Group items tagged

Episode 55: This week we discuss why some people believe that their devices are always watching or listening to them. Oh, and whether or not it's true. Companies are finding more creative ways to target ads and content and consumers are consenting every day. But what are they consenting to?

A security researcher accessed personal information stored by Hello Barbie, a Wi-Fi-enabled talking Barbie doll.

Virgin sends subscribers a letter with details of what they watched

Virgin media sends Matt Webb a summary of what he's watched. Creepy!

New York City amasses data on habits, health and security of its citizens to cope with spiralling growth

Critical and insightful reading list about IOT

"A vulnerability in FitBit fitness trackers first reported to the vendor in March could still be exploited by the person you sit next to on a park bench while catching your breath."

"A Ford safety feature has also turned out to be a way to track badly-behaved drivers."

Uber, realizing it had hundreds of millions of people in its back seats as a captive audience, just announced it will allow third-party apps the ability to serve up notifications, content, or ads to its users during their trips. The notifications would be customized to the length and locations of each person's ride, but only with permission from the user.

"Amazon, Apple and Google all employ staff who listen to customer voice recordings from their smart speakers and voice assistant apps."

The makers of the We-Vibe, a line of vibrators that can be paired with an app for remote-controlled use, have reached a $3.75 million class action settlement with users following allegations that the company was collecting data on when and how the sex toy was used. Standard Innovations, the Canadian manufacturer of the We-Vibe, does not admit any wrongdoing in the settlement finalized Monday.

The blind application of machine learning runs the risk of amplifying biases present in data. Such a danger is facing us with word embedding, a popular framework to represent text data as vectors which has been used in many machine learning and natural language processing tasks. We show that even word embeddings trained on Google News articles exhibit female/male gender stereotypes to a disturbing extent. This raises concerns because their widespread use, as we describe, often tends to amplify these biases. Geometrically, gender bias is first shown to be captured by a direction in the word embedding. Second, gender neutral words are shown to be linearly separable from gender definition words in the word embedding. Using these properties, we provide a methodology for modifying an embedding to remove gender stereotypes, such as the association between between the words receptionist and female, while maintaining desired associations such as between the words queen and female. We define metrics to quantify both direct and indirect gender biases in embeddings, and develop algorithms to "debias" the embedding. Using crowd-worker evaluation as well as standard benchmarks, we empirically demonstrate that our algorithms significantly reduce gender bias in embeddings while preserving the its useful properties such as the ability to cluster related concepts and to solve analogy tasks. The resulting embeddings can be used in applications without amplifying gender bias.

Say you're scrolling through your Facebook Newsfeed and you encounter an ad so eerily well-suited, it seems someone has possibly read your brain. Maybe your mother's birthday is coming up, and Facebook's showing ads for her local florist. Or maybe you just made a joke aloud about wanting a Jeep, and Instagram's promoting Chrysler dealerships. Whatever the subject, you've seen ads like this. You've wondered - maybe worried - how they found their way to you.

"A trove of internal documents sheds light on the algorithms that Facebook's censors use to differentiate between hate speech and legitimate political expression."

Session is a secure messaging app that protects your metadata, encrypts your communications and makes sure your messaging activities leave no digital trail behind.

Our system collects torrent files in two ways: parsing torrent sites and listening DHT network. We have more than 500.000 torrents which where classified and which are using now for collecting peer sharing facts (up to 700.000.000 daily). We don't guarantee we can show ALL peer sharing facts:

"Only a couple of weeks ago, there were a lot of news headlines about how Germany had banned an internet-connected doll called "Cayla" over fears hackers could target children. One of their primary concerns was the potential risk to the privacy of children:"

"Facebook's collection of data makes it one of the most influential organisations in the world. Share Lab wanted to look "under the bonnet" at the tech giant's algorithms and connections to better understand the social structure and power relations within the company. "