AI Filter Launched To Block Twitter Cyberflashing

It seems weird to report, but a tiny but determined group of Twitter users think it's a good idea to send a message (DM) photo of male genitals to complete strangers.

Does this sound a bit like street harassment flickering in digital form?

It was done to developer Kelsey Bressler after receiving such an unsolicited picture as a DM via Twitter last August. Later, she told the BBC:

You’re not giving them a chance to consent, you are forcing the image on them, and that is never okay.

Instead of shrugging it off, she and a friend had the idea of using AI pattern recognition to scan the images out before they were used. But that AI still needed a set of – ahem – images to practise on, which Bressler requested via Twitter.

Bressler has reportedly received over 4,000 photos in response – enough to train the system to a state where it has just been released as a Safe DM service that anyone can sign up for.

Media site Buzzfeed checked Safe DM against a set of photos taken from Wikimedia Commons and found that it was working well, although with a lag of a few minutes.

In experiments, the filter blocked penises in a variety of states, including full body shots and condoms and drawings. It even blocked examples that looked like a penis without being a single one.

Conclusion: recipients can see an image if it is opened immediately, otherwise it should be secure. Bressler told Naked Protection that it will also block pictures of female genitals, but no proof of its efficacy has yet been made public.

For now, Safe DM is only available on Twitter, but other channels may be included in future updates, Buzzfeed said.

The filter asks for a lot of permissions, but doesn't read the text content of the DMs, she said. That's because:

Unfortunately, Twitter doesn’t allow us to pick and choose. It’s all or nothing.

Cyberflashing continues to be a rising threat on several platforms. Huffington Post UK published an article last May citing hundreds of women who had encountered it via email, SnapChat, Instagram, Twitter, Facebook Messenger, and – the most frequently discussed platform – AirDrop (which has been in trouble for this sort of abuse before).

Despite more general abuse regulations, the chances of conviction for cyberflashing remain between very low and non-existent. But at least with Safe DM, Twitter users need to switch to something.

Explore Instagram posts for tag #cyberflashing - Instastory.net