Instagram abnormal activity authenticates and logs in (Instagram starts warning users before posting potentially offensive headings)

Starting today, at the beginning of Instagram, they will upload photos or a "may be unpleasant" caption to their main feed video to warn users that the company has announced. If an Instagram user publishes content that the service's AI driver tool believes may be harmful, the application generates a notification stating that the title "looks similar to other reported headings." It then encourages users to edit the title. But they can also choose not to change the publication time.

The new feature is based on reviews similar to AI power tools launched by insta gram in July. The company said urging people to reconsider making potentially harmful comments had produced "promising results" in the fight against cyber bullying.

This is just the latest in a series of measures taken by Instagram to address bullying on its platform. In October, the service launched a new "limit" feature that allows users to stop their bullies. Last year, AI was used to filter offensive comments and actively detect bullying in photos and subtitles.

Unlike other audit tools, Instagram relies on users to find out if any of their comments are out of line. It is unlikely to stop the more determined bullies on the platform, but I hope it will protect people from frivolous insults.

Instagram said the new feature was being launched "exclusively" but would be expanded globally in the coming months.