Red And Blue Algorithms: Study Finds Political Bias In How Major ESPs Decide What Spam Is

Partisanship exists everywhere. Why not in the spam box? Political bias has crept in to the spam filter algorithms  (SFAs) of three major email service providers: Gmail, Outlook and Yahoo, according to “A Peek into the Political Biases in Email Spam Filtering Algorithms During US Election 2020,” a study from North Carolina State University.  

Don’t think it’s dominated by any either side. “Gmail leaned towards the left (Democrats) whereas Outlook and Yahoo leaned towards the right (Republicans),” the study reports. 

To study the problem, the authors created 102 email accounts and subscribed to the emails of two Presidential, 78 Senate, and 156 House candidates, collecting 318,108 emails across the three services from July 1, 2020 to November 30, 2020.

They found that Gmail marks 67.5% from the right (i.e., Republican candidates) as spam, compared to 8.2% from the left. Gmail marked only 10.12% of the left-wing emails as spam while up to 77.2% ended up in the spam box.



In contrast, Outlook is unfriendly to all campaign emails, marking 95.8% those from the left as spam versus 75.4% from the right. 

Yahoo marks 14.2% more left-wing emails as spam than right-wing, while designating 55.2% of political emails as spam. Outlook filtered out over 71.8%. 

Is all this due to conspiracy on either side?

“While we have no reason to believe that there were deliberate attempts from these email services to create these biases to influence the voters, the fact remains there that their SFAs have learnt to mark more emails from one political affiliation as spam compared to the other,” the authors write.

What is true is that SFAs adapt to user preferences, as when a recipient moves emails as spam, or moves them from the spam folder to the inbox. That said, “we did not find any consistent actions that tone could recommend to users to help them reduce the bias in the way the SFA treats political emails that are sent to them.”

Did it affect the election? Possibly. 

‘As these prominent email services are actively used by a sizable chunk of voting population and as many of the voters today rely on the information they see (or don’t see) online, such biases may have an unignorable impact on the outcomes of an election,” the authors conclude. 

The authors are Hassan Iqbal, Usman Mahmood Khan, Hassan Ali Khan, Muhammad Shahzad Department of Computer Science, North Carolina State University, Raleigh, North Carolina.


1 comment about "Red And Blue Algorithms: Study Finds Political Bias In How Major ESPs Decide What Spam Is".
Check to receive email when comments are posted.
  1. Dan Ciccone from STACKED Entertainment, April 5, 2022 at 11:01 a.m.

    Are they moved to SPAM because of the content of the email or because of the naming convention used in the email address or IP from the sender?  There's a big difference.

Next story loading loading..