Nieman Foundation at Harvard
HOME
          
LATEST STORY
From shrimp Jesus to fake self-portraits, AI-generated images have become the latest form of social media spam
ABOUT                    SUBSCRIBE
Oct. 25, 2017, 1:04 p.m.
Reporting & Production
LINK: qz.com  ➚   |   Posted by: Ricardo Bilton   |   October 25, 2017

To ferret out some of the many bots on Twitter that push out political opinions, Quartz has created a bot of its own.

The new bot, @probabot_, is designed to search Twitter for accounts that are heavily focused on political topics. These accounts are fed into a tool called Botometer, which uses machine learning to determine the probability that an account is assisted by or run exclusively by a bot. Quartz has already used the tool to classify over thirty accounts, many of which have tweeted hundreds of thousands of times and have thousands of followers. One bot that Quartz’s bot identified, for example, has tweeted or has been retweeted 1.14 million times since April. Quartz has created a Twitter list for all the accounts that @probabot_ has identified as bots.

The bot’s launch is timely given the ongoing investigation into Twitter’s role in Russia’s campaign to influence the 2016 presidential election. Twitter estimates that false and spam accounts represent less than 5 percent of its monthly active user base. The company said last month that its automated systems flag more than 3.2 million suspicious accounts per week, but @probabot_ shows that not even Twitter is able to spot them all.

@probabot_ is part of an ongoing effort by Quartz to use bots to both inform the site’s reporting and break news. _ransombot, an earlier project, was first to spot when the hackers behind the WannaCry ransom bot cashed out the bitcoin accounts victims had to pay into before regaining access to their machines. Both efforts were created by Keith Collins, a technology reporter at Quartz, who said that Quartz plans to use the bot’s early results to tweak its parameters and improve results in the future.

Show tags
 
Join the 60,000 who get the freshest future-of-journalism news in our daily email.
From shrimp Jesus to fake self-portraits, AI-generated images have become the latest form of social media spam
Within days of visiting the pages — and without commenting on, liking, or following any of the material — Facebook’s algorithm recommended reams of other AI-generated content.
What journalists and independent creators can learn from each other
“The question is not about the topics but how you approach the topics.”
Deepfake detection improves when using algorithms that are more aware of demographic diversity
“Our research addresses deepfake detection algorithms’ fairness, rather than just attempting to balance the data. It offers a new approach to algorithm design that considers demographic fairness as a core aspect.”