Twitter’s “Potentially Offensive” Prompt Kills 30% of Posts When Displayed

Nudges towards self-censorship.

A smartphone showing a Twitter prompt reading, Want to review this before Tweeting? regarding harmful or offensive language.

If you’re tired of censorship and surveillance, subscribe to Reclaim The Net.

Last year, Twitter re-launched its system that inserts a prompt whenever it detects that a user is about to post a “potentially offensive” statement.

The introduction of the feature was somewhat controversial, allowing Twitter to determine what was “offensive” and resulted in Twitter being able to nudge the public conversation in a direction that it chooses.

Now, some data has been collected to reveal that impact that these prompts have.

Twitter app prompt asking Want to review this before Tweeting? over a draft reply with potentially harmful language.

In the report, Twitter says that, when issued with the prompt, 30% of users backtrack on what they were about to post, changing it or deleting it.

Bar chart showing actions taken by prompt exposure, including percentages for Cancel Tweet, Revise Tweet, and Send Tweet.

30% is no small number when the impact is extrapolated to Twitter’s almost 200 million users.

Explore more on these topics