Koo introduces new safety features for content moderation

As per Koo, the microblogging platform has introduced new content moderation features that aim to enhance the safety and security of users while using social media. The platform’s in-house developed features can detect and block nudity and child sexual abuse materials in less than five seconds, label misinformation, and conceal toxic comments and hate speech. 

Here are details on Koo’s safety features:

Nudity: Koo’s in-house “No Nudity Algorithm” is designed to proactively detect and instantly block any attempt by a user to upload pictures or videos containing child sexual abuse materials, nudity, or sexual content, as per the company. Koo says that the algorithm can detect and block such content in less than five seconds. When a user posts sexually explicit content, they are immediately prevented from posting content, being discovered by other users, being featured in trending posts, or engaging with other users on the platform.

Toxic Comments and Hate Speech: According to Koo, its in-house technology proactively detects and removes toxic comments and hate speech in less than 10 seconds to ensure they are not visible to the public.

Violence: Users are provided with a warning overlay for content that contains excessive blood, gore, or acts of violence. 

Impersonation: Koo employs an in-house algorithm called ‘MisRep Algorithm’ to regularly scan its platform for profiles that use content, photos, videos, or descriptions of famous individuals in an attempt to impersonate them. When detected, the algorithm swiftly removes any pictures or videos of the well-known individuals from the profiles and flags the accounts for future monitoring of potential improper conduct.

Misinformation and Disinformation: The company revealed that it developed an in-house algorithm known as the ‘Misinfo and Disinfo Algorithm,’ which continuously and instantly scans all viral and reported fake news on the platform using both public and private sources of fake news. The algorithm identifies and labels misinformation and disinformation present in a post, effectively reducing the dissemination of viral false information on the platform.