Ms. Susan Wojcicki,
It’s been almost two years since the COVID-19 pandemic started. The world has seen time and time again how destructive disinformation and misinformation can be for social harmony, democracy, and public health; too many lives and livelihoods have been ruined, and far too many people have lost loved ones to disinformation. As an international network of fact-checking organizations, we monitor how lies spread online — and every day, we see that YouTube is one of the major conduits of online disinformation and misinformation worldwide. This is a significant concern among our global fact-checking community.
What we do not see is much effort by YouTube to implement policies that address the problem. On the contrary, YouTube is allowing its platform to be weaponized by unscrupulous actors to manipulate and exploit others, and to organize and fundraise themselves. Current measures are proving insufficient. That is why we urge you to take effective action against disinformation and misinformation, and to elaborate a roadmap of policy and product interventions to improve the information ecosystem — and to do so with the world’s independent, nonpartisan fact-checking organizations.
In the last year, we have seen conspiracy groups thriving and collaborating across borders, including an international movement that started in Germany, jumped to Spain and spread through Latin America, all on YouTube. Meanwhile, millions of other users were watching videos in Greek and Arabic that encouraged them to boycott vaccinations or treat their COVID-19 infections with bogus cures. Beyond COVID-19, YouTube videos have been promoting false cures for cancer for years.
In Brazil, the platform has been used to amplify hate speech against vulnerable groups, reaching tens of thousands of users. Elections are not safe either. In the Philippines, false content with over 2 million views denying human rights abuses and corruption during the Martial law years are being used to burnish the reputation of the late dictator’s son, one of the candidates in the 2022 elections. In Taiwan, the last election was marred by unsubstantiated accusations of fraud. The whole world witnessed the consequences of disinformation when a violent mob assaulted the U.S. Capitol last year. From the eve of the U.S. presidential election to the day after, YouTube videos supporting the “fraud” narrative were watched more than 33 million times.
The examples are too many to count. Many of those videos and channels remain online today, and they all went under the radar of YouTube’s policies, especially in non-English speaking countries and the Global South. We are glad that the company has made some moves to try to address this problem lately, but based on what we see daily on the platform, we think these efforts are not working — nor has YouTube produced any quality data to prove their effectiveness.
Your company platform has so far framed discussions about disinformation as a false dichotomy of deleting or not deleting content. By doing this, YouTube is avoiding the possibility of doing what has been proven to work: our experience as fact-checkers together with academic evidence tells us that surfacing fact-checked information is more effective than deleting content. It also preserves freedom of expression while acknowledging the need for additional information to mitigate the risks of harm to life, health, safety and democratic processes. And given that a large proportion of views on YouTube come from its own recommendation algorithm, YouTube should also make sure it does not actively promote disinformation to its users or recommend content coming from unreliable channels.
With all this in mind, we propose some solutions that would make a great deal of difference in reducing the dissemination of disinformation and misinformation on YouTube.
- A commitment to meaningful transparency about disinformation on the platform: YouTube should support independent research about the origins of the different misinformation campaigns, their reach and impact, and the most effective ways to debunk false information. It should also publish its full moderation policy regarding disinformation and misinformation, including the use of artificial intelligence and which data powers it.
- Beyond removing content for legal compliance, YouTube’s focus should be on providing context and offering debunks, clearly superimposed on videos or as additional video content. That only can come from entering into a meaningful and structured collaboration taking the responsibility and systematically investing in independent fact-checking efforts around the world that are working to solve these issues.
- Acting against repeat offenders that produce content that is constantly flagged as disinformation and misinformation, particularly those monetizing that content on and outside the platform, notably by preventing its recommendation algorithms from promoting content from such sources of misinformation.
- Extend current and future efforts against disinformation and misinformation in languages different from English, and provide country- and language-specific data, as well as transcription services that work in any language.
We hope you will consider implementing these ideas for the public good and to make YouTube a platform that truly does its best to prevent disinformation and misinformation being weaponized against its users and society at large. We are ready and able to help YouTube. We wish to meet with you to discuss these matters and find ways forward on a collaboration, and look forward to your response to this offer.
Africa Check (Senegal, Nigeria, Kenya, South Africa) / Animal Político – El Sabueso (Mexico) / Aos Fatos (Brazil) / Bolivia Verifica (Bolivia) / BOOM (India, Myanmar and Bangladesh) / Check Your Fact (USA) / Code for Africa – PesaCheck (Burkina Faso, Burundi, Cameroon, Central African Republic, Côte d’Ivoire, Ethiopia, Ghana, Guinea, Kenya, Mali, Niger, Nigeria Senegal, South Africa, Sudan, Tanzania, Uganda and Zimbabwe) / Colombiacheck (Colombia) / CORRECTIV (Germany) / Cotejo.info (Venezuela) / Chequeado (Argentina) / Delfi Lithuania (Lithuania) / Demagog Association (Poland) / Doğruluk Payı (Turkey) / Dubawa (Nigeria, Ghana, Sierra Leone, Liberia and The Gambia) / Ecuador Chequea (Ecuador) / Ellinika Hoaxes (Greece) / Fact Crescendo (India) / Fact-Check Ghana / FactCheck.org (USA) / FactSpace West Africa / Facta (Italy) / Factcheck.NI (UK) / Factly (India) / Factual.ro (Romania) / FactWatch (Bangladesh) / Fakenews.pl (Poland) / Faktisk.no (Norway) / Faktograf.hr (Croatia) / Faktoje (Albania) / Fast Check CL (Chile) / Fatabyyano (Middle East and North Africa) / Full Fact (UK) / GRASS – FactCheck Georgia / India Today Group (India) / Istinomer (Serbia) / Istinomjer (Bosnia i Herzegovina) / Hibrid.info (Kosovo) / Knack Magazine (Belgium) / La Silla Vacía (Colombia) / Lead Stories (USA) / Les Surligneurs (France) / Logically (UK) / Lupa (Brazil) / Maldita.es (Spain) / MediaWise (USA) / Mongolian Fact-checking Center (Mongolia) / MyGoPen (Taiwan) / Myth Detector (Georgia) / NewsMobile (India) / Newschecker (India and South Asia) / Newtral (Spain) / Observador – Fact Check (Portugal) / Open Fact-checking (Italy) / OŠTRO (Slovenia) / Pagella Politica (Italy) / Poligrafo (Portugal) / PolitiFact (USA) / Pravda (Poland) / Rappler (Philippines) / Raskrinkavanje (Bosnia i Herzegovina) / Re:Check/Re:Baltica (Latvia) / RMIT ABC Fact Check (Australia) / Rumor Scanner (Bangladesh) / Science Feedback (France) / StopFake (Ukraine) / Taiwan FactCheck Center / Tempo (Indonesia) / Teyit (Turkey) / The Healthy Indian Project/THIP Media (India) / The Journal FactCheck (Ireland) / The Logical Indian (Indian) / The Quint (India) / The Washington Post Fact Checker (USA) / The Whistle (Israel) / Univision – elDetector (USA) / VERA Files (Philippines) / Verificat (Spain) / Vishvas News (India) / Vistinomer (North Macedonia) / VoxCheck (Ukraine) / 15min (Lithuania)