Content Moderation at Koo

By Koo App

Koo’s core mission is to provide users a safe and trusted micro blogging platform in a language of their choice. We believe in people power and that’s what micro blogging platforms are all about. It’s all about the users that use and adopt the platform.

In order to provide our users with a wholesome community and meaningful engagement users are expected to abide by our Community Guidelines. Free exchange of thoughts and ideas is valued, while at the same time adhering to our legal responsibility of complying with the letter of law and removing content which violates Community Guidelines. These Community Guidelines are designed to foster a safe and respectful space for our users to express themselves, with the highest regard for users’ freedom of speech and expression.

We respect the local laws of the country we operate in. We believe that lawmakers are in the best position to understand the local cultural context within which laws are formed. We believe that creating our own legal rules and structures over and above the local law is too authoritarian. We wouldn’t want to do anything beyond the local laws that may come in the way of freedom of speech.

We hope this write up helps users understand Koo’s approach to Content Moderation.

Content that violates the law of the land or is the subject of a legal order

Content that violates the law of the land or is the subject matter of a valid legal order will be removed. Users are informed of the reasons for removal and they may take recourse to such legal remedies that are available to them including an appeal process.

Categories of Moderation

Broadly, moderation actions are taken on Content, User Profiles and Spam.

(i) Action on Content: Koos, Re-Koos, comments, profile photos, handle names and/or profile names, may be removed with or without prior notice, if they are found to violate Koo’s Community Guidelines. However, this action will not affect the user account or other data connected with such a user. The principle is to ensure that bad content is not in circulation.

(ii) Action on User Profiles: If a user repeatedly and egregiously violates our Community Guidelines or engages in any illegal activity, action may be taken to restrict their visibility or interaction with other accounts.

A user’s freedom to express themself is curtailed only on the basis of their actions. Only in extreme and repeated violations or in compliance with government orders, we may permanently remove a user from the platform.

(iii) Action on spam and malicious content: In the interest of user safety and safety of the platform, Koo adopts a wide definition of spam to include content that, among others is, irrelevant or repeated or malicious or violates rate limits set by Koo.

How are violations identified

(i) Human Moderation: Any registered user can report potential violations using the in-app reporting feature (two dots on the top right corner of a Koo/Comment/Re-Koo) and selecting an appropriate reason for reporting. Our team of moderators will review the reported Koo and take action, as required.

(ii) Automated tools: Koo deploys and continues to examine several automated detection tools to help in content moderation and keeping Koo platform safe:

  • Koo has created a corpus of expressions including words, phrases, abbreviations and acronyms that are considered offensive or sensitive across its operating languages. This corpus is created in collaboration with government and non-governmental organizations and also based on the frequency of their use and context as seen on the platform. Koo uses self-developed automated tools to identify and remove such content. This is an effort to reduce abuse and promote a fair use of language.
  • Koo engages with various vendors and uses in house tools to develop cloud-based AI models to create visual content moderation tools especially in the context of nudity and child sexual abuse.

While we take utmost care and exercise due caution in our content moderation practices, on occasion we may make a mistake. If you are of the view that your content was removed in error and wish to reinstate the content, you are welcome to submit an appeal for reinstatement and we are happy to reconsider.

Handling Misinformation

Koo strives to encourage healthy discussions, foster creativity of thought and enable exchange of views and opinions. To achieve this, Koo Community Guidelines specifically prohibit sharing of false, misleading or unverified information.

While the guideline is fairly simple, handling misinformation is one of the most difficult problems in the social media world today.

There is a growing expectation that social media companies will step in to solve the problem for users and society. This sounds attractive at first, but the reality is that misinformation does not start with the social media company. Misinformation starts with users posting something that they either (i) know to be false or misleading, or (ii) unknowingly posting material that is false, but the user believes to be true.

Koo’s approach to handling misinformation is based on first principle rules and is built on several layers of checks and balances.

While anonymity on the internet has its benefits, it is also a harbor for individuals who propagate misinformation.

Koo provides its users the means to voluntarily self verify themselves as real humans who have a genuine voice on the internet. We believe that users are more likely to trust accounts which have declared themselves to be “not bots”.

