The new autoblock feature, revealed last year in September, is now being tested out by Twitter to block potentially harmful interactions.
Twitter is now testing the autoblock feature they announced last year. When Twitter announced the feature back in September it was being called “Safety Mode” and was undergoing a beta test with a “small feedback group.” Senior product manager, Jarrod Doherty, described the feature in a blog post as such:
“When the feature is turned on in your Settings, our systems will assess the likelihood of a negative engagement by considering both the Tweet’s content and the relationship between the Tweet author and replied. Our technology takes existing relationships into account, so accounts you follow or frequently interact with will not be autoblocked.” Accounts are temporarily blocked for seven days, but the company says you can undo autoblocks made in error.
Twitter displayed Safety Mode in February at its Analyst Day presentation. At the time, we didn’t know when it would officially roll out
Now, people are reporting the feature’s effects, which are beginning to be seen by more users on the platform.
Never saw this before: Twitter autoblocked someone from even *viewing* ABC News.
(Shared from elsewhere.) pic.twitter.com/dEH2Ip7Vfm
Love Tech?
Get the latest Tech news, features, updates and giveaways straight to your inbox Learn more
— Larry Sanger (@lsanger) March 16, 2022
Twitter is testing a new autoblock feature which automatically blocks ‘potentially harmful language or repetitive, unwelcome interactions’ from accounts. pic.twitter.com/fZT5m3kIYw
— Pop Base (@PopBase) March 17, 2022
When someone is “autoblocked” people have reported that they have been prevented from seeing the page’s content and receive this message:
Autoblocked by Twitter
You’re temporarily autoblocked from viewing and interacting with @[user]’s Tweets. this happened when they were in Safety Mode, and we flagged your interactions as potentially abusive or spammy.
We know we don’t get autoblocks right all the time, so we’re working to improve our detections.
What’s Safety Mode?
Here are ways to keep your conversations healthy— and avoid getting autoblocked.
Remember the human
Avoid repetitive, uninvited replies
They can overwhelm people, even if the content is innocent.
Skip the insults, name-calling, and harmful remarks
Even if they seem playful to you, others might find them hurtful.