In its efforts to improve performance, the platform "Twitter" has turned to internet hackers to identify "alleged biases" in its image cropping algorithm after facing accusations of racism. Twitter has long been accused of favoring white faces over black ones in its automated algorithms. This led to a study that reportedly nearly resolved the issue, although it seems to still persist, according to the tech site "Verge."
The image cropping algorithms cut parts of images to prevent them from taking up too much space in users' timelines and to display more images in tweets. Twitter has opened the door for anyone who can solve the problem, whether they are hackers, researchers, or enthusiasts, in a competition that will award winners with cash prizes through the tech site "Engadget."
The microblogging site hopes that by allowing hackers and others access to the image cropping algorithms, they can find out how these algorithms can be harmful, such as cropping images in a way that perpetuates stereotypes or removes them altogether. Participants in the competition must provide a summary of their findings, along with operational data from the algorithms themselves to prove the existence of the problem.
After this, Twitter will assess the most harmful points and their impact on users and then address them. The winner will receive a prize of up to $3,500, with other prizes of $1,000 available as well. Twitter states that submissions for the competition are open to everyone until August 6. Typically, Twitter pays $2,940 to anyone who discovers a limited error, such as in "retweeting and liking," as part of its bug bounty program. Meanwhile, the reward for discovering a vulnerability that results in a hack of an account on the site is $7,700.
