Get the latest on global AI regulations, legal risk, and safety-by-design strategies. Read the Report

About header image

ActiveFence Blog

Digital world globe loop blue version
10 min read

Mapping Internet Regulations: Online Hate Speech

10 min read

ActiveFence provides a searchable interactive guide to the legislation of over 70 countries that govern online hate speech content.
Gideon Freud
May 17, 2022
READ MORE
Digital world globe loop green version
9 min read

Mapping Internet Regulations: Online Terrorist Content

9 min read

ActiveFence provides a searchable interactive guide to the legislation of over 60 countries that govern online terrorist content.
Gideon Freud
Apr 12, 2022
READ MORE
From Our Research Desk Digital world in one place/ Digitalization
5 min read

Russia, Ukraine, and Information Warfare

5 min read

ActiveFence shares information warfare tactics being used to help teams secure their platforms from becoming weapons of war.
Ariella Rothschild
Mar 31, 2022
READ MORE
From Our Research Desk Miniature soldiers positioned on a map of Eastern Europe with a red gradient overlay.
6 min read

The Evolution of a Disinformation Trend

6 min read

ActiveFence shares how one false narrative evolved, creating a wide-spread trend of disinformation and misinformation.
Ariella Rothschild
Mar 08, 2022
READ MORE
two hands directed to the sun showing the sign of hope
3 min read

Child Safety Spotlight: INHOPE x ActiveFence

3 min read

ActiveFence and INHOPE partner to fight the spread of CSAM online and promote the mental wellbeing of digital first responders.
Ariella Rothschild
Jan 26, 2022
READ MORE
Frost and sullivan
2 min read

ActiveFence Honored with Trust & Safety Award

2 min read

ActiveFence has been awarded the Frost & Sullivan 2021 European Technology Innovation Leadership Award in the online trust and safety industry.
Ariella Rothschild
Jan 25, 2022
READ MORE
Storm near the sea
8 min read

All Eyes On 2022: Trust & Safety Challenges for the New Year

8 min read

Emerging threats, technological evolution and legislation revolution. As 2021 nears its end, ActiveFence identifies the challenges to be navigated by the Trust and Safety community in the coming year.
Gideon Freud
Dec 02, 2021
READ MORE
From Our Research Desk Reichstag during the sunset
5 min read

False Narratives, Elections, and the Quest to Secure German Democracy

5 min read

ActiveFence assesses the false narratives that have spread across German social media during the country’s 2021 federal elections.
Gideon Freud
Nov 15, 2021
READ MORE
ActiveFence staff alltogether
3 min read

ActiveFence Announces $100 Million in Funding to Keep the Internet Safe

3 min read

With a unique proactive approach, the company’s technology detects and protects against disinformation, child abuse, hate speech, terror, and other malicious content and activities online.
Liron Hakim Bobrov
Jul 27, 2021
READ MORE

Categories

Latest Posts

Cover image of ActiveFence report: Essential AI Red Teaming Tools and Techniques for Product Teams
AI Red Teaming: The New Discipline Every Product Team Needs
A young girl talks to an emotional support chatbot in a dark room.
When Comfort Turns Harmful: How Emotional Support Chatbots Enable Self-Harm
Aligning to OWASP LLM Top Ten
Aligning AI Safety and Security Policies with the OWASP LLM Top Ten

Algeria

Algeria passed new legislation in 2020 to make the dissemination of online discrimination or insult based on “gender, race, color, descent, national or ethnic origin, language, geographical affiliation, disability or state of health” a criminal act.

International Covenant on the Elimination of All Forms of Racial Discrimination (1965); Law 20.05 (2020)

Egypt

Egypt’s Supreme Council for Media Regulation can issue an order to block a website that hosts content that incites violence, discrimination, sectarianism, racism, or offends the state institutions.

International Covenant on the Elimination of All Forms of Racial Discrimination (1965); Media Regulation Law and the Supreme Council for Media Regulation promulgated by Law No. 180 (2018)

Ethiopia

Ethiopia holds internet intermediaries to be liable for hosted user-generated content. Platforms must remove hate speech based upon ethnicity, religion, race, gender or
disability within 24 hours of notification from the authorities.

International Covenant on the Elimination of All Forms of Racial Discrimination (1965); Hate Speech and Disinformation Prevention and Suppression Proclamation (2020)

Kenya

Kenya requires internet service providers to identify dangerous users and collect and provide information to assist law enforcement. Platforms are also required to remove content that constitutes hate speech and constitutes ethnic incitement on receipt of a court order within 24 hours.

International Covenant on the Elimination of All Forms of Racial Discrimination (1965); Computer Misuse and Cybercrimes Act (2018)

Mali

Mali criminalized the spreading of racist, xenophobic acts, threats and insults through an internet system. Internet service providers must also put in place mechanisms to monitor their platforms for illegal activities and communications.

Failure to inform the competent authorities of illegal activities risks a fine of CFA 500,000 to 2,000,000 (between $790 and $3,200).

International Covenant on the Elimination of All Forms of Racial Discrimination (1965); Suppression of Cybercrime (2019)

Morocco

In Morocco network providers are obliged to suppress, prohibit and restrict access to online content that is considered a threat to the kingdom, its security, and symbols within 24 hours of the content’s upload. Failure to act proactively will prompt the service provider to receive formal notice to remove the illegal content within 5 days. Failure to comply with this could result in a penalty of nearly €46,000 Euros as well as a temporary suspension from operating in Morocco. This draft law was suspended until the end of the COVID-19 crisis.

International Covenant on the Elimination of All Forms of Racial Discrimination (1965); Additional Protocols to the Convention on Cybercrime (2003)

Niger

Niger criminalized the creation, dissemination, and making available ideas or theories, of a racist, regionalist, ethnic, religious, or xenophobic nature. This covers all communication from text to audio, image, and video that is shared on an information system. Platform operators may be required to intercept communications on request from relevant authorities with regard to illegal online activity.

International Covenant on the Elimination of All Forms of Racial Discrimination (1965); Additional Protocols to the Convention on Cybercrime (2003); Cybercrime Law (2020)

Nigeria

International Covenant on the Elimination of All Forms of Racial Discrimination (1965); Additional Protocols to the Convention on Cybercrime (2003)

South Africa

South Africa requires platform operators to identify and remove user-generated content that incites violence due to any group characteristic on receipt of a court order.

