Hello, and welcome back to The Ethical Technologist. I don’t know about you, but it really feels like things are coming to head, for better or worse. On one hand, we’re well into a second wave of COVID here in the Netherlands where I live. On the other, Election Day in America will soon be upon us, and it looks like maybe Trump has caused enough problems to cost him the election. And certainly the warm weather is very over: The change of season from summer to autumn is always welcome to me.
There’s no particular theme to this week’s issue; the topics range from surveillance capitalism to high-speed trains, from racist chatbots to colorzing the historical record. As always I hope you find them thought-provoking.
In this issue:
If you have a recommendation for a theme, or an article you’ve seen recently that you think I ought to share, please do let me know. You can reply to this email, or hit me up on Twitter
As always, you can find back issues of The Ethical Technologist in the archives. And if you found this issue thought-provoking and informative, don’t forget to share!
Rite Aid, a chain of drugstores across North America, was recently discovered to be using facial recognition technology to covertly track customers, predominantly in low-income areas. Today, facial recognition technology is rapidly improving, and can be used for so much more than identifying individuals. A recent project from SHERPA demonstrates just how easy it is to leverage facial recognition to judge us in surprising, and often unfairly biased ways.
GPT-3 is the hot new AI algorithm that takes natural-language requests, and outputs a surprisingly convincing response in English. No matter how convincing it is though, like all AI techniques, it’s a mathematical model that’s trained on (in this case) massive amounts of text scraped from the web. And as many of us know, the web can be an awfully sexist, racist place. So we shouldn’t be terribly surprised when it turns out that GPT-3 is a bigot.
Cameras have historically been held up as objective observers of history. We all know now that photographs and movies are easy to alter to create lies, but what about colorizing and upscaling historical footage? Surely there’s nothing wrong with it? There’s a small movement dedicated to doing just that, and the results are nothing short of stunning. But, as I’ve discussed before, cameras lie in surprising ways. And some historians believe that these enhancements add details that can only be guessed at, and risk creating a false historical narrative.
I’m a serious train nerd, and it’s easy for someone like me who has experienced Japan’s high-speed rail (shinkansen) network to think that it’s a modern miracle. But building and expanding the shinkansen network has helped to drain the countryside’s population into Tokyo and Osaka, and shifted funds away from rural transit, leaving small towns in the lurch. But now in the time of COVID and working from home, small towns are learning how to use the appeal of the shinkansen to lure city dwellers back.
New advances in computer vision are revealing things otherwise invisible. Whether it’s seeing behind our backs or around corners by examining faint shadows, or recording audio from the microscopic vibrations of a bag of potato chips, researchers are discovering new ways to observe the world around us. What’s particularly interesting to me is the way that the researchers have grappled with the obvious military implications of their work, ultimately deciding that they can’t do anything about it, and that the research is far too interesting to resist.
And that’s it for Issue 7! Thanks once again to our amazing content editor, Amy Goodman-Wilson, for all her hard work getting our second issue out the door.
And thank you for subscribing. If you enjoyed this issue, share with your friends! If you have an article that should be featured in an upcoming issue of The Ethical Technologist, let me know by either replying to this email (what, a newsletter you can reply to!?), or pinging me on Twitter.
Until next time,
yours,
Don Goodman-Wilson