This does not mean that Koo stores any data about an individual, a green tick means that the account has been identified as related to a human being. Self verification on Koo is free and will continue to be so.

Any user on the platform can report content that they believe to be false or inaccurate using our in app reporting mechanisms. This content is flagged to our trust and safety for review.

Users have also been empowered with tools to directly reach out to fact checkers. We constantly keep curating and listing fact checkers on our website and users can interact with them without any interference from the platform.

The third component of our approach is to proactively detect, flag and if found egregious, remove false content. Our algorithms look for spikes related to dramatic events, trending hashtags and abnormal actions by user handles.

Based on the severity and frequency of these spikes, we will remove content which is outright false while in other cases we may take a combination of steps, including, slowing circulation, elevating content from experts accounts or overlaying with warnings.

Overall, our approach is to be a neutral intermediary and assist in democratizing the fact checking process. We will continue to create features that empower users to fact check with a choice of fact checker and on their own terms.

Fact checking is a shared community responsibility and Koo is willing to work with governments and society to find ways and means to create more resilience and transparency in the online world.

Handling Hate Speech

The world today is more super connected than ever. With increase in internet penetration more and more people are connecting with each other and interacting with each other’s thoughts and opinions. This is a super positive feeling for users.

Hand in hand with this positivity is also the natural feeling of dislike, discomfort and disagreement. All these reactions are natural and naturally human.

Social media platforms have the unique responsibility and privilege of respecting all of these very human reactions and giving them space and freedom to exist. However, in the ebb and flow of everyday human emotions, dislike, discomfort and disagreement can very easily become disgust, resent and discord. Social media platforms must also handle these and put them in the right boxes.

Koo’s response to this challenge is based on three pillars (i) Definition (ii) Identification, and (iii) Management.


Koo defines hate speech to mean content which contains excessively inflammatory or instigating or provoking or demeaning language against a person, religion, nation or any group of persons. This definition is in line with the legal requirements of most countries and universal human rights. The rules are simple:

(i) Treat others with dignity, respect, and empathy. The rules that apply to human interactions in the offline world are applicable equally (if not more) in the online world.

(ii) Criticism or dislike does not amount to hate speech. Valid and well-intentioned expressions of healthy disagreement are welcome. Any personal attacks and appeals to prejudices are not.

(iii) Avoid discourteous or impolite or rude statements to express disagreement while making your point. Sarcasm and parody should be used in context.


In the online world hate speech manifests itself as (i) abusive text (ii) abusive handle or user names, and (iii) pictures and videos. In order to identify/detect these versions of hate speech, we rely on our community of users as well as in house algorithmic tools.

All users have the ability to report content which they find offensive, using our in-app reporting features. After reporting, the content is handled by our content moderators based on severity and frequency.

Additionally, abusive words and slurs are also detected algorithmically using our well-researched and constantly updated dictionary of abuse. The dictionaries are not just a storehouse of prude language, but are a real reflection of language used everyday, by everyone.

For example, the word f**** may not get removed if its an exclamation, however f*** your m**** will certainly get attention. A similar algorithmic process is used for abuse through visual means.


The management of reported content is triangulated on meaning, severity and frequency. Prioritization is given to the most consequential cases first, e.g. if there is a call for physical harm.

Another important principle we follow is to manage the content first (and not the creator). While content that falls within the definitions is removed, at the same time the creator is provided reasons for removal and an ability to appeal and ask for reinstatement. Tell us why something you wrote is ok, and we will listen and are happy to reverse.

If a user is flagged to be a repeat and persistent violator, the user is assisted with reminders. These reminders are sent at thresholds linked to frequency and severity of violations.

In extreme cases we implement our version of digital law enforcement (also similar to offline law enforcement). We may take steps to restrict visibility to others or of others; ability to notify others or receive notifications from others or restrict ability to comment, like or ReKoo.

However, users also have the ability to show that they are reformed and we will always welcome them back.

At Koo we believe that respectful communication and communities are based on everyday common sense rules that are already practised in the offline world. We will continue to ensure that speech remains free, respectful and meaningful for everyone.

To know more about us and our approach do click through to the following links.

Community Guidelines

Content Moderation Guidelines

Profile Actions Policy

Spam Policy, and

Reporting & Redressal