Inkcazelo engeyonyani
Iinkcukacha zepolisi
TSHINTSHA ILOGU
Namhlanje
Izizathu Zepolisi
Inkcazelo engeyonyani yahlukile kwezinye iintlobo zezinto ezithethwayo ezishukuxwe kwiMigangatho Yoluntu yethu kuba akunakukwazeka ukuchaza uluhlu olupheleleyo lwezinto ezingavunyelwayo. Ngokomzekelo, xa kufikwa kubundlobongela okanye kwizenzo ezinentiyo, iipolisi zethu ziyayichaza ikhontenti esingayivumeliyo, ibe nkqu nabantu abangavumelaniyo nezo polisi bangakwazi ukuzilandela. Kodwa ke, xa kufikwa kwinkcazelo engeyonyani, asikwazi kukrwela umgca ngolo hlobo. Ihlabathi lihlala litshintsha, ibe into eyinyani ngalo mzuzu kungenzeka ibe ingabiyonyani kumzuzu olandelayo. Enye into nabantu banamanqanaba angafaniyo enkcazelo ngehlabathi elibajikelezileyo, ibe bangayikholelwa into njengeyinyani ngoxa ingeyonyani. Ipolisi eyalela nje "inkcazelo engeyonyani" ayiyi kubanceda bazi abantu abasebenzisa iinkonzo zethu kwaye ayinakukwazi ukunyanzeliswa, kuba asikwazi kuyifikelela ngokugqibeleleyo inkcazelo.
Kunoko, iipolisi zethu zichaza iindidi ngeendidi zenkcazelo engeyonyani zize zizame ukunikela ngokhokelo olucacileyo ngendlela esiyiphatha ngayo intetho enjalo xa siyibona. Kudidi ngalunye, indlela esenza ngayo ibonisa ukuzama kwethu ukubhalansisa imigangatho yethu yokuveza izimvo, ukhuseleko, isidima, ubunyani, nobumfihlo.
Siyayisusa inkcazelo engeyonyani xa kusenokwenzeka ifake isandla ngokungqalileyo kumonzakalo wasemzimbeni okhawulezileyo. Siyayisusa ikhontenti naxa emaninzi amathuba okuba ifake isandla ngokungqalileyo ekuphazamiseni iinkqubo zepolitiki. Xa kugqitywa ukuba yintoni inkcazelo engachananga kwezi ndidi, sisebenzisana neengcali ezizimeleyo ezinolwazi nobugcisa bokuhlola inyaniso ngaloo khontenti nokuba maninzi na amathuba okuba ifake isandla ngokungqalileyo kumonzakalo oza kwenzeka msinya. Oku kuquka, ngokomzekelo, ukusebenzisana nemibutho yamalungelo abantu ephakathi kwabantu elizweni ukuze sigqibe ngobunyani bamarhe emfazwe yamakhaya.
Kuyo yonke enye inkcazelo engeyonyani, sijolisa ekunciphiseni ukunwenwa kwayo okanye ekuyileni imeko edala uthethwano oluncedayo. Siyazi ukuba abantu badla ngokusebenzisa inkcazelo engeyonyani ngeendlela ezingeyongozi, njengokubaxa izinto xa kuncokola ("Eli qela lineyona rekhodi imbi kwimbali yemidlalo!") okanye xa kuqhulwa okanye kuhlekiswa ("Umyeni wam usandula ukuphumelela iwonga loMyeni Wonyaka.") Kwakhona bangabalisa amava abo ngamabali aquka izinto ezingachananga. Kwezinye iimeko, abantu baveza iimbono zabo ezendeleyo ezigqalwa ngabanye abantu njengezibubuxoki okanye bashere inkcazelo abakholelwa ukuba iyinyaniso kodwa abanye bona bayigqale njengengaphelelanga okanye elahlekisayo.
Ngenxa yokuba siyiqonda indlela exhaphake ngayo intetho elolu hlobo, sijolisa ekunikeni iiuser inkcazelo eluncedo xa kunokwenzeka kubekho ikhontenti engeyonyani okanye edidayo. Njengenxalenye yaloo mzamo, kwiindawo ezingaphandle kwelizwe laseUnited States sisebenzisana neminye imibutho ehlola ubunyani kwiinxalenye zehlabathi ezininzi ukuze zihlolisiswe uze ureyithe ukuchana kweyona khontenti isasazekileyo kwiiplatfomu zethu (jongaapha uze ujonge nalapha ubone indlela esisingatha ngayo izinto eUnited States). Ukuqala eUnited States, iprogramu yeCommunity Notes ivumela abantu bongeze ingcombolo engakumbi kwiipowusti zikaFacebook, uInstagram noThreads ezisenokulahlekisa okanye zidide abantu (jonga apha). Kwakhona sibonelela ngezinto zokusebenza ukuze ikhule into yokufundiswa kwabantu ngemidiya nezedijithali ukuze benze isigqibo sokufunda, ukuthemba, kunye nokukwazi ukushera ngokwabo. Siye sifune ukuba abantu bachaze, besebenzisa isixhobo sethu sokuchaza seAI, xa bepowusta ikhontenti yokoqobo enevidiyo ebonakala iyeyokwenyani okanye eneodiyo evakala iyeyokwenyani kodwa iyilwe okanye itshintshwe ngedijithali, ibe sisenokukhupha izohlwayo xa bengayenzi loo nto. Singayifaka neleyibhuli kwikhontenti ethile eyilwe okanye yatshintshwa ngedijithali enyusa umngcipheko wokulahlekisa abantu kuba obalulekileyo kuwonkewonke.
Okokugqibela, asiyivumeli ikhontenti nezenzo kwezinye iinkalo edla ngokugudla ukusasaza inkcazelo engeyonