Home / Technology / Tech-news /  Cutting-edge tech to combat extremists, hate speech, fake drugs et al

From developing insect-like robots to building sensors that can detect bad medicine, researchers across the world are experimenting with cutting-edge technologies in their labs. Here are some of them:

Robotic Skins

Researchers at Yale School Of Engineering and Applied Sciences are developing “skins" that allow users to turn soft inanimate objects into robots. These skins, developed in assistant professor Rebecca Kramer-Bottiglio’s lab, are made from elastic sheets embedded with sensors and actuators. Draped around a deformable object—a stuffed animal or a foam tube, for instance—the skins can animate them from their surfaces. The makeshift robots can perform different tasks depending on the properties of the soft objects and how the skins are applied.

Flying Insect bots

Researchers from the Micro Air Vehicle Laboratory of the Netherlands-based Delft University of Technology have developed an insect-inspired flying robot, according to a 13 September press release. While MAVLab has been developing insect-inspired flying robots for over 10 years under the DelFly project, the robot in this study—named the DelFly Nimble—uses off-the-shelf components, and has flight endurance long enough to be of interest for real-world applications. The robot has a top speed of 25 kilometre per hour and can even perform aggressive manoeuvres, such as 360-degree flips, loops and barrel rolls.

Bad drugs, bad music

A new sensor based on a 3,000-year-old African musical instrument can be used to identify substances, including a poisonous chemical sometimes mistakenly added to medicines. The device, built by an engineer at the University of California, Riverside (UCR), and described in a paper published in ACS Omega, can accurately measure the density of any liquid, according to a 12 September press release. William Grover, an assistant professor of bioengineering in UCR’s Marlan and Rosemary Bourns College of Engineering, and his team modified an mbira—an instrument made of metal prongs attached to a soundboard—for this purpose. The sensor, the researchers believe, could help drug companies or compounding pharmacists in the developing world verify bottles of glycerol that might have been mislabelled, unintentionally or not, by the manufacturer or distributor. It can also help consumers detect counterfeit drugs by comparing the density of a suspect sample to a known good sample of the same drug.

Identify extremists

New research has found a way to identify extremists, such as those associated with the terror group like the Islamic State (IS), by monitoring their social media accounts. Such individuals can be identified even before they post threatening content. The study, “Finding Extremists In Online Social Networks", was recently published in Operations Research journal . The researchers, according to a 12 September press release, collected Twitter data from approximately 5,000 “seed" users who were either known IS members or who were connected to many known IS members as friends or followers. Using statistical modelling of extremist behaviour with optimized search policies and actual IS user data, the researchers developed a method to predict new extremist users, identify if more than one account belongs to the same user, as well as predict network connections of suspended extremist users who start a new account.

AI fooled

A new study by the Aalto University’s Secure Systems research group (ssg.aalto.fi) has discovered weaknesses in many machine learning (ML) detectors that are currently used to recognize and keep hate speech at bay, according to a 14 September press release. Bad grammar and awkward spelling—intentional or not—might make toxic social media comments harder for AI detectors to spot, a team of researchers led by professor N. Asokan has shown. The researchers inserted typos, changed word boundaries and even added neutral words to the original hate speech.

AI to alter videos

Researchers at Carnegie Mellon University (CMU) have devised a way to use artificial intelligence (AI) to automatically transform the content of one video into the style of another, making it possible to, say, transfer the facial expressions of comedian John Oliver to a cartoon character, or to make a daffodil bloom in much the same way a hibiscus would, according to a 11 September press release.

The trick is performed by a class of algorithms called generative adversarial networks (GANs). The researchers developed a technique, called Recycle-GAN, which incorporates not only spatial but temporal information.

They showed that Recycle-GAN can be used to transform the video of John Oliver into what appears to be fellow comedian Stephen Colbert and back into Oliver. Such effects might also be useful in developing self-driving cars that can navigate at night or in bad weather, according to the researchers.

According to Aayush Bansal, a Ph.D student in CMU’s Robotics Institute, the technology also has potential to be used for detecting so-called “deep fakes"—videos in which a person’s image is inserted without permission, making it appear that the person has done or said things that are out of character.

Catch all the Technology News and Updates on Live Mint. Download The Mint News App to get Daily Market Updates & Live Business News.
More Less

Recommended For You

Trending Stocks

Get alerts on WhatsApp
Set Preferences My ReadsWatchlistFeedbackRedeem a Gift CardLogout