Instagram using AI to scan for bullying in photos and captions

Mail Online | 10/9/2018 | Annie Palmer For Dailymail.com
mel4 (Posted by) Level 4
Click For Photo: https://i.dailymail.co.uk/1s/2018/10/09/20/4899266-0-image-a-10_1539113912671.jpg


Click For Video: https://videos.dailymail.co.uk/video/mol/2018/09/25/8168096590838648488/1024x576_MP4_8168096590838648488.mp4

Instagram is taking action against trolls and bullies.

The social media giant on Tuesday rolled out new machine learning tools that are able to 'proactively detect' bullying in photos and captions.

App - Weeks - Firm

It's expected to arrive on the app in the coming weeks, according to the firm.

Instagram already uses AI to filter out comments that violate its Community Guidelines, but this feature takes it a step further.

Ranks - Rates - Images - Captions - Bullying

The algorithm ranks and rates images and captions to detect bullying in photos.

For example, it can detect if a photo is criticizing a person's appearance or character, as well as if the photo is making a threat to someone's wellbeing.

Photo - Instagram - Moderators - Review

Once a photo is flagged, it's sent to Instagram's human moderators for further review.

'This change will help us identify and remove significantly more bullying — and it’s a crucial next step since many people who experience or observe bullying don’t report it,' Instagram CEO Adam Mosseri wrote in a blog post.

'It - Community - Members - Teens - Rates

'It will also help us protect our youngest community members since teens experience higher rates of bullying online than others.'

The firm is also extending its bullying filter to live videos.

Tuesday - Instagram - Comments - Videos - Place

Starting Tuesday, Instagram will be able to block offensive comments on live videos to make sure it 'remains a safe and fun place to authentically...
(Excerpt) Read more at: Mail Online
Wake Up To Breaking News!
Keep the change!
Sign In or Register to comment.

Welcome to Long Room!

Where The World Finds Its News!