchip implants, and devices that monitor employee behavior are beginning to replace traditional ID badges, with the potential to improve workplace security and navigability. Innovations like face and iris scans are harder to spoof than plastic ID cards, and companies can easily upgrade office cameras with facial-recognition technology. Facial recognition is especially popular at businesses with large workforces, but problems identifying people of color have prompted legislation to limit its usage. The U.S. Intelligence Advanced Research Projects Activity's Lars Ericson suggested smartphones could easily replace conventional IDs, with employees downloading apps to signal their arrival to office sensors, while phone-based motion sensors can record their daily movements and confirm their identity.
*May Require Paid Registration
|
PGP Keys, Software Security, Much More Threatened by New SHA1 Exploit
Ars Technica
Dan Goodin
January 7, 2020
Researchers at Inria France and Singapore's Nanyang Technological University have unveiled an exploit that is more potent than the previous SHA1 cryptographic hash algorithm attack. The new collision exploit makes it practical to create PGP encryption keys that can impersonate a target when digitally signed using the SHA1 algorithm; it also generates the same hash for two or more attacker-selected inputs by appending data to each. SHA1 remains the default hash function for certifying PGP keys in the legacy 1.4 version branch of GnuPG, the heir to PGP application for email and file encryption. Johns Hopkins University's Matt Green said these findings highlight the fact that SHA1 can no longer be considered secure, adding, “We knew that one shoe had dropped, and this is the next shoe dropping.”
|
Facebook Bans 'Deepfake' Videos
The Guardian
Alex Hern
January 7, 2020
Facebook has banned doctored videos, or deepfakes, designed to spread misinformation as the 2020 U.S. presidential election approaches. The prohibition only applies to artificial intelligence (AI)-produced content, and does not ban misleading videos made with conventional editing tools. Facebook’s Monika Bickert said the company, along with Instagram (a social networking service that it owns), will eliminate deepfakes that have been edited or synthesized in ways that are not apparent to the average person, and that would likely trick viewers into attributing words to the subject of the video they never actually spoke. Another requirement for removal is for the material to be produced by AI or machine learning that combines, substitutes, or superimposes content onto a video, making it seem authentic. Said Bickert, "Videos that don't meet these standards for removal are still eligible for review by one of our independent third-party fact-checkers, which include over 50 partners worldwide fact-checking in over 40 languages."
|
A Lazy Fix 20 Years Ago Means the Y2K Bug Is Taking Down Computers Now
New Scientist
Chris Stokel-Walker
January 7, 2020
A glitch connected to the Y2K bug addressed two decades ago is currently crashing payment systems and computers, due to a poor fix. Coders had to fix the Y2K bug either by completely rewriting their code or using a windowing technique, which would treat all dates from 00 to 20 as from the 2000s, rather than the 1900s. Programmers opted for 1920 to 2020 as their standard window because many coding languages and systems manage dates and times as seconds from 1970/01/01, with the expectation that the windowed systems would be obsolete by 2020. The windowed systems have now rolled back to 1920, giving rise to the Y2020 bug. Problems attributed to Y2020 include cash registers from manufacturer Novitus that cannot print receipts, and a professional wrestling video game that stopped working on Jan. 1, 2020.
|
Dating Apps Need Women, Advertisers Need Diversity. AI Companies Offer a Solution: Fake People
The Washington Post
Drew Harwell
January 7, 2020
Artificial intelligence (AI) startups are marketing images of computer-generated faces, giving companies opportunities to create imaginary models and boost diversity in ads without the use of actual people. The software for generating images is freely available, and trained on vast datasets of real faces to reproduce their features in new configurations. AI experts are concerned this could further fraud and manipulation by allowing the construction of fictitious online personas, concealing discrimination in hiring, undermining industry diversification, and escalating distrust of the Internet. The University of California, Los Angeles' Elana Zeide said the technology “showcases how little power and knowledge users have in terms of the reality of what they see online.” Meanwhile, Valerie Emanuel of the Role Models Management talent agency said the fake photos could give rise to a monoculture of similar-looking faces.
*May Require Paid Registration
|
Fake Clicks on Online Ads Costing Companies 'Tens of Billions' a Year
Financial Times
Tim Bradshaw
December 30, 2019
Researchers estimate scammers are stealing tens of billions of dollars from companies every year by faking clicks on online ads. Ad fraud investigator Augustine Fou thinks the scale of the problem is "many times" greater than the official estimate of about $6 billion annually, compared to more than $300 billion spent on digital ads this year. Methods for perpetrating such crimes include programming smartphones to click on ads from complicit websites or ad networks, or using software to generate virtual phones and spoof downloads. Gary Danks with ad fraud tracking company Machine said 49% of over 150 million ad-driven app downloads in 2019 were bogus, while 90% of the 200 networks that sell "app-install ads" were "knowingly selling fraudulent inventory."
*May Require Paid Registration
|
|
