Machine Learning – Page 2 – PossibleThreat Articles

Google purges 600 Android apps for “disruptive” pop-up ads

February 24, 2020 0 Comments ad fraud, adware, Android, Android apps, bad apps, cheetah mobile, disruptive ads, Google, Google Play, Google Play Store, Machine Learning, Mobile, out-of-app ads, policy, Spam

Credit to Author: Lisa Vaas| Date: Mon, 24 Feb 2020 12:26:22 +0000

These apps plunk ads in front of us when we’re trying to do something else, often leading to inadvertent ad clicks and much cursing.<img src=”http://feeds.feedburner.com/~r/nakedsecurity/~4/xlWz9s2MFTA” height=”1″ width=”1″ alt=””/>

Security Sophos

Self-driving car dataset missing labels for pedestrians, cyclists

February 14, 2020 0 Comments AI, Artificial intelligence, autonomous vehicles, datasets, elaine herzberg, Machine Learning, pedestrians, self-driving cars, udacity, udacity dataset 2

Credit to Author: Lisa Vaas| Date: Fri, 14 Feb 2020 11:10:23 +0000

Udacity Dataset 2, used to train thousands of engineers, contained thousands of unlabeled vehicles and hundreds of unlabeled pedestrians.<img src=”http://feeds.feedburner.com/~r/nakedsecurity/~4/O3azSOGnuHw” height=”1″ width=”1″ alt=””/>

Security Sophos

Facebook, Google, YouTube order Clearview to stop scraping faceprints

February 7, 2020 0 Comments AI, Artificial intelligence, biometrics, BIPA, CFAA, class action, clearview ai, david mutnick, DCMA, facebook, faceprints, facial recognition, first amendment, Google, hiQ Labs, hoan ton-that, Illinois, illinois biometric information privacy act, Instagram, Law & order, law enforcement, lawsuit, LinkedIn, Machine Learning, Privacy, privacy law, Social networks, twitter, youtube

Credit to Author: Lisa Vaas| Date: Fri, 07 Feb 2020 11:30:03 +0000

It’s my First Amendment right to scrape publicly available face images, its CEO says. Besides, we’re just doing what Google Search does.<img src=”http://feeds.feedburner.com/~r/nakedsecurity/~4/X7rOITrf5_c” height=”1″ width=”1″ alt=””/>

Security Sophos

Twitter bans deepfakes, but only those ‘likely to cause harm’

February 6, 2020 0 Comments alphabet, assembler, cheapfakes, deepfakes, disinformation, facebook, fake news, gan, generative adversarial network, Google, jigsaw, Machine Learning, shallowfakes, Social networks, stylegan, the current, twitter

Credit to Author: Lisa Vaas| Date: Thu, 06 Feb 2020 13:52:30 +0000

Twitter isn’t interested in how the “synthetic or manipulated” media is created, but if it has the potential to cause harm it’ll be removed.<img src=”http://feeds.feedburner.com/~r/nakedsecurity/~4/6IJPjMtC2sA” height=”1″ width=”1″ alt=””/>

Security Sophos

Facial recognition firm sued for scraping 3 billion faceprints

January 28, 2020 0 Comments AI, Artificial intelligence, biometrics, BIPA, class action, clearview ai, david mutnick, facebook, faceprints, facial recognition, hoan ton-that, Illinois, illinois biometric information privacy act, Law & order, law enforcement, lawsuit, Machine Learning, Privacy, privacy law, Social networks, twitter, Venmo, youtube

Credit to Author: Lisa Vaas| Date: Tue, 28 Jan 2020 10:51:21 +0000

A potential class action says Clearview AI is breaking biometrics privacy law by ransacking social media so police can match photos with IDs.<img src=”http://feeds.feedburner.com/~r/nakedsecurity/~4/9Wtt8BCE-R8″ height=”1″ width=”1″ alt=””/>

Security Sophos

Tinder to get panic button, catfish-fighting facial recognition

January 27, 2020 0 Comments AI, Artificial intelligence, catfishing, dating apps, emergency assistance, facial recognition, fake identity, location tracking, Machine Learning, match group, Mobile, noonlight, online dating, panic button, photo verification, Privacy, selfies, tinder

Credit to Author: Lisa Vaas| Date: Mon, 27 Jan 2020 13:28:05 +0000

It’s both a genius move to protect from assault and fraud and a personal data grab.<img src=”http://feeds.feedburner.com/~r/nakedsecurity/~4/knmm8T6klTY” height=”1″ width=”1″ alt=””/>

Security Sophos

Nobody boogies quite like you

January 22, 2020 0 Comments Artificial intelligence, biometrics, dancing, finland, identification, journal of new music research, Machine Learning, music, Privacy, surveillance, university of jyväskylä

Credit to Author: Lisa Vaas| Date: Wed, 22 Jan 2020 10:38:05 +0000

Our unique dancing style can be used by a machine-learning model to ID us, regardless of musical genre. Unless it’s Metal. We all headbang.<img src=”http://feeds.feedburner.com/~r/nakedsecurity/~4/uprytRWl518″ height=”1″ width=”1″ alt=””/>

Security Sophos

Facial recognition is real-life ‘Black Mirror’ stuff, Ocasio-Cortez says

January 17, 2020 0 Comments AI, alexandria ocasio-cortez, Artificial intelligence, biometrics, black mirror, facial recognition, house oversight committee, house oversight committee hearing, joy buolamwini, Law & order, Machine Learning, meredith whittaker, Privacy, surveillance, willie lynch

Credit to Author: Lisa Vaas| Date: Fri, 17 Jan 2020 10:59:27 +0000

“People think they’re going to put on a cute filter and have puppy dog ears, and not realize that that data’s being collected.”<img src=”http://feeds.feedburner.com/~r/nakedsecurity/~4/WEGyDh5kmBU” height=”1″ width=”1″ alt=””/>

Security Sophos

Facebook bans deepfakes, but not cheapfakes or shallowfakes

January 8, 2020 0 Comments cheapfakes, deepfakes, facebook, fake news, Machine Learning, Mark Zuckerberg, shallowfakes, Social networks

Credit to Author: Lisa Vaas| Date: Wed, 08 Jan 2020 11:20:02 +0000

Quick-n-sleazy edits are still OK, such as the 75% slowdown that made Nancy Pelosi slur or the edit that turned Joe Biden into a racist.<img src=”http://feeds.feedburner.com/~r/nakedsecurity/~4/7w8AhUNJatM” height=”1″ width=”1″ alt=””/>

Security Sophos

Emotion-detection in AI should be regulated, AI Now says

December 16, 2019 0 Comments affect recognition, AI, ai now, ai now institute, Artificial intelligence, BIPA, cogito, emotion detection, hirevue, Illinois, kate crawford, Law & order, Machine Learning, oxygen forensics, paul ekman

Credit to Author: Lisa Vaas| Date: Mon, 16 Dec 2019 10:42:34 +0000

It’s built on junk science, yet it’s being used to determine who gets hired, fired, insured, medicated and more, the research institute says.<img src=”http://feeds.feedburner.com/~r/nakedsecurity/~4/MWAhLktWa8c” height=”1″ width=”1″ alt=””/>