Resource Guide

To begin, this document maps each company’s efforts that fulfill the GIFCT membership criteria. These are criteria that all tech companies must have publicly in place before being considered for GIFCT membership. First, companies must have public platform standards that explicitly prohibit the promotion of terrorism and/or violent extremism in their terms of service, community guidelines, or other publicly available content policies. This first section explores how companies discuss their frameworks for prohibiting the use of their platforms by terrorist and/or violent extremists.

The second section of this document links to member companies’ Transparency Reports. All GIFCT member companies must publish transparency reports on at least an annual cadence. While we have seen larger GIFCT member companies continue to develop more granular and robust transparency reports, the initial development of such a report can be the most daunting piece of the GIFCT membership criteria for smaller companies.

The third section covers the safety portals and wider safety information that platforms make available so that their users are aware of the range of tools they can use to stay safe. Safety information is particularly crucial for activists, academics, journalists, and users that might be vulnerable to violent extremist groups. On some platforms, this information includes how to report hacked accounts, how to better ensure privacy in your account set-up, and how to flag abuse and harassment to the company.

The fourth section reviews the ways that GIFCT members ensure they have the ability to receive and act on reports of illegal activity or activity violating their terms of service from their users.This is also a GIFCT membership requirement. Reporting capabilities include internal flagging tools, reporting portals, and outreach emails. These mechanisms vary across platforms based on how they operate. Whatever form it takes, it is important to ensure that the community using a platform can highlight abuses to that company that might be missed by any proactive detection efforts.

We know that content removal alone addresses a symptom and not the root causes of radicalization leading to violence. The fifth section of this document highlights resources related to GIFCT’s membership criteria to support the capacity of civil society organizations to challenge violent extremism. Under this remit, some of the larger social media companies have developed international counter-narrative programs and tooling to help activists and NGOs scale and optimize their voices online as part of their efforts to push back on hate speech and extremism.

While community standards, transparency reports, safety outlets, and reporting guidance work to define and disallow certain types of content and activities on a given platform, we know that content removal alone addresses a symptom and not the root causes of radicalization leading to violence. The fifth section of this document highlights resources related to GIFCT’s membership criteria to support the capacity of civil society organizations to challenge violent extremism. Under this remit, some of the larger social media companies have developed international counter-narrative programs and tooling to help activists and NGOs scale and optimize their voices online as part of their efforts to push back on hate speech and extremism.

The final sections of this document highlight preventative measures to counter extremism online, specifically, digital literacy training and tooling. We have included, where available, open-access resources related to digital literacy along with relevant resources that guide activists and NGOs on how to utilize advertising and marketing tools to optimize the reach of their campaigns to their target audiences and how they can measure these efforts.

For ease of use, the appendix gives all the resource links in one comprehensive table by company, theme and URL link. GIFCT hopes that this document will serve as a useful resource and guide for better understanding and joint efforts in combating terrorism and violent extremism online.

A key piece of the GIFCT Membership criteria is that members must prohibit terrorist and/or violent extremist exploitation on their services and include this explicitly in their publicly-available terms of service or content standards. Just like governments, intergovernmental institutions, civil society organizations, and academics – tech companies often have slightly different definitions of “terrorism,” “terrorist content” and “violent extremism.” While there is no one globally agreed upon definition of terrorism or violent extremism, most tech companies, in their independent capacity, have developed definitions and approaches based on existing resources and in consideration of what will work best based on how their platforms operate.

Larger platforms often have content standards that address a range of harmful content and abuse related to violence and criminal behavior. For example, Facebook’s dangerous individuals and organizations policies explicitly prohibits any organizations or individuals that proclaim a violent mission or are engaged violence to have a presence on their platform. This includes organizations or individuals involved in terrorist activity, organized hate, mass murder, organized violence and large-scale criminal activity. Likewise, Instagram’s community guidelines outline a clear set of policies for what is and is not allowed on their platform. These guidelines explicity prohibit material that supports or praises terrorism, organized crime, or identified hate groups.

Also explicit in their guidelines, Twitter and YouTube do not allow violent organizations, including terrorist organizations, violent extremist groups, or individuals who affiliate with and promote the illicit activities of these groups to use their platforms for any purposes, including recruitment. These policies are part of the companies’ respective sets of comprehensive guidelines against hate speech and harmful content.

