The YouTube algorithm that I helped build in 2011 still recommends the flat earth theory by the *hundreds of millions*. This investigation by @RawStory shows some of the real-life consequences of this badly designed AI.

This spring at SxSW, @SusanWojcicki promised "Wikipedia snippets" on debated videos. But they didn't put them on flat earth videos, and instead @YouTube is promoting merchandising such as "NASA lies - Never Trust a Snake". 2/
A few example of flat earth videos that were promoted by YouTube #today:
https://t.co/TumQiX2tlj 3/
https://t.co/uAORIJ5BYX 4/
https://t.co/yOGZ0pLfHG 5/
https://t.co/Tj1zVkoIFG 6/
https://t.co/uUXCFjXxHg 7/
And this one would be hilarious if not recommended so massively to people who are the most vulnerable to conspiracies
https://t.co/EKed0B1XhD
8/
@nasa's @mlthaller nicely shows in 3 minutes why the earth is round and wonders: Why do so many people believe in flat earth? But the conspiracy is promoted by YouTube 100,000,000s of times, so I'm actually surprised that *so few* people fall for it 9/

https://t.co/yZpqdiJgsR
NBA star @KyrieIrving fell for it and explained that he was convinced both by Instagram and YouTube recommendations 10/

https://t.co/cqYz3SbDO8
So basically we have the two best AIs of the world, on Instagram and YouTube, competing to convince people that the earth is flat. Because it yields large amounts of watch time, and watch time yields ads. This is a #raceToTheBottom 11/
Flat Earth is not a "small bug". It reveals that there is a structural problem in Google's and Facebook's AIs: they exploit weaknesses of the most vulnerable people, to make them believe the darnedest things 12/
Another real-life harm of this conspiracy promoted by AI: in Nigeria, Boko Haram used the flat earth theory to motivate the killings of more than 600 geography teachers 😰 13/

https://t.co/M2mVtZRut9
https://t.co/myxPsrhlKa
More info on the link between Boko Haram and flat earth in this detailed @latimes article 14/
Full @latimes article here:
https://t.co/M2mVtZRut9
15/
How do we know if harmful videos get promoted by the AI? To monitor that, I created https://t.co/0RjD2PtTvV, a site that scans YouTube recommendations daily. If harmful videos get recommended by the AI from many channels, they will appear at the top of the list there. 16/
For now, it is available in English for YouTube yet, but with the help of universities and @HumaneTech_, it will soon cover more languages and platforms 17/
The real danger of AI is not that it becomes "self-conscious". It's that we trust it too much and slowly become imbeciles

Flat-earthers are the canaries in the coalmine /18
I've seen some people state that: "canaries are only a small fraction of our users!"

I think they're missing the point. 19/
"Ban all the canaries from the coal mine!" also misses the point.

With AI in charge of our information, we're facing a brand new, existential problem that concerns all of us. We need to develop tools to understand it better. 20/
One more key to understand this: this @KELLYWEILL's article shows that some people spent *days* on YouTube after being converted to flat earth

From the algorithm's point of view, flat earth is a gold mine.

Full article: https://t.co/LPjCKpbwXj

21/
Also fascinating in this article is that Logan Paul was a speaker at a Flat-Earth conference, where he said "I’m coming out of the Flat Earth closet". He has 18,610,154 followers on YouTube, mostly pre-teens and teens

https://t.co/LPjCKpbwXj 22/

More from Tech

The entire discussion around Facebook’s disclosures of what happened in 2016 is very frustrating. No exec stopped any investigations, but there were a lot of heated discussions about what to publish and when.


In the spring and summer of 2016, as reported by the Times, activity we traced to GRU was reported to the FBI. This was the standard model of interaction companies used for nation-state attacks against likely US targeted.

In the Spring of 2017, after a deep dive into the Fake News phenomena, the security team wanted to publish an update that covered what we had learned. At this point, we didn’t have any advertising content or the big IRA cluster, but we did know about the GRU model.

This report when through dozens of edits as different equities were represented. I did not have any meetings with Sheryl on the paper, but I can’t speak to whether she was in the loop with my higher-ups.

In the end, the difficult question of attribution was settled by us pointing to the DNI report instead of saying Russia or GRU directly. In my pre-briefs with members of Congress, I made it clear that we believed this action was GRU.

You May Also Like