Your AI Friday – 10101

Welcome to the weekly selection of links compiled by PAC's AI team. This week, we'd like to invite you to a little game: Who can define a suitable image for AI? Enjoy reading!

 

#MUSTREAD

 

#AIForGood #ClimateChange #AIAsAGameChanger

Global warming is – as the name implies – a ‘global’ issue and one to which the entirety of nations have decided to find a common response trough the 2015 United Nations Climate Change Conference. To win this battle, artificial intelligence could in fact prove to be a valuable tool: For the area of smart agriculture, the optimization of energy consumption or the protection of the oceans, the article below presents concrete cases where artificial intelligence already provides real added value.

"Uriarte says her work could not be done without artificial intelligence. "AI is going to revolutionize this field," she said. "It's becoming more and more important for everything that we do. It allows us to ask questions at a scale that we could not ask from below. There's only so much that one can do [on the ground] … and then there are areas that are simply not accessible. The flyovers and the AI tools are going to allow us to study hurricanes in a whole different way. It's super exciting."

Artificial intelligence – A game changer for climate change and the environment - Renee Cho
Source: phys.org

 

#MinorityReportInJapan #PreventionBeforeAction #CheckThatGuy

In Japan, too, AI solutions are gradually, but surely, integrated into surveillance processes, be it in shopping centers to identify suspicious behavior, i.e. behavior that deviates from that of a reference group, or on the streets of Kyoto to detect certain types of wrongdoing that may occur in a given neighborhood.

"When a security camera detects movements typical of a shoplifter, such as pacing back and forth along product shelves or a person stopping and then checking left and right several times, built-in AI technology kicks into gear. In addition to sending a message and visual images to the smartphones of store employees and security guards, a speaker attached to the camera broadcasts a warning such as “a store employee is coming” to the person acting suspiciously."

Police departments turn to AI in bid to anticipate crime
Source: The Yomiuri Shimbun

 

#AIExperiment #AIBias #TheDataSetIsCrucial

MIT's experiment emphasizes the importance of the kind of data used to train artificial intelligence. To avoid bias, or worse, to avoid ending up with an AI like Norman, it is essential to have abundant data, but, above all, such of quality. In fact, data quality will become a major challenge for AI-based solutions.

"With AI, you get out what you put in, so it’s important that these platforms are trained to avoid bias, and preferably not left to browse the darker corners of Reddit for long periods of time."

MIT creates AI-Powered psychopath called ‘Norman’ - Ryan Whitwam
Source: Extremetech.com

 

#AIpictures #NonSense #RobotsEverywhere

Here comes a blog post that brings in a little humour, sketching out a real problem in the marvellous world of AI: the databases of images used to illustrate articles about AI! They are always pretty similar: We get to see futuristic robots, brains that are half human, half machine, or other fluorescent neural networks. Even if they are only a nice extra, these images do have an impact on our collective unconscious and convey an idea of AI that is far from reality. It's a pity that the author of the article does not provide any recommendation himself as to what he would deem more suitable.

"It seems like every article on AI has a picture like this — just a glowing head overflowing with the power of computers. I guess part of the problem is that stories without pictures are boring and it’s really hard to illustrate the idea of AI. So they have to throw in something."

The Real Scandal of AI: Awful Stock Photos - Adam Geitgey
Source: Medium

 

    


#2MinutesLeft

AI in all its facets: ads, videos, partnerships, computer graphics... to read, see and listen to this week.

#AllIsAboutDebuggingSkills  - Why is machine learning 'hard'? - ai.stanford.edu

#MeetWanda - Unit4 & AI: Help me, Wanda! - pac-online.com

#ConservationOfTheEnvironment  - Use artificial intelligence to identify, count, describe wild animals - sciencedaily.com

#AIToImproveHealthcare  - Four ways AI can make healthcare more efficient and affordable  - Medium

#DontBe(toomuch)Evil - Google reportedly leaving Project Maven military AI program after 2019 - TheVerge.com

 

    


   Follow the A(I) Team on Twitter:

    'Hannibal' @OlivierRafal 

    'Looping' @dagautier 

    'Futé' @VicCrete 

    'Barracuda' @StephaneLinise