Similarly, LinkedIn is committed to ensuring their platforms remain safe for all their users and do not tolerate content that depicts terrorist activity, that is intended to recruit for terrorist organizations, or threatens, promotes, or supports terrorism. To help guide users, LinkedIn provides publishing guidelines. In addition to prohibiting hate speech and content that advocates for violence against others, Microsoft’s standards also prohibit terrorist content. This is defined as material posted by or in support of organizations included on the Consolidated United Nations Security Council Sanctions List that depicts graphic violence, encourages violent action, endorses a terrorist organization or its acts, or encourages people to join such groups. JustPaste.It’s Terms of Service prohibits terrorist content, which it defines as content in violation of EU Directives and EU Member State laws on terrorist offenses, or content produced by or attributable to terrorist groups or entities designated by the European Union or by the United Nations.

Dropbox has put in place an an Acceptable Use Policy page which explicitly condemns the use of their platform to publish or share materials that contain extreme acts of violence or terrorist activity, including terrorist propaganda. Similarly, MailChimp’s Acceptable Use Policy is clear about suspending or terminating accounts which contain behavior, content, and in particular any campaigns promoting or supporting terrorism or violent extremism. Tumblr’s Community Guidelines also state that the platform does not tolerate content that promotes, encourages, or incites acts of terrorism, including content which supports or celebrates terrorist organizations, their leaders, or associated violent activities.

WhatsApp’s responsible use guide includes a video, a list of best practices, and what to avoid on the messaging platform. Though WhatsApp uses end-to-end encryption – meaning that the company does not have access to users’ communications – it has clear reporting mechanisms and explicitly condemns illegal, obscene, defamatory, threatening, intimidating, harassing, hateful, racially, or ethnically offensive content. While MEGA’s platform also features end-to-end encryption, its Terms of Service are explicit in not condoning, authorizing, supporting or facilitating the storage of any materials relating to violent extremism.

Platforms not typically considered to be social media platforms also recognize the importance of providing clear guidelines and rules to keep their users safe. For example, Amazon explicitly prohibits any activity that encourages or supports behavior that is illegal, including violence or content encouraging physical or financial harm, including terrorism. In Airbnb’s trust and safety hub, its community standards refer to five pillars to guide its users: safety, security, fairness, authenticity and reliability. WordPress.com’s User Guidelines clarify what is and is not allowed on user sites, while its Terrorist Activity page specifically explains how websites for United States government-designated terrorist groups and calls for violence are prohibited on websites that it hosts. Zoom also explicitly prohibits accounts that are affiliated with organizations that use or promote violence against civilians to further their causes, either on or off the platform. Their Terrorism and Violent Extremism Policy outlines how they determine whether a group or individual is a terrorist or violent extremist.

While the public tends to focus on the four to five larger GIFCT member companies, this criterion is necessary for all GIFCT members, large or small. Airbnb, Discord and Pinterest have reviewed and refreshed their Terms of Service and Community Guidelines through their membership process. While undergoing the GIFCT membership process, applying companies are given access to the Tech Against Terrorism Mentorship Program, sponsored by GIFCT. This supports tech companies who want to further develop policy language references to terrorism and/or violent extremism in their Terms or Service or Community Guidelines to better combat terrorist use of their services.

In addition to the resources from Tech Against Terrorism, GIFCT also ensures there is a large body of academic and expert insights that members can review as they further develop their policies and understanding of how terrorist and violent extremists exploit online services. GIFCT works in close partnership with Global Network on Extremism and Technology (GNET), an independent but GIFCT-funded global academic network. Topical insights, reports, and workshops with global experts are designed to highlight trends and adversarial shifts around the nexus between extremism and technology.

NGOs, academics, and governments can utilize this information to better understand how platforms enforce community standards, respond to violating content, and what the landscape of violating content looks like on a platform. This can inform how NGOs communicate about violent extremism on the platform.

Some of GIFCT’s members include specific categories in their transparency data for terrorism and violent extremism, while others might have this information incorporated into other areas of violating or removed content. Member companies that include specific data on content flagged and/or removed for promoting terrorism are Facebook, JustPaste.It, Microsoft, Twitter, WordPress.com, YouTube, and Zoom based on their distinct, respective definitions of this content. These members are able to provide more granular data due to a number of different factors, including greater resources and additional personnel, as well as reporting by government entities, all of which support more robust efforts to detect, label, and review content.