International Covenant on the Elimination of All Forms of Racial Discrimination (1965); Promotion of Equality and Prevention of Unfair Discrimination Act (2000); Films Publication Act (1965); FBP Online Regulation Policy (2016)

Uganda

Platforms are required to fulfill the “duties of a licensee and producer,” and ensure that hosted content does not promote ethnic prejudice and violence and is not likely to create public insecurity or violence. Failure to comply risks fines of up to 10% of a platform’s gross annual revenue, and a suspension or revocation of a license.

International Covenant on the Elimination of All Forms of Racial Discrimination (1965); Uganda Communications Act (2013); The Computer Misuse Act (2011)

Argentina

Argentina gave accession to the 2003 Additional Protocols to the Convention on Cybercrime making it illegal to publish xenophobic and racist incitement. The country requires platform operators to take preventative measures against illegal acts motivated by racism, xenophobia, and other types of intolerance and hatred.

International Covenant on the Elimination of All Forms of Racial Discrimination (1965); Additional Protocols to the Convention on Cybercrime (2003); Law No.23.592: Discriminatory Acts; Law No. 26.522: Law of Audiovisual Communication

Brazil

Brazil criminalizes hate speech based on race, color, ethnicity, religion, or national origin. The country’s approach to internet regulation is similar to that of the US, with platforms exempted from liability for illegal or harmful content on their platforms of which they were unaware. However, the country gave accession to the 2003 Additional Protocols to the Convention on Cybercrime making it illegal to publish xenophobic and racist incitement.

International Covenant on the Elimination of All Forms of Racial Discrimination (1965); Additional Protocols to the Convention on Cybercrime (2003); Marco Civil Law of the Internet (2014); Law 7.716 (1989)

Chile

Chile gave accession to the 2003 Additional Protocols to the Convention on Cybercrime making it illegal to publish xenophobic and racist incitement. The country criminalizes online harassment motivated by ​​race, religion, nationality, ethnicity, or cultural origin. Liability for illegal content rests with the user rather than the service provider.

International Covenant on the Elimination of All Forms of Racial Discrimination (1965); Additional Protocols to the Convention on Cybercrime (2003); Statute on Freedom of Opinion and Information and the Performance of Journalism (2001)

Colombia

Colombia gave accession to the 2003 Additional Protocols to the Convention on Cybercrime making it illegal to publish xenophobic and racist incitement. The country criminalizes online harassment motivated by ​​race, religion, sex, sexual orientation, or disability. Internet Service Providers are required to monitor for illegal activity.

International Covenant on the Elimination of All Forms of Racial Discrimination (1965); Additional Protocols to the Convention on Cybercrime (2003); Colombian Criminal Code Article 134

Mexico

Mexico gave accession to the 2003 Additional Protocols to the Convention on Cybercrime making it illegal to publish xenophobic and racist incitement. The country criminalizes online harassment motivated by ethnicity, nationality, sex, age, disability, health, pregnancy, language, religion, or sexuality. Liability for illegal content rests with the user rather than the service provider.

International Covenant on the Elimination of All Forms of Racial Discrimination (1965); Additional Protocols to the Convention on Cybercrime (2003); Federal Law to Prevent and Eliminate Discrimination (2003)

Nicaragua

Nicaragua criminalizes online harassment motivated by race, nationality, sex, or sexual orientation. Liability for illegal content rests with the user rather than the service provider.

International Covenant on the Elimination of All Forms of Racial Discrimination (1965); Criminal Code (2008)

Afghanistan

While the Islamic Republic of Afghanistan was party to international UN treaties, its successor state the Islamic Emirate of Afghanistan is not bound by these agreements.

International Covenant on the Elimination of All Forms of Racial Discrimination (1965)

China

China requires companies responsible for the information published on their platforms. They must also operate content management mechanisms, have clear rules on content governance and response systems to detect ‘rumors’; and receive user complaints and reports. Failure to comply carries civil and criminal liabilities for operators. Companies must also provide law enforcement with information to identify users, as well as encryption keys. Any organization using networks shall abide by the Constitution, observe public order, and respect social morality; they must use the internet to engage in activities endangering national security, national honor, and national interests; they must not incite subversion of national sovereignty, overturn the socialist system, incite separatism, … advocate ethnic hatred and ethnic discrimination.”

International Covenant on the Elimination of All Forms of Racial Discrimination (1965); Cyber Security Law (2017); The Provisions on the Governance of Network Information Content Ecology (2015)

Spain

Spain defines public incitement to violence and racist and xenophobic hate speech as illegal. This is specifically directed towards a person or group of persons by reference to race, skin color, nationality, ethnicity, religion, or sexual orientation. Signatories to the EU Code of Conduct on countering illegal hate speech online must remove ‘illegal hate speech’ within 24 hours on receipt of a removal notice from Spanish authorities.

International Covenant on the Elimination of All Forms of Racial Discrimination (1965); Additional Protocols to the Convention on Cybercrime (2003); EU Framework Decision 2008/913/JHA (2008); EU Code of Conduct on countering illegal hate speech online (2016)

Albania

Albania has aligned its legislation with the EU. It has criminalized the sharing of threats online based on immutable characteristics including race, gender, sexuality, and discrimination. Anti-discrimination Commissioner can order content to be removed with the power to give fines for non-compliance.

International Covenant on the Elimination of All Forms of Racial Discrimination (1965); Additional Protocols to the Convention on Cybercrime (2003); Law Against Discrimination (2010)

Iceland

Iceland has criminalized the dissemination of incitement to violence, hatred, or discrimination against a person or group of persons due to their group characteristics.

International Covenant on the Elimination of All Forms of Racial Discrimination (1965); Additional Protocols to the Convention on Cybercrime (2003)

North Macedonia

North Macedonia has criminalized the dissemination of incitement to violence, hatred, or discrimination against a person or group of persons due to their group characteristics.

International Covenant on the Elimination of All Forms of Racial Discrimination (1965); Additional Protocols to the Convention on Cybercrime (2003)

Norway

Norway defines hate speech as speech that threatens or insults a person or incites hatred, persecution, or contempt for someone because of their skin color, national or ethnic origin, religion, sexual orientation, or disability.

International Covenant on the Elimination of All Forms of Racial Discrimination (1965); Additional Protocols to the Convention on Cybercrime (2003)

Switzerland

Switzerland has criminalized the dissemination of incitement to violence, hatred, or discrimination against a person or group of persons due to their group characteristics.

