Harassment over the internet is nothing new and Twitter is no exception. Whether you make an account to have political debates, share memes, or conduct business, it does not warrant any unsolicited NSFW messages from strangers. Twitter acknowledged the growing problem and offered a few different ways to protect your DMs. However, if you need an extra layer of protection the Safe DM filter is a nice little tool. It filters out unsolicited pictures sent to your Twitter DMs.
What is Safe DM?
Safe DM is an AI-based filter developed by twitter user @raeBress. It basically scans your Twitter DMs for any unsolicited pictures and automatically deletes them for you. Using the filter is absolutely free and you just need to enable it once. After that, it will scan new incoming messages that contain an image and identify the ones that may be unsolicited nudes from strangers. Post identification, the image is deleted and a message is sent to the user stating that the image was deleted because it was inappropriate.
Safe DM can access your messages as stated in the terms and conditions of the app. It also takes a few minutes to process and delete the unsolicited NSFW pictures.
Enable the Safe DM filter on your Twitter account
Log in to your Twitter account in a web browser. After logging in, go to the Safe DM website and click the Sign-up button.
Next, click on the “Register to start using the filter” button and enter an email account to proceed.
Once that’s done, you’ll see a status page that would show the filter configuration. Now, your Tweets are protected by the filter.
Now, anyone who starts a thread with you will see a message “This account is protected by Safe DM filter“. That should be enough to discourage them from sending an unsolicited picture but even if they do, the filter would automatically delete the image from your thread and show this message “This image was NSFW. Deleted!” instead.
The images below are from the perspective of a person who’s sending the Nudes.
The image below is from a protected account and you only see a prompt that the image was deleted.
During my brief testing, I tried a bunch of test images. These images included NSFW images from the internet, some original NSFW pics of me, and a few control images. The filter always deleted the ones that were NSFW. An important point to note is that the filter would delete the NSFW images regardless of gender which makes it useful for everyone.
The filter is designed to block only images, but it also successfully blocked some NSFW GIFs. I contacted the developers and they confirmed that certain GIFs and videos may get blocked, only if those are NSFW. Another thing that I noticed that the filter takes a few minutes to actually process the image and delete it. So if you immediately open the thread after getting notification, you may accidentally see the image before the filter has had a chance of deleting it.
Safe DM offers an added layer of protection to your Twitter DMs so that you can use your account without worrying about any unsolicited images. However, If you ever want to remove the filter from your account, you can simply navigate to Settings>Content preferences> App Access> DM-Filter> revoke Access.
How does it compare with Twitter’s Inbuilt Quality Filter?
Twitter also offers a little bit of inbuilt protection to its users and you can enable the Quality Filter to bring it into effect. It hides the threads of incoming users and shows it under Message Requests instead. You can then choose to accept or reject the message but you may still end up seeing the unsolicited image. In my experience, the “Quality filter” doesn’t work reliably and at times you can still see the images in the thread. Having said that, you should still enable it to add a layer of protection.
This was a quick way to protect your Twitter account against random strangers filling in your DMs with NSFW images. The filter works like a charm and even though developers say that it works 99% of the time. The added layer of protection would surely set you free to do much better things on Twitter. What do you think of this filter, let me know in the comments below or reach out to me on Twitter if you have any questions?