As if tax reform wasn’t enough to worry about this week, add cybersecurity threats, biased robots and data-hungry Facebook to the mix. The government is still dealing with Kaspersky and WannaCry (thanks to North Korea); but on a positive note, our solar system is now tied for the most number of planets around a single star. Machine learning for the win!
Kaspersky Lab is Fighting Back Against U.S. Software Ban
The Moscow-based security software company is claiming the U.S. government’s ban of its products deprived if of due process, and is asking a federal court to overturn the ruling. The Department of Homeland Security ordered civilian government agencies to remove the software from their networks in September (Kaspersky products are already not allowed on military networks), after government officials worried the software was enabling Russian spies — but Kaspersky has denied its ties to any government or cyber espionage.
The lab’s founder, Eugene Kaspersky, is appealing the decision saying it harmed his company’s reputation and commercial operations without any evidence. The lawsuit also alleges that the government “largely relied on uncorroborated news media reports as evidence,” according to Reuters, asking to overturn the ban and claiming the company’s products are not a national security threat to the U.S. Reuters
Security Robot Fired After Targeting the Homeless
A 400-pound Autonomous Data Machine named “K-9” has been relieved of its crime-fighting duties in San Francisco after it was accused of discriminating against homeless people encampment on the sidewalks it patrols.
This robot security program was a pilot from the the San Francisco chapter of the animal rescue group SPCA, and the bot could be rented from Silicon Valley start-up Knightscope for $6 an hour. San Francisco SPCA launched the program to improve safety around the area for staff, by patrolling the parking lot and sidewalk outside the animal shelter after the building had already been broken into twice, and employees faced harassment and catcalls. It was able to take photos, record security footage and notify employees or police during an emergency.
But rumors started spreading that the robot was hired to target the homeless or even “chase” them away, as one Twitter user put it. After threats and vandalism against the shelter, they’ve decided to end the pilot program, as they claimed this was not their intention. The Washington Post
Facebook is Living Up to its Name
The social media giant is launching some facial recognition features that users will be given access to when they allow Facebook to use their face data (by turning the face recognition switch on). The company says these tools will help protect privacy and block catfishing attempts, but first, you’ll have to let them access, store and match your face data with uploads across the site.
One of the features will alert users when photos of them are uploaded, even if they haven’t been tagged in them and even if they aren’t friends with the person who uploaded the photo. Another will alert users if their face is included in a profile picture, and the last feature verbally tells visually impaired users who is takked in a photo alongside them.
It’s not that Facebook isn’t already using facial recognition in some areas, but these new privacy and security schemes are definitely increasing facial data intake. Gizmodo
White House Blames North Korea for WannaCry
The ransomware attack in May that affected more than 300,000 computers in 150 countries, locking users and hospital networks from information, has been officially traced back to Pyongyang. In fact, Homeland Security adviser Tom Bossert said the country was “directly responsible" for the WannaCry virus, and that there’s enough evidence to support it.
NPR described some of the clues leading up to the accusations, like the lines of code used in WannaCry are identical to what the Lazarus Group hackers used, which also link back to North Korea. Britain's Minister of State for Security, Ben Wallace, said in October that his government was "as sure as possible" that Pyongyang was behind the attack. So, Bossert said, the U.S. will continue to use its pressure strategy to stop North Korea’s ability to lay out these attacks. NPR
NASA Used AI to Find a New Planet
The agency recently discovered an eighth planet circling Kepler-90, a sun-like star thousands of lightyears from Earth. The planet, being called Kepler-90i, was found in data from NASA’s Kepler Space Telescope and with the help of machine learning from Google. The computers were taught to identify planets by finding instances where the telescope recorded signals from planets outside our solar system, and did so in archived Kepler data.
The method, artificial “neural network,” came about when researchers Christopher Shallue, a senior engineer with Google AI, and Andrew Vanderburg trained a computer to learn how to identify planets outside the solar system in the light readings recorded by Kepler. It’s done so by a small change in brightness when a planet passes in front of a star, and was inspired by the way neurons connect in the human brain. This neural network went through Kepler data and found the transmit signals from this newly discovered planet orbiting Kepler-90.
Using neural networking could mean even more discoveries for NASA, and the agency even held a Reddit “Ask Me Anything” on Dec. 14 on the news. NASA
-- Sign up for our weekly newsletter to receive the latest analysis and insights on emerging federal technologies and IT modernization.