😷 AI spotted Coronavirus before it was announced, Londontown starts real-time AI-powered surveillance

PETA wants a robo groundhoggo to tell us when spring is coming

The Good

Image result for global disease spreading gif

China has not been very forthcoming about the Coronavirus. The first known case was December 1st in a city called Wuhan, but Chinese officials didn’t report it to the WHO until December 31st (and punished some doctors for trying to raise the alarm before then), which didn’t announce it to the public until early January. A Canadian AI company called BlueDot, however, spotted it on the 31st and had already alerted its clients to the possible outbreak.

BlueDot's system looks at billions of data points, like airline ticket sales and news stories, uses that to map how viruses could spread by planes and trains, and then predicts areas that are at the highest risk of an outbreak. The AI even correctly predicted where the Coronavirus would jump to after Wuhan.

With AI, unforthcoming governments won’t be able to hide certain information, like outbreaks and air pollution stats, from all the different data crunching algos out there. In matters where time is of the essence (like this one), AI might be a lifesaver.

The Bad

The London Metropolitan Police announced last week that it will start using real-time face rec tech to identify suspects. London famously has the largest network of cameras for any city outside of China, numbering 500,000 (!!!).

This rollout is smaller than that for now, only in a few spots run for a few hours each day with “bespoke” list of suspects it is on the lookout for. The police say this system generates false alerts only .1 percent of the time, which sounds pretty remarkable considering some of the horrendous stats other trials in the U.K. has produced.

Even with this small test, groups are sounding the alarm about how few guardrails there are for this tech in the U.K and how many unanswered questions remain. In an opinion piece, one researcher points out:

British law enforcement requires warrants to search homes, businesses and phones. Why do these areas enjoy greater legal protections than our faces, which links our physical and digital selves? And how is it reasonable and proportionate to subject us to a real-time, 24/7 digital dragnet that jeopardizes our privacy, anonymity and many of our civil liberties, when we could create a legal requirement for targeted, geographic-and-time limited face searches backed by warrants? These are questions the Metropolitan Police must answer.

It’s easy to disregard China’s system as something that wouldn’t happen in democracies. What’s going on in London is harder to dismiss for the West.

More New

PETA, my favorite stunt factory, wants to replace the real animals used on Groundhog Day with robot ones (and those could *actually* predict the weather).

U.S. college students don’t want face rec tech on their campuses.

The EU will be building its own cloud service for data sovereignty security Jeff Bezos can’t offer.

Clearview AI, a system used by 600 law enforcement agencies (and written about in this newsletter last week), got banned for New Jersey cops.

AI is making license plate reader tech very cheap, so every podunk police department will be running your plate soon.

Tinder wants AI to save you from shitty messages.

As someone who dealt her A/C unit die in the middle of summer, I think the idea that Nest could use AI to spot early issues in HVAC units is a great idea.

Last night a robot DJ took my radio job and made a playlist based off an algo.


Until AI can assure me this cold is nbd,