AI MUST READS — W9 2018, by City AI
Artificial Intelligence, Machine Learning and related fields are in a constant state of change. We want to inform but also encourage discussions on well presented topics we think are necessary in the context of putting AI into production. Every week we’re picking applied AI’s best articles plus adding a discussion starter
1. Family fun with deepfakes. Or how I got my wife onto the Tonight Show
Once again there has been a piece of new technology with AI as its driving force that has been in the news for all the wrong reasons — Deepfakes.
If you haven’t heard of them yet, ‘deepfaking’ consists of using an algorithm that allows for the user to replace the face of an actress or actor with that of someone else of their choosing, and its startlingly good at what it does.
This article highlights a technology that is currently causing great controversy with sites like reddit and pornhub issuing blanket bans across the materials as people pump out deepfake porn videos, having replaced the actors face with the ones from Hollywood! However, this article spends just as much time focusing on the potential productive applications for this technology as on the controversial topics surrounding deepfakes which have inundated media sites for the past week. Sven Charleer gives an explanation that makes sense, without being too technical for those that want to understand just how it works.
This article is genuinely enjoyable to read whilst also instructive. If that’s not a sign of a good article, then what is?
2. Good News: A.I. Is Getting Cheaper. That’s Also Bad News.
Again and again on an almost daily basis there are articles and ‘reports’ that are published about the evils and dangers of AI and how it’s going to steal your job. It seems to be an ever growing bandwagon that often serves not to cause people to ask productive questions and promote change, but instead to fear this technology and everything that will come with it.
Whilst not as guilty as a lot of the articles that have been recently published, Cade Metz does fall into a number of pitfalls that I find are common. Within the second paragraph, talking about a startup’s recently unveiled drone, he refers to it as a bloodhound, a term clearly used because of its negative connotations. Further on within the article he delves further into other simple applications of AI and its relevant technology, however once again he centres around the dangers and problems that can be caused by AI. Not once does he draw any parallel between the dangers of this technology and its benefits.
Now, to deny that artificial technology, machine learning, etc. has problems would be naive. Like every new technology it has its problems and its potential power means that the conversations being held within the community are much needed to prevent more abuses of the technology like the recent ‘deepfake’ debacle. However, we also need to talk about its benefits, where will AI improve the general quality of life? How can AI save lives or reduce the stress placed upon doctors and nurses? And we’re not talking about the hype…
A permanent focus on the dangers of AI will only lead to fulfil the future we have already predetermined for it. Read the full report and think about it.
3. Bad Data Is Ruining Machine Learning, Here’s How To Fix It
Short, sweet and straight to the point. There isn’t much else to say about this article other than perhaps the ending was abrupt. Yves Mulkers cuts straight down to the point in an impressively efficient matter, consume this article and take on board the lessons within and it wont be able to lead you astray.
Have we missed any well thought out articles? Please send them our way!