Transparency reports also include data on the enforcement of community standards, requests for information or content removal by governments and law enforcement, as well as how content is restricted based on local laws. LinkedIn, MEGA and Mailchimp’s and Tumblr’s transparency reports all feature a combination of these elements.

Transparency reports from GIFCT members Airbnb, Amazon, Pinterest, Dropbox, and Zoom primarily provide data focused on government and law enforcement requests due to how their platforms are set up and how on-platform abuse takes place. Discord’s most recent transparency report focuses primarily on the company’s responses to reports made by users, as well as efforts by the company to proactively enforce community standards by removing racist and extremist groups organizing on its servers.

Recognizing that tech companies, governments, and NGOs can always improve how they convey transparency and what is added to transparency data, GIFCT hosts a Transparency Working Group that brings together international, multi-sector stakeholders to foster better collaboration on how best to approach transparency in the context of tech platforms. This multi-sector effort ensures that GIFCT can best support its mission to facilitate dialogue between tech companies and other stakeholders on key issues.

For tech companies seeking membership that do not currently produce regular transparency reports, GIFCT provides guidance by connecting companies with our partner, Tech Against Terrorism, whose mentorship program facilitates developing best practices around transparency in a way that works for the specific platform. As a reminder, all platforms will naturally have different approaches to transparency reporting both in terms of frequency and data included. Some larger companies give transparency updates as often as quarterly, while that sort of frequency would put a massive strain on the communication, policy, data and legal teams at smaller companies. It should also be noted that depending on how a platform is set up, there might naturally be less data available to report on.

Finally, GIFCT has its own commitment to transparency and produces an annual Transparency Report, which documents our organization’s growth, year’s activities and progress in joint-tech innovations, and knowledge-sharing initiatives.

Companies such as Discord, Facebook, Microsoft, YouTube, and Zoom have extensive safety hubs that cover a wide range of abuse topics and risk mitigation best practices. For example, Discord provides account tips, server management and how to report problems directly to their Trust and Safety team as well as a Parents and Educators section. Facebook also ensures that its safety hubs cater to a range of different users, thereby addressing the needs of law enforcement, parents, and young people. Similarly, YouTube has safety resources focusing on online harms as well as parent, teens and educator resources. Zoom also has a number of resources including tips on how to prevent uninvited guests from joining meetings, blogs on security and privacy, and classroom safety tips for parents and teachers.

Sometimes safety resources include third party facilitation and external experts to ensure safety measures develop as adversarial shifts occur. For example, Twitter has a dedicated Twitter account with the latest safety tools, resources, and updates to support its community. Twitter’s Trust and Safety Council is a global group of independent expert organizations that look into a wide array of issues from online safety and harassment to human and digital rights. Twitter also provides tools geared towards brand safety. Meanwhile, platforms such as Facebook have a Safety Advisory Board, composed of leading internet safety organizations from around the world, who provide expertise, perspective, and insights that inform their approach to safety. The tools provided address a range of safety concerns, such as privacy measures and account security.

Instagram’s Safety Hub provides comprehensive safety tools and a help center that gives safety guidelines. The center provides resources around privacy and security, reporting abuse, and digital wellbeing resources for its community.Microsoft’s reports hub provides a Digital Safety Content Report that explicitly details their mechanisms for prohibiting certain content and conduct. In particular, it dedicates a significant section on what they do to help to prevent terrorists and violent extremists from exploiting digital platforms, including by addressing images and videos on their hosted services that include terrorist or violent extremist content.

WhatsApp aims to ensure that all users remain safe while using its platform through tools, including privacy controls and reporting mechanisms. It has also published a white paper on how WhatsApp prevents abuse of automated and bulk messaging. Similarly, MEGA provides users with a range of tools to remain safe across their different streams.

Dropbox and Pinterest, have developed principles and platform-specific approaches to tackle terrorist exploitation on their platforms with a focus on keeping users safe. Additionally, Airbnb provides safety guidance for both online and offline components of their service, providing users with proper information throughout their experience. On Airbnb’s Trust and Safety page and in the Airbnb Host Resource Center, both hosts and travelers can learn more about the wider safety resources and information offered by the platform.

The evolution of safety tools and risk mitigation strategies also comes from external dialogues and international commitments with bodies such as the United Nations Counterterrorism Executive Directorate, the European Union Internet Forum, and the Christchurch Call to Action. GIFCT members Amazon, Facebook, Microsoft, Twitter and YouTube signed onto the Christchurch Call to Action in 2019 as part of their commitment to address the exploitation of technology by users looking to spread terrorist and violent extremist agendas online. These companies and GIFCT committed to a nine-step plan for individual and collective action to prevent and respond to terrorist or violent extremist content.

In the GIFCT-funded Campaign Toolkit, there is also a Digital Security Checklist, which lays out the basics for digital security and steps to take if a campaign is threatened, harassed, or targeted by bad actors.

Overall, safety across digital platforms is a shared industry responsibility and efforts are furthered by a range of stakeholders including private sector, academic researchers, civil society, governmental and intergovernmental actors. Multi-stakeholder engagement to further this is one of GIFCT’s foundational goals. When different stakeholders work together to address this ever evolving and complex challenge, we see progress that would be difficult for any one entity to tackle alone.

In addition to outreach portals, other member companies use a combination of tools including content-adjacent reporting mechanisms and interactive online forms. Some larger companies have tools for giving more granular feedback on the type of abuse violating content might fall under. Many of these reporting tools also provide guidance on how to report content using the aforementioned content-adjacent mechanisms, ensuring that users are able to more easily report content in the future.

Discord, Dropbox, and Pinterest make use of general-purpose forms for reporting violating content. Discord requires users to share links to reported messages in their online form and provides instructions for users to do so successfully. In addition to their online forms, Dropbox and Pinterest also provide instructions for users to report content where it appears on the platform. Similarly, WordPress.com’s reporting mechanisms include on-platform options for users logged into WordPress.com accounts and a form for users to directly contact the company. Amazon allows customers to report violating content via email and content adjacent reporting mechanisms. Information on both of these tools is located at the base of Amazon’s Community Guidelines Page. Zoom users can report abuse either in-meeting, through a web report form, or through the ‘Meetings’ menu via the Zoom web portal. Like Discord, Zoom has a space for sharing additional files, screenshots, and transcripts that may help the Trust and Safety team pass judgement.

Airbnb provides guidance for reporting and blocking other users and reporting discrimination on their platform. Additionally, Airbnb provides users with a neighborhood support portal to call or chat with a representative about safety and broader neighborhood concerns.

Tumblr provides guidance to report blogs or offensive content where they are found on the platform, as well as a form for reporting by type of violation, including reporting suspected terrorist content.

Facebook, Instagram, LinkedIn, Microsoft, and Twitter rely on similarly structured pages in their help centers to provide information about reporting mechanisms and instructions for how to report content, profiles, and other platform aspects. In addition to on-platform reporting resources, each of these companies also provides direct contact reporting options and guidance for reporting content without having an active user account.

Counterspeech efforts supported by tech companies often focus on helping activists and organizations produce, launch and evaluate localized content to challenge engaging narratives created by violent extremists. Other forms of counterspeech aim to create friction between individuals searching for violent extremist content by proactively surfacing alternative voices or local resources for disengagement. These efforts rely on strong partnerships between the tech companies and local civil society organizations around the world, reinforcing GIFCT’s belief that multi-sector collaboration is a critical element of effectively combating extremism in online spaces.

Facebook’s Counterspeech Hub houses a range of tips, tools, research and examples for activists and organizations to better understand Facebook’s current initiatives and resources in the counterspeech space. The site’s Initiatives page showcases the range of global projects, partnerships, and positive intervention approaches already underway. These include programs they run, such as the Redirect Initiative, as well as programs they have in partnership with bodies like the United Nations, such as Extreme Lives. The Counterspeech Resources section offers broader guidance for nonprofit organizations on how to take advantage of Facebook’s tools and products, as well as more specific information on crafting counterspeech projects.

YouTube’s Creators for Change program is a global initiative that partners with content creators from over 20 countries to promote positive messages, give space to community voices, and empower youth voices as drivers of social good. Creators – known as Ambassadors and Fellows – receive mentorship and promotional assistance to work on Impact Projects, which engage viewers on a variety of challenging topics, including combating extremism, hate, and intolerance. YouTube was also an early pioneer of The Redirect Method, a tool created by Jigsaw and Moonshot CVE to “redirect” Google Search and YouTube users away from violent extremist recruitment content to YouTube videos that credibly refuted extremist content and recruitment materials. Modified versions of the initial pilot are being used across Microsoft, Facebook, and Instagram to address a variety of challenges.