International Covenant on the Elimination of All Forms of Racial Discrimination (1965); Additional Protocols to the Convention on Cybercrime (2003)

Austria

Austria defines illegal hate speech as racism or xenophobia directed towards a group defined by race, color, religion, descent, or national or ethnic origin. Platforms operating within Austria must remove illegal hate crimes on social networks within 24 hours of notification (with a maximum of 7 days for complex cases).

Fines for non-compliance range between €10,000 and €58,000.

International Covenant on the Elimination of All Forms of Racial Discrimination (1965); Additional Protocols to the Convention on Cybercrime (2003); EU Framework Decision 2008/913/JHA (2008); EU Code of Conduct on countering illegal hate speech online (2016); Communications Platform Act (2021)

Belgium

Belgium defines racist and xenophobic hate speech as illegal. This is specifically directed towards a person or group of persons by reference to race, color, religion, descent, or national or ethnic origin. Signatories to the EU Code of Conduct on countering illegal hate speech online must remove ‘illegal hate speech’ within 24 hours on receipt of a removal notice from Belgian authorities.

International Covenant on the Elimination of All Forms of Racial Discrimination (1965); Additional Protocols to the Convention on Cybercrime (2003); EU Framework Decision 2008/913/JHA (2008); EU Code of Conduct on countering illegal hate speech online (2016)

Bulgaria

Bulgaria defines racist and xenophobic hate speech as illegal. This is specifically directed towards a person or group of persons by reference to race, color, religion, descent, or national or ethnic origin. Signatories to the EU Code of Conduct on countering illegal hate speech online must remove ‘illegal hate speech’ within 24 hours on receipt of a removal notice from Belgian authorities.

International Covenant on the Elimination of All Forms of Racial Discrimination (1965); EU Code of Conduct on countering illegal hate speech online (2016)

Slovakia

Slovakia defines public incitement to violence and racist and xenophobic hate speech as illegal. This is specifically directed towards a person or group of persons by reference to race, skin color, nationality, ethnicity, religion, or sexual orientation. Signatories to the EU Code of Conduct on countering illegal hate speech online must remove ‘illegal hate speech’ within 24 hours on receipt of a removal notice from Slovakian authorities.

International Covenant on the Elimination of All Forms of Racial Discrimination (1965); Additional Protocols to the Convention on Cybercrime (2003); EU Framework Decision 2008/913/JHA (2008); EU Code of Conduct on countering illegal hate speech online (2016)

Singapore

Singapore prohibits the publication of communication that “incites feelings of enmity, hatred, or ill-will between different groups of persons.” Internet intermediaries must block content on receipt of a notiice. Failure to comply risks daily fines of up to $20,000 for platforms in or outside of the country.

The law created a maximum penalty of $500,000 per infraction.

International Covenant on the Elimination of All Forms of Racial Discrimination (1965); Protection from Online Falsehoods and Manipulation Act (2019)

Germany

Germany defines illegal hate speech as racism or xenophobia directed towards a group defined by race, color, religion, descent, or national or ethnic origin. Social media and video sharing platforms operating within Germany must remove illegal content within 24 hours of notification. They must also produce bi-annual transparency reports on their moderation process. Fines for non-compliance range up to €50 million.

International Covenant on the Elimination of All Forms of Racial Discrimination (1965); Additional Protocols to the Convention on Cybercrime (2003); EU Framework Decision 2008/913/JHA (2008); EU Code of Conduct on countering illegal hate speech online (2016); Network Enforcement Act (2017, 2019)

Croatia

Croatia defines racist and xenophobic hate speech as illegal. This is specifically directed towards a person or group of persons by reference to race, color, religion, descent, or national or ethnic origin. Signatories to the EU Code of Conduct on countering illegal hate speech online must remove ‘illegal hate speech’ within 24 hours on receipt of a removal notice from Croatian authorities.

International Covenant on the Elimination of All Forms of Racial Discrimination (1965); Additional Protocols to the Convention on Cybercrime (2003); EU Framework Decision 2008/913/JHA (2008); EU Code of Conduct on countering illegal hate speech online (2016)

 

Sweden

Sweden defines public incitement to violence and racist and xenophobic hate speech as illegal. This is specifically directed towards a person or group of persons by reference to race, skin color, nationality, ethnicity, religion, or sexual orientation. Signatories to the EU Code of Conduct on countering illegal hate speech online must remove ‘illegal hate speech’ within 24 hours on receipt of a removal notice from Sweden authorities.

International Covenant on the Elimination of All Forms of Racial Discrimination (1965); Additional Protocols to the Convention on Cybercrime (2003); EU Framework Decision 2008/913/JHA (2008); EU Code of Conduct on countering illegal hate speech online (2016)

Cyprus

Cyprus defines racist and xenophobic hate speech as illegal. This is specifically directed towards a person or group of persons by reference to race, color, religion, descent, or national or ethnic origin. Signatories to the EU Code of Conduct on countering illegal hate speech online must remove ‘illegal hate speech’ within 24 hours on receipt of a removal notice from Cypriot authorities.

International Covenant on the Elimination of All Forms of Racial Discrimination (1965); Additional Protocols to the Convention on Cybercrime (2003); EU Framework Decision 2008/913/JHA (2008); EU Code of Conduct on countering illegal hate speech online (2016)

Czechia

Czechia defines racist and xenophobic hate speech as illegal. This is specifically directed towards a person or group of persons by reference to race, color, religion, descent, or national or ethnic origin. Signatories to the EU Code of Conduct on countering illegal hate speech online must remove ‘illegal hate speech’ within 24 hours on receipt of a removal notice from Czech authorities.

International Covenant on the Elimination of All Forms of Racial Discrimination (1965); Additional Protocols to the Convention on Cybercrime (2003); EU Framework Decision 2008/913/JHA (2008); EU Code of Conduct on countering illegal hate speech online (2016)

Indonesia

Indonesia requires social media companies to ensure that they do not host or facilitate the distribution of “restricted content”, which includes hate speech and blasphemy. Social media companies that do not comply risk a fine of between c.€580 and €2800 Euros as well as administrative sanctions and suspending their networks. Platforms must remove prohibited content within 24 hours of a government notice and must monitor for prohibited content.

International Covenant on the Elimination of All Forms of Racial Discrimination (1965); Digital Platform Law (PP PSTE no. 71/2019 on the Government Regulation on the Implementation of Electronic and Transaction System (2019); Ministerial Regulation 5 (2021)

Slovenia

Slovenia defines public incitement to violence and racist and xenophobic hate speech as illegal. This is specifically directed towards a person or group of persons by reference to race, skin color, nationality, ethnicity, religion, or sexual orientation. Signatories to the EU Code of Conduct on countering illegal hate speech online must remove ‘illegal hate speech’ within 24 hours on receipt of a removal notice from Slovenian authorities.

