Welcome to the News Byte, a smaller piece of the News Beat that lets us hone in on one single important story about the internet. This week...
“Millions of Kenyans are being manipulated on Twitter.” That’s how Mozilla Fellow Brian Obilo put it, describing research he conducted with fellow Mozilla Fellow Odanga Madung. The two led an investigation into disinformation on Twitter and how it affects Kenyan society. Their findings?
Madung and Obilo uncovered at least 11 disinformation campaigns involving 3,700 Twitter accounts and over 23,000 tweets. Eight of the campaigns even trended on Twitter. The campaigns involved influencers, who were paid to spread disinformation and harassment, and even some verified Twitter users with blue checkmarks.
Journalists, judges and other civil society members in Kenya were all subject to coordinated disinformation campaigns on Twitter and the site did very little to stop it — until Madung and Obilo’s report. Twitter then responded by removing 100 accounts involved in spreading disinfo on the platform.
We chatted with Odanga Madung about the reception to his and Obilo’s findings, what he thought about Twitter’s response and what’s next.
Xavier Harding: Where did the idea of investigating Kenya-related disinformation on Twitter come from? When did you first realize this was happening?
Odanga Madung: It’s been happening forever. If you speak to any Kenyan on Twitter, we are so used to it. Manufactured trends have been part of our ecosystem for the longest time. It’s quite the phenomenon here.
What’s been the response to your investigation so far?
We’ve had many activists thank us for shining a light on this, saying it’s a very important report for our digital ecosystem. Other activists have read the report and told us, “this is exactly what I’ve been going through.” It’s resonated with a lot of people, and that feedback has been very welcome.
Then of course there’s the Twitter response to the report. It’s been interesting in understanding its capacities in terms of dealing with this problem.
Twitter removed 100 accounts from the platform that had a hand in spreading the disinformation you focused on. Does that fully fix the problem?
It does not, because it does not address the systemic issues and incentives that allow these campaigns to exist in the first place. Twitter continuously resurfaces these sorts of disinformation campaigns through its trending algorithms.
One really interesting part of the report was where you mentioned how Twitter is able to profit from this sort of content on its platform via advertisements. With that in mind, can we trust Twitter to properly address the problem if the company stands to profit from it?
Absolutely! Because brand safety matters. It’s an issue that platforms have been trying to deal with for a long time. No advertiser wants to be associated with hate speech.
This takes me back to 2014, 2015 all over again when the biggest incentive to fix the Islamic radicalization problems was advertisers speaking out because they realized their ads were running against ISIS propaganda. The situation we talk about in this report is the same situation.
What else should Twitter do so we can see lasting change here?
We need to start pushing Twitter to publish regular policy enforcement reports, similar to how Facebook does with its Coordinated Inauthentic Behavior reports that show the deceptive campaigns the site removes from its platform each month. Twitter needs to copy Facebook here.
What do you want people to ultimately take away from this report?
Ultimately I think Twitter needs to pay attention. I know as a Kenyan who logs onto Twitter every day, you’ll notice a bunch of weird hashtags trending on there. This is negligence in regards to this feature. Twitter doesn’t really pay attention to the ways trending hashtags can be misused and, if it does, it isn’t doing anything about it.
I know getting rid of the trending section may not get rid of this information, but it would make spreading it a lot harder.
The News Byte
Anna Jay, Kaili Lambe