Instagram is a popular photo-sharing app/website that offers hundreds of millions of users access to a humongous userbase sharing their work, daily life and more. But this also means it is a home ground for spreading fake photos and news and that’s something Instagram is trying to curb with the new feature that flags fake photos from the platform.
Apparently, the feature which works on feedback from the user community and uses an image-matching and combination of other algorithms is occasionally tagging photoshopped images as fake. This does hurt artists and photographers who use Facebook and Instagram among other handles to showcase their work. This indeed makes photographers worried about where to share their work.
The feature in place hides images that it finds fake. Here, it overlays a warning that the image is reportedly fake and uses external fact-checkers and other services to validate the same. This overlay is similar to how Facebook puts an overlay over explicit images although showcases the images once the user accepts the warning and intentionally wants to see it.
Although many people have observed this, Toby Harriman, a San Francisco-based photography received a notification just after posting an image that had a person standing against rainbow-colored mountains. Although the image was photoshopped, Instagram’s algorithm flagged it as a fake photo and thus, it went under an overlay titled ‘False Information’. There are independent fact-checkers that explain a bit about the mountain aka its ‘alt’ description marking it as ‘False’.
Moreover, Instagram removes these false images from its Hashtag and Explore pages as well and report them flagged for future references. Today when photos are posted on Instagram after a bit of post-processing which could be anything from adding color gamut to adjusting focus, color saturation, cropping images, etc which could be tagged as photoshopped and so will be removed from the respective Instagram handle.
Instagram is fighting hard against bullying content as well as it recently unveiled a feature that warns the user if the caption contains any bullying words or phrase that was earlier reported by other users.