International Covenant on the Elimination of All Forms of Racial Discrimination (1965); Additional Protocols to the Convention on Cybercrime (2003); EU Framework Decision 2008/913/JHA (2008); EU Code of Conduct on countering illegal hate speech online (2016)

Vietnam

Vietnam prohibits the sharing of ‘anti-Vietnamese’ propaganda, which causes divisions between Vietnamese ethnic groups, religions, and people of all countries. Internet companies who are at risk of hosting such content are required to implement technical measures to prevent, detect, and remove such material on request from the relevant authority. If they do not take steps they can be held liable for its contents.

International Covenant on the Elimination of All Forms of Racial Discrimination (1965); Cybersecurity Law 2018

 

Denmark

Denmark defines racist and xenophobic hate speech as illegal. This is specifically directed towards a person or group of persons by reference to race, skin color, nationality, ethnicity, religion, or sexual orientation. Signatories to the EU Code of Conduct on countering illegal hate speech online must remove ‘illegal hate speech’ within 24 hours on receipt of a removal notice from Danish authorities.

International Covenant on the Elimination of All Forms of Racial Discrimination (1965); Additional Protocols to the Convention on Cybercrime (2003); EU Framework Decision 2008/913/JHA (2008); EU Code of Conduct on countering illegal hate speech online (2016); Criminal Code (2005)

Australia

Australia regards it to be a criminal offense to use the internet in a menacing, harassing or offensive manner. In addition, cyber-abuse material directed at either adults or children must be removed within 48 hours of a court order.

International Covenant on the Elimination of All Forms of Racial Discrimination (1965); Additional Protocols to the Convention on Cybercrime (2003); Australia: Section 474.17 of the Criminal Code; The Online Safety Act (2021)

Estonia

Despite being subject to EU law, Estonia has no national laws against hate speech which is not defined in the country. However, under EU law hate speech is defined as racist and xenophobic hate speech and is illegal. This is specifically directed towards a person or group of persons by reference to race, color, religion, descent, or national or ethnic origin. Signatories to the EU Code of Conduct on countering illegal hate speech online must remove ‘illegal hate speech’ within 24 hours on receipt of a removal notice from authorities.

International Covenant on the Elimination of All Forms of Racial Discrimination (1965); Additional Protocols to the Convention on Cybercrime (2003); EU Framework Decision 2008/913/JHA (2008); EU Code of Conduct on countering illegal hate speech online (2016)

Finland

Finland does not have a legal definition of hate speech. However, it defines ethnic agitation as illegal. However, under EU law, hate speech is defined as insulting or threatening speech directed towards a person or group of persons by reference to race, skin color, nationality, ethnicity, religion, or sexual orientation. Signatories to the EU Code of Conduct on countering illegal hate speech online must remove ‘illegal hate speech’ within 24 hours on receipt of a removal notice from Finnish authorities.

International Covenant on the Elimination of All Forms of Racial Discrimination (1965); Additional Protocols to the Convention on Cybercrime (2003); EU Framework Decision 2008/913/JHA (2008); EU Code of Conduct on countering illegal hate speech online (2016)

France

France defines public incitement to violence and racist and xenophobic hate speech as illegal. This is specifically directed towards a person or group of persons by reference to race, skin color, nationality, ethnicity, religion, or sexual orientation. Signatories to the EU Code of Conduct on countering illegal hate speech online must remove ‘illegal hate speech’ within 24 hours on receipt of a removal notice from French authorities. French internet platforms must create simple mechanisms to report hate speech content online.

International Covenant on the Elimination of All Forms of Racial Discrimination (1965); Additional Protocols to the Convention on Cybercrime (2003); EU Framework Decision 2008/913/JHA (2008); EU Code of Conduct on countering illegal hate speech online (2016); Loi d’Avia (2020)

Greece

Greece defines public incitement to violence and racist and xenophobic hate speech as illegal. This is specifically directed towards a person or group of persons by reference to race, skin color, nationality, ethnicity, religion, or sexual orientation. Signatories to the EU Code of Conduct on countering illegal hate speech online must remove ‘illegal hate speech’ within 24 hours on receipt of a removal notice from Greek authorities.

International Covenant on the Elimination of All Forms of Racial Discrimination (1965); Additional Protocols to the Convention on Cybercrime (2003); EU Framework Decision 2008/913/JHA (2008); EU Code of Conduct on countering illegal hate speech online (2016)

Hungary

Hungary defines public incitement to violence and racist and xenophobic hate speech as illegal. This is specifically directed towards a person or group of persons by reference to race, skin color, nationality, ethnicity, religion, or sexual orientation. Signatories to the EU Code of Conduct on countering illegal hate speech online must remove ‘illegal hate speech’ within 24 hours on receipt of a removal notice from Hungarian authorities. Hungary’s civil courts hold internet intermediaries liable if they enable the publication of third-party content on their sites which is illegal, this liability is applied whether or not the platform operator is aware of the content.

International Covenant on the Elimination of All Forms of Racial Discrimination (1965); EU Framework Decision 2008/913/JHA (2008); EU Code of Conduct on countering illegal hate speech online (2016)

Romania

Romania defines public incitement to violence and racist and xenophobic hate speech as illegal. This is specifically directed towards a person or group of persons by reference to race, skin color, nationality, ethnicity, religion, or sexual orientation. Signatories to the EU Code of Conduct on countering illegal hate speech online must remove ‘illegal hate speech’ within 24 hours on receipt of a removal notice from Romanian authorities.

International Covenant on the Elimination of All Forms of Racial Discrimination (1965); Additional Protocols to the Convention on Cybercrime (2003); EU Framework Decision 2008/913/JHA (2008); EU Code of Conduct on countering illegal hate speech online (2016)

Netherlands

The Netherlands defines public incitement to violence and racist and xenophobic hate speech as illegal. This is specifically directed towards a person or group of persons by reference to race, skin color, nationality, ethnicity, religion, or sexual orientation. Signatories to the EU Code of Conduct on countering illegal hate speech online must remove ‘illegal hate speech’ within 24 hours on receipt of a removal notice from Dutch authorities.

International Covenant on the Elimination of All Forms of Racial Discrimination (1965); Additional Protocols to the Convention on Cybercrime (2003); EU Framework Decision 2008/913/JHA (2008); EU Code of Conduct on countering illegal hate speech online (2016)

Poland

Poland defines public incitement to violence and racist and xenophobic hate speech as illegal. This is specifically directed towards a person or group of persons by reference to race, skin color, nationality, ethnicity, religion, or sexual orientation. Signatories to the EU Code of Conduct on countering illegal hate speech online must remove ‘illegal hate speech’ within 24 hours on receipt of a removal notice from Polish authorities.

International Covenant on the Elimination of All Forms of Racial Discrimination (1965); Additional Protocols to the Convention on Cybercrime (2003); EU Framework Decision 2008/913/JHA (2008); EU Code of Conduct on countering illegal hate speech online (2016)

New Zealand

New Zealand prohibits online communications that share private information, threaten, intimidate, are grossly offensive, or are harassing in nature. In addition, it prohibits content that incites or denigrates based on color, race, religion, sexual orientation, or disability. Failure to remove harmful content carries a fine of up to NZ$20,000.

International Covenant on the Elimination of All Forms of Racial Discrimination (1965); Additional Protocols to the Convention on Cybercrime (2003); Harmful Digital Communications Act 2015

Ireland

Ireland defines public incitement to violence and racist and xenophobic hate speech as illegal. This is specifically directed towards a person or group of persons by reference to race, skin color, nationality, ethnicity, religion, or sexual orientation. Signatories to the EU Code of Conduct on countering illegal hate speech online must remove ‘illegal hate speech’ within 24 hours on receipt of a removal notice from Irish authorities.

International Covenant on the Elimination of All Forms of Racial Discrimination (1965); EU Framework Decision 2008/913/JHA (2008); EU Code of Conduct on countering illegal hate speech online (2016)

Italy

Italy defines public incitement to violence and racist and xenophobic hate speech as illegal. This is specifically directed towards a person or group of persons by reference to race, skin color, nationality, ethnicity, religion, or sexual orientation. Signatories to the EU Code of Conduct on countering illegal hate speech online must remove ‘illegal hate speech’ within 24 hours on receipt of a removal notice from Italian authorities. Video-sharing platforms are, requested to implement effective systems for detecting and reporting both offenses and online perpetrators, and should produce quarterly reports on the monitoring and identification of online hate content.

Should platforms not comply with the regulation or subsequent warnings, the Authority may apply fines of between €10,300 to €258,000.

International Covenant on the Elimination of All Forms of Racial Discrimination (1965); Additional Protocols to the Convention on Cybercrime (2003); EU Framework Decision 2008/913/JHA (2008); EU Code of Conduct on countering illegal hate speech online (2016); Regulation on respect for human dignity and the principle of non-discrimination in relation to hate speech (2019)

Malta

Malta defines public incitement to violence and racist and xenophobic hate speech as illegal. This is specifically directed towards a person or group of persons by reference to race, skin color, nationality, ethnicity, religion, or sexual orientation. Signatories to the EU Code of Conduct on countering illegal hate speech online must remove ‘illegal hate speech’ within 24 hours on receipt of a removal notice from Maltese authorities.

International Covenant on the Elimination of All Forms of Racial Discrimination (1965); Additional Protocols to the Convention on Cybercrime (2003); EU Framework Decision 2008/913/JHA (2008); EU Code of Conduct on countering illegal hate speech online (2016)

Pakistan

Pakistan requires platforms to remove, suspend, or disable access to violative accounts and content within 24 hours of the government’s notification. This includes accounts that spread hate speech or other content that might negatively affect the religious, cultural, ethnic, or national security sensitivities of Pakistan

International Covenant on the Elimination of All Forms of Racial Discrimination (1965); Citizen’s Protection (Against Online Harm) Rules 2020

United States

The First Amendment guarantees the right to freedom of speech and US platforms hold a liability exemption for user-generated content and shields platforms from a legal obligation to remove hate speech content.

International Covenant on the Elimination of All Forms of Racial Discrimination (1965); Section 230 of the Communication Decency Act (1996)

Canada

Canadian Digital Charter 2019 stipulates that the online space should be “free from hate and violent extremism.” Section 320.1 of the Canadian Criminal Code to order that publicly available online ‘hate propaganda’ be taken down and deleted by platform operators.

International Covenant on the Elimination of All Forms of Racial Discrimination (1965); Canadian Criminal Code Section 320.1

Bangladesh

Bangladesh criminalized the use of digital media to intimidate people or cause damage to the state they face a 14-year jail sentence. If someone uses a website to intimidate anyone they face a 3-year jail sentence. The Bangladesh Telecommunications and Regulatory Commission may block any online information that harms religious value or incites racial violence. Service providers do not have liability if they act against such content.

International Covenant on the Elimination of All Forms of Racial Discrimination (1965); Digital Security Act (2018)

Portugal

Portgual defines public incitement to violence and racist and xenophobic hate speech as illegal. This is specifically directed towards a person or group of persons by reference to race, skin color, nationality, ethnicity, religion, or sexual orientation. Signatories to the EU Code of Conduct on countering illegal hate speech online must remove ‘illegal hate speech’ within 24 hours on receipt of a removal notice from Portuguese authorities.

International Covenant on the Elimination of All Forms of Racial Discrimination (1965); Additional Protocols to the Convention on Cybercrime (2003); EU Framework Decision 2008/913/JHA (2008); EU Code of Conduct on countering illegal hate speech online (2016)

International Covenant on the Elimination of All Forms of Racial Discrimination (1965); Additional Protocols to the Convention on Cybercrime (2003); EU Framework Decision 2008/913/JHA (2008); EU Code of Conduct on countering illegal hate speech online (2016)

Lithuana

Lithuania defines public incitement to violence and racist and xenophobic hate speech as illegal. This is specifically directed towards a person or group of persons by reference to race, skin color, nationality, ethnicity, religion, or sexual orientation. Signatories to the EU Code of Conduct on countering illegal hate speech online must remove ‘illegal hate speech’ within 24 hours on receipt of a removal notice from Lithuania authorities.

International Covenant on the Elimination of All Forms of Racial Discrimination (1965); Additional Protocols to the Convention on Cybercrime (2003); EU Framework Decision 2008/913/JHA (2008); EU Code of Conduct on countering illegal hate speech online (2016)

Luxembourg

Luxembourg defines public incitement to violence and racist and xenophobic hate speech as illegal. This is specifically directed towards a person or group of persons by reference to race, skin color, nationality, ethnicity, religion, or sexual orientation. Signatories to the EU Code of Conduct on countering illegal hate speech online must remove ‘illegal hate speech’ within 24 hours on receipt of a removal notice from Luxembourg authorities.

International Covenant on the Elimination of All Forms of Racial Discrimination (1965); Additional Protocols to the Convention on Cybercrime (2003); EU Framework Decision 2008/913/JHA (2008); EU Code of Conduct on countering illegal hate speech online (2016)

India

India requires social media companies to remove material that is “libelous, racially or ethnically objectionable… is patently false and untrue, and is written or published in any form, with the intent to mislead or harass a person, entity or agency for financial gain or to cause any injury to any person.” This content action must be taken within 36 hours of receipt of notification and platforms may need to provide identifiable information about those users uploading this material on request.

International Covenant on the Elimination of All Forms of Racial Discrimination (1965); The Republic of India’s Information Technology (Intermediary Guidelines and Digital Media Ethics Code) Rules (2021)

United Kingdom

The UK criminalizes the sending of grossly offensive, indecent, or menacing communications online. Providers of video-sharing platforms must take steps to prevent harmful content as set out in Schedule 15A and must operate an out-of-court resolution service for disputes.

Failure to operate an approved mechanism carries a fine of up to £250,000.

International Covenant on the Elimination of All Forms of Racial Discrimination (1965); Communications Acts 2003

Armenia

Armenia prohibits the use of computer systems to make threats of violence against a person or group of persons. Platforms that permit this risk their services being blocked or a fine of ֏5 million ($11,100).

International Covenant on the Elimination of All Forms of Racial Discrimination (1965); Additional Protocols to the Convention on Cybercrime (2003); Law on Electronic Communications (2005)

Belarus

Belarus criminalized content that aims to incite based on race, national, religious, linguistic, or other social affiliation, with prison terms established for persons who share such prohibited content publicly.

International Covenant on the Elimination of All Forms of Racial Discrimination (1965); Article 130 Belarussian Criminal Code (1999, 2019)

Georgia

Georgia has criminalized the dissemination of incitement to violence, hatred, or discrimination against a person or group of persons due to their group characteristics.

International Covenant on the Elimination of All Forms of Racial Discrimination (1965); Additional Protocols to the Convention on Cybercrime (2003)

Russia

Russia requires that internet service providers will restrict access to websites containing calls for extremist activities. Failure to comply with legal entities risks fines of between ₽3,000,000 and ₽8,000,000. The Russian regulator Roskomnadzor can request the courts to block platforms that host content that incites national, class, social, or religious intolerance.

International Covenant on the Elimination of All Forms of Racial Discrimination (1965); Law on Mass Media 1991; Federal law N511-FZ (2020)

Turkey

Turkey criminalized the provocation of a group of people belonging to a different social class, religion, race, or sect. Social media companies must remove unlawful content and respond within 48 hours of a judicial order to remove prohibited material, or a user complaint.

Failure to comply risks fines of up to $1.5 million.

International Covenant on the Elimination of All Forms of Racial Discrimination (1965); Additional Protocols to the Convention on Cybercrime (2003); Law No. 5651 on The Regulation of Publications made in the Internet Environment and Combatting Crimes Committed through these Publications (2020)

Nepal

Nepal expanded the applicability of its National Broadcasting Rules to apply to internet platforms hosting video content. This prohibits the broadcast of material that could create “enmity or social mistrust (ill feeling) amongst the people of various caste, tribe, religion, race, region, community.”

International Covenant on the Elimination of All Forms of Racial Discrimination (1965); National Broadcasting Rules (1995, 2022 Amendment)

Sri Lanka

Sri Lanka does not have a specific law that regulates online hate speech, although it did give accession to the CoE’s Convention on Cybercrime. Sri Lanka instead relies upon the Telecommunication Regulatory Commission (TRCS) which orders internet service providers to block access to specific domains for hosting destabilizing content.

International Covenant on the Elimination of All Forms of Racial Discrimination (1965); Additional Protocols to the Convention on Cybercrime (2003)

Philippines

The Philipines does not have a specific law that regulates online hate speech, although it did give accession to the CoE’s Convention on Cybercrime.

International Covenant on the Elimination of All Forms of Racial Discrimination (1965); Additional Protocols to the Convention on Cybercrime (2003)

Austria

Internet platforms must remove terrorist content within 24 hours of notification (7 days in more complex cases), and one hour after a formal removal notice. They must determine which measures to put in place to reduce their exposure to terrorist content while maintaining the Right to Information.

Fines for non-compliance range between €10,000 and €58,000.

Audiovisual Media Services Directive 2010; Regulation of the European Parliament and of the Council dissemination of terrorist content online 2021; Communications Platform Act (January 1, 2021)

Bulgaria

Internet platforms must remove terrorist content one hour after the receipt of a removal notice, they must determine which measures to put in place to reduce their exposure to terrorist content while maintaining the Right to Information.

Audiovisual Media Services Directive 2010; Regulation of the European Parliament and of the Council dissemination of terrorist content online 2021

Croatia

Internet platforms must remove terrorist content one hour after the receipt of a removal notice, they must determine which measures to put in place to reduce their exposure to terrorist content while maintaining the Right to Information.

Audiovisual Media Services Directive 2010; Regulation of the European Parliament and of the Council dissemination of terrorist content online 2021

Cyprus

Internet platforms must remove terrorist content one hour after the receipt of a removal notice, they must determine which measures to put in place to reduce their exposure to terrorist content while maintaining the Right to Information.

Audiovisual Media Services Directive 2010; Regulation of the European Parliament and of the Council dissemination of terrorist content online 2021

Czechia

Internet platforms must remove terrorist content one hour after the receipt of a removal notice, they must determine which measures to put in place to reduce their exposure to terrorist content while maintaining the Right to Information.

Audiovisual Media Services Directive 2010; Regulation of the European Parliament and of the Council dissemination of terrorist content online 2021

Denmark

Internet platforms must remove terrorist content one hour after the receipt of a removal notice, they must determine which measures to put in place to reduce their exposure to terrorist content while maintaining the Right to Information.

Audiovisual Media Services Directive 2010; Regulation of the European Parliament and of the Council dissemination of terrorist content online 2021

Estonia

Internet platforms must remove terrorist content one hour after the receipt of a removal notice, they must determine which measures to put in place to reduce their exposure to terrorist content while maintaining the Right to Information.

Audiovisual Media Services Directive 2010; Regulation of the European Parliament and of the Council dissemination of terrorist content online 2021

Finland

Internet platforms must remove terrorist content one hour after the receipt of a removal notice, they must determine which measures to put in place to reduce their exposure to terrorist content while maintaining the Right to Information.

Audiovisual Media Services Directive 2010; Regulation of the European Parliament and of the Council dissemination of terrorist content online 2021

France

Audiovisual Media Services Directive 2010; Regulation of the European Parliament and of the Council dissemination of terrorist content online 2021; Law on strengthening the provisions relating to the fight against terrorism 2014; Respect for the Principles of the Republic 2021 2021 Amendment

Germany

Failure to comply carries fines of up to EUR 50M.

Audiovisual Media Services Directive; Regulation of the European Parliament and of the Council dissemination of terrorist content online 2021; Network Enforcement Act 2017; 2020 amendment

Greece

Internet platforms must remove terrorist content one hour after the receipt of a removal notice, they must determine which measures to put in place to reduce their exposure to terrorist content while maintaining the Right to Information.

Audiovisual Media Services Directive 2010; Regulation of the European Parliament and of the Council dissemination of terrorist content online 2021

Hungary

Internet platforms must remove terrorist content one hour after the receipt of a removal notice, they must determine which measures to put in place to reduce their exposure to terrorist content while maintaining the Right to Information.

Audiovisual Media Services Directive 2010; Regulation of the European Parliament and of the Council dissemination of terrorist content online 2021

Ireland

Internet platforms must remove terrorist content one hour after the receipt of a removal notice, they must determine which measures to put in place to reduce their exposure to terrorist content while maintaining the Right to Information.

Audiovisual Media Services Directive 2010; Regulation of the European Parliament and of the Council dissemination of terrorist content online 2021

Italy

Audiovisual Media Services Directive 2010; Regulation of the European Parliament and of the Council dissemination of terrorist content online 2021

Article 2 of Decree Law 7/2015

Latvia

Internet platforms must remove terrorist content one hour after receiving a removal notice, and they must determine which measures to put in place to reduce their exposure to terrorist content while maintaining the Right to Information.

Audiovisual Media Services Directive 2010; Regulation of the European Parliament and of the Council dissemination of terrorist content online 2021

Lithuana

Internet platforms must remove terrorist content one hour after receiving a removal notice, and they must determine which measures to put in place to reduce their exposure to terrorist content while maintaining the Right to Information.

Audiovisual Media Services Directive 2010; Regulation of the European Parliament and of the Council dissemination of terrorist content online 2021

Luxembourg

Internet platforms must remove terrorist content one hour after receiving a removal notice, and they must determine which measures to put in place to reduce their exposure to terrorist content while maintaining the Right to Information.

Audiovisual Media Services Directive 2010; Regulation of the European Parliament and of the Council dissemination of terrorist content online 2021

Malta

Internet platforms must remove terrorist content one hour after receiving a removal notice, and they must determine which measures to put in place to reduce their exposure to terrorist content while maintaining the Right to Information.

Audiovisual Media Services Directive 2010; Regulation of the European Parliament and of the Council dissemination of terrorist content online 2021

Netherlands

Internet platforms must remove terrorist content one hour after receiving a removal notice, and they must determine which measures to put in place to reduce their exposure to terrorist content while maintaining the Right to Information.

Audiovisual Media Services Directive 2010; Regulation of the European Parliament and of the Council dissemination of terrorist content online 2021

Poland

Internet platforms must remove terrorist content one hour after receiving a removal notice, and they must determine which measures to put in place to reduce their exposure to terrorist content while maintaining the Right to Information.

Audiovisual Media Services Directive 2010; Regulation of the European Parliament and of the Council dissemination of terrorist content online 2021

Portugal

Internet platforms must remove terrorist content one hour after receiving a removal notice, and they must determine which measures to put in place to reduce their exposure to terrorist content while maintaining the Right to Information.

Audiovisual Media Services Directive 2010; Regulation of the European Parliament and of the Council dissemination of terrorist content online 2021

Romania

Internet platforms must remove terrorist content one hour after receiving a removal notice, and they must determine which measures to put in place to reduce their exposure to terrorist content while maintaining the Right to Information.

Audiovisual Media Services Directive 2010; Regulation of the European Parliament and of the Council dissemination of terrorist content online 2021

Slovakia

Internet platforms must remove terrorist content one hour after receiving a removal notice, and they must determine which measures to put in place to reduce their exposure to terrorist content while maintaining the Right to Information.

Audiovisual Media Services Directive 2010; Regulation of the European Parliament and of the Council dissemination of terrorist content online 2021

Slovenia

Internet platforms must remove terrorist content one hour after receiving a removal notice, and they must determine which measures to put in place to reduce their exposure to terrorist content while maintaining the Right to Information.

Audiovisual Media Services Directive 2010; Regulation of the European Parliament and of the Council dissemination of terrorist content online 2021

Spain

Internet platforms must remove terrorist content one hour after receiving a removal notice, and they must determine which measures to put in place to reduce their exposure to terrorist content while maintaining the Right to Information.

Audiovisual Media Services Directive 2010; Regulation of the European Parliament and of the Council dissemination of terrorist content online 2021

Sweden

Internet platforms must remove terrorist content one hour after receiving a removal notice, and they must determine which measures to put in place to reduce their exposure to terrorist content while maintaining the Right to Information.

Audiovisual Media Services Directive 2010; Regulation of the European Parliament and of the Council dissemination of terrorist content online 2021

Russia

Article 4 of the Law on Mass Media 1991; Federal Law No. 405-FZ, December 2019

Belarus

The Belarusian Ministry of Interior can block access to an internet resource without a court order, if that outlet has received two written warnings in one year, or publishes content “threatening national security.

Mass Media Law: amended April 21, 2021

Australia

Broadcasting Services Act 1992; The Online Content Scheme; Sharing of Abhorrent Violent Material – Division 474.32 of the Criminal Code; The Online Safety Act 2021
signatory to the Christchurch Call To Action

Canada

Canada Anti-Terrorism Act 2015; Christchurch Call To Action May 2019

New Zealand

Films, Videos, and Publications Classification Act 1993; Broadcasting Act 1989

United Kingdom

Interim codes of practice on terrorist content and activity online

United States

The First Amendment of the US Constitution; Section 230 of the Communication Decency Act

Argentina

Argentina is a signatory to the Christchurch Call To Action. This is a multi-national accord that lays out a best practice framework for technology platforms to identify, review and censor terrorist content and activity on social media, and for governments to build the framework and legal structures for this work.

Christchurch Call To Action

Chile

Chile is a signatory to the Christchurch Call To Action. This is a multi-national accord that lays out a best practice framework for technology platforms to identify, review and censor terrorist content and activity on social media, and for governments to build the framework and legal structures for this work.

Christchurch Call To Action

Colombia

Colombia is a signatory to the Christchurch Call To Action. This is a multi-national accord that lays out a best practice framework for technology platforms to identify, review and censor terrorist content and activity on social media, and for governments to build the framework and legal structures for this work.

Christchurch Call To Action

Costa Rica

Costa Rica is a signatory to the Christchurch Call To Action. This is a multi-national accord that lays out a best practice framework for technology platforms to identify, review and censor terrorist content and activity on social media, and for governments to build the framework and legal structures for this work.

Christchurch Call To Action

Peru

Peru is a signatory to the Christchurch Call To Action. This is a multi-national accord that lays out a best practice framework for technology platforms to identify, review and censor terrorist content and activity on social media, and for governments to build the framework and legal structures for this work.

Christchurch Call To Action

Japan

Japan is a signatory to the Christchurch Call To Action. This is a multi-national accord that lays out a best practice framework for technology platforms to identify, review and censor terrorist content and activity on social media, and for governments to build the framework and legal structures for this work.

Christchurch Call To Action

Indonesia

Digital Platform Law (PP PSTE no. 71/2019 on the Government Regulation on the Implementation of Electronic and Transaction System 2020; Article 14 of Law No. 15/2003 on the elimination of acts of terrorism; Christchurch Call To Action

Singapore

Internet Regulatory Framework Internet Code of Practice 2016; Protection from Online Falsehoods and Manipulation Act 2019

South Korea

South Korea is a signatory to the Christchurch Call To Action. This is a multi-national accord that lays out a best practice framework for technology platforms to identify, review and censor terrorist content and activity on social media, and for governments to build the framework and legal structures for this work.

Christchurch Call To Action

China

Cyber Security Law 2017; Provisions on the Governance of Network Information Content Ecology 2015; China’s Counter-Terrorism Law 2015

Vietnam

Spreading false information is criminalized under specified conditions, and is intended to be used against “enemy and reactionary forces” who incite protests, riots, and terrorism.

Law 24 on Cybersecurity

Bangladesh

A seven-year prison sentence and a fine of up to 500,000 BDT (roughly 5,800 USD) for creating anarchy using digital media. The law does not extend liability for UGC to platforms that exercise due diligence.

Digital Security Act 2018

India

Information Technology (Intermediary Guidelines and Digital Media Ethics Code) Rules, 2021; Christchurch Call To Action

Switzerland

Switzerland is a signatory to the Christchurch Call To Action. This is a multi-national accord that lays out a best practice framework for technology platforms to identify, review and censor terrorist content and activity on social media, and for governments to build the framework and legal structures for this work.

Christchurch Call To Action May 2019

Pakistan

Social media companies must “deploy proactive mechanisms to ensure prevention of live streaming through Online Systems in Pakistan any content related to terrorism, extremism, hate speech, defamation, fake news, incitement to violence, and national security.”

Citizen’s Protection (Against Online Harm) Rules 2020

South Africa

Platforms should remove user-generated content that contains sexual conduct, that “violates or shows disrespect for the right to human dignity of any person; degrades a person; or constitutes incitement to cause harm; advocates propaganda for war; incites violence; or advocates hatred based on any identifiable group characteristic and that constitutes incitement to cause harm.” The Review Board can order its removal.

FBP Online Regulation Policy 2016; Films Publication Act 1965

Uganda

The Anti-Terrorism Act 2002; The Uganda Communications Act 2013; The Computer Misuse Act, 2011

Egypt

Social media accounts, blogs, or websites with more than 5,000 followers are classified as media outlets and placed under the supervision of the Supreme Council for the Administration of the Media. This entity has the power to block websites and file criminal complaints against platforms and users accused of offenses that are considered to constitute a threat to national security, disturb public peace, promote discrimination, violence, racism, hatred, or intolerance.

Law on the Organization of Press, Media and the Supreme Council for Media No.180

Ethiopia

Internet intermediaries are liable for policing content. Social media services must act within 24 hours to remove incendiary content upon receiving notifications about such communication or post.

Hate Speech and Disinformation Prevention and Suppression Proclamation 2020 2020:

Morocco

Law relating to the fight against terrorism 2003

Nigeria

Nigeria’s counter-terrorism framework contains the provisions for gathering intelligence from service providers and sharing information with foreign governments on terrorist use of the internet.

The Anti-Terrorism Act, 2013

Kenya

Service providers are required to assist in the investigation of offenses expeditiously, identify dangerous users, and collect and provide information to investigation officers. Platforms are also required to remove content on receipt of a court order.

Computer Misuse and Cybercrimes Act 2018

Israel

Israel’s Attorney General’s Office currently requests that inciting content be removed from social media platforms. This is not backed by regulation and therefore while some platforms comply, others do not.

Jordan

The Cybercrime Act 2019 added ‘smartphone applications’ into the scope of the preexisting Telecommunications Act. As a result, they must now comply with Article 29 of the Act. This considers any media or publishing material that “facilitates the commission and promotion of terrorist acts” to be terrorism, and requires applications to allow the monitoring of communication when a terrorist offense is suspected to be in progress.

Telecommunications Law (1995); Cybercrime Law 2019

Turkey

Anti-Terrorism Law; Defamation Law; Law 7253 amending “Law No. 5651 on The Regulation of Publications

Mexico

Mexico is a signatory to the Christchurch Call To Action. This is a multi-national accord that lays out a best practice framework for technology platforms to identify, review and censor terrorist content and activity on social media, and for governments to build the framework and legal structures for this work.

Christchurch Call To Action

Brazil

Brazil explicitly exempts platforms from liability for user-generated content. However, this liability exemption is removed if a court order states the content was illegal. Once content is found to be illegal, platforms must remove the content or face legal liability.

Marco Civil Law of the Internet 2014

Belgium

Internet platforms must remove terrorist content one hour after the receipt of a removal notice, they must determine which measures to put in place to reduce their exposure to terrorist content while maintaining the Right to Information.

Audiovisual Media Services Directive 2010
Regulation of the European Parliament and of the Council dissemination of terrorist content online 2021