To further foster this, GIFCT worked to gather resources that serve activists and organizations in their counterspeech efforts, working with ISD Global to produce the online Campaign Toolkit, a digital resource trove for anyone interested in producing and promoting counterspeech campaigns to global audiences. The Campaign Toolkit provides users with guidance as they develop campaigns and counter-narrative projects, as well as resources from many of our member companies to optimize the use of their platforms. Additionally, GIFCT hosts a Content-Sharing Algorithms, Processes, and Positive Interventions Working Group to ensure we further dialogues around best practices, monitoring and evaluation, and innovations.

To learn more about counterspeech and counter-narrative deployment, you can also look to our academic partners at the Global Network on Extremism and Technology (GNET) for published work by academics and researchers on this important and ever-evolving subject.

Some of our companies have built out substantial educational resources for digital literacy on their platforms and the internet at-large. Within Discord’s Safety Center, users can find sections on “Controlling Your Experience” and a section for “Parents and Educators,” which provide in-depth information for users to make their own experience safer. This also includes resources for those that might have responsibility for the safety of others on the platform.

Facebook hosts a Digital Literacy Library, which offers thorough modules for comprehension on several topics including; Security, Privacy and Reputation, and Community Engagement. Likewise, Instagram’s Community page provides users with information about how the platform aims to protect its users, as well as a broad range of safety tools and programs to ensure safe, positive user experiences.

LinkedIn’s Staying Safe page highlights best practices for LinkedIn users to maximize their use of the platform’s products safely and effectively. The page also offers guidance for protecting personal information and resources for parents, educators and teenagers using the platform. Microsoft’s digital literacy resources include two LinkedIn Learning courses, as well as a series of courses presented in a series of downloadable videos. Microsoft also hosts resources dedicated to Digital Civility online and open source Resources and Research.

Twitter, in collaboration with the United Nations Educational, Scientific and Cultural Organization (UNESCO), produced a document titled Teaching and Learning with Twitter: Media and Information Literacy. Additionally, A Safer Twitter, located within Twitter’s Help Center, hosts a series of short videos that explain how to use some of the basic mechanisms offered on the platform to enhance the user’s safety and experience.

Finally, YouTube provides a suite of Safety Resources, which cover a broad list of digital literacy subjects with specific information catering to parents, educators and teenagers. Additionally, YouTube also provides general-purpose resources for staying safe online and staying safe on YouTube, which educate users on phishing, account safety and basic mechanisms such as reporting and blocking.

For example, Facebook for Business is a free resource that allows small and medium businesses to expand their internet presence for greater visibility and reach. Meanwhile, the Facebook Social Impact site provides resources around a range of impact aimed streams: charitable causes, crisis response to vulnerable and affected people, health and mentorship.

The YouTube Advertising team helps organizations connect with their target audiences by supporting development content. YouTube Social Impact also helps users harness YouTube’s scale, technology and talent to ignite and sustain movements that drive measurable social change. In this way YouTube also partners with others such as Google.org, Google for Nonprofits, Jigsaw and YouTube Creators For Change.

Twitter for business provides a range of free resources around campaign types, takeover products, advertising best practices and analytics. Especially for small businesses these are valuable tools on how to navigate their platform with recommendations, ideas and educational resources. Similarly, Microsoft provides a range of different open-source tools to support individuals, organizations, activist and business growth. Microsoft’s Advertising provides insights, technology and proven expertise that empower organizations to deliver better-performing marketing campaigns. Microsoft Advertising Learning Lab offers free, on-demand, online learning courses to help users better understand the Microsoft Advertising interface. Tools and best practices for maximizing campaigns and investment with Microsoft Advertising are also available. Finally, Microsoft Nonprofit ensures the benefits of world-class cloud technology is accessible and affordable for nonprofits. They provide nonprofit grants and discounts as well as industry-leading solutions help accelerate their impact. Microsoft provides step-by-step instructions to make this process as easy as possible.

Finally, above and beyond GIFCT member resources, additional resources include a Campaign Toolkit, which act as an information hub, providing activists and organizations with resources on a range of useful information, in particular for advertising, audience targeting and social good.