Anti Harmful Content

About Us


Supported by UNESCO and the European Union, the Anti Harmful Content project aims to create a better understanding of the dynamics and the impacts of harmful content spreading in Indonesia. We hope that the information provided on this website can help users to report and avoid potentially harmful content and promote more peacebuilding narratives and initiatives in Indonesia.

We will use the information submitted through this form for our research on the effectiveness of the current reporting mechanisms in eliminating harmful content on social media. The data will only be used within the research purposes of this project. We will keep your information confidential and we will anonymize the data when we use your submission in our research publication. Anonymized data sharing to a third party (i.e. government, social media companies, and non-governmental organizations) will be strictly limited to research and research-adjacent activities.
File types that can be uploaded: .jpg, .png Maximum size: 5 MB
We will use the information submitted through this form for our research on the effectiveness of the current reporting mechanisms in eliminating harmful content on social media. The data will only be used within the research purposes of this project. We will keep your information confidential and we will anonymize the data when we use your submission in our research publication. Anonymized data sharing to a third party (i.e. government, social media companies, and non-governmental organizations) will be strictly limited to research and research-adjacent activities.
File types that can be uploaded: .jpg, .png Maximum size: 5 MB

How Do I Report Harmful Content?

Data Analysis

We use sample data from Twitter to analyze harmful content on the platform. We collect tweets using queries related to hate speech or minority groups between November 1st, 2021 and December 31st, 2021. However, exceptions apply for some of the case studies. For tweets using #tolakLGBT and #daruratLGBT hashtags, we collect sample tweets that were posted in 2016. For Ahok's blasphemy case, we collect sample tweets from May 24, 2016 to April 27, 2018

Many tweets using #tolakLGBT and #daruratLGBT hashtags highlight the need to save the young generation from LGBTQIA+ influence as they consider it as a deviation. Many religious wordings and hashtags such as “Adam”, “Hawa” (Eve), “Allah”, “Islam”, and #IslamTolakLGBT are also frequently used in expressing their rejection towards LGBTQIA+, which signifies the important role of religion in shaping public opinion.
Our sample tweets reflect the proliferation of hate speech towards Shia and Ahmadiyya on social media. Approximately 11.5% of all sample tweets contain the word “sesat” (deviant), showing that many users consider Shia and Ahmadiyya as devious beliefs that cannot be grouped together with Islam.
Tweets containing neutral words are also more likely to have positive or neutral sentiment than tweets containing slurs. Approximately 23.79% of tweets with neutral words have positive sentiment, while only 15.85% of tweets with LGBTQIA+ slurs do so. Queer-related slurs are often used by Indonesians as a derogatory word, not only towards queer people, but also towards non-queer people. “Banci” and “bencong”, for example, are often used to denote the cowardice act of a man.
Both topics related to political disinformation are dominated by tweets with negative sentiment. Approximately 83.07% of the tweets in the “buzzer” dataset have negative sentiment. This percentage increases to more than 90% in the “Ahok” dataset. These numbers show that political discussion on social media tends to be hostile and offensive, especially when it comes to a polarizing political figure such as Ahok.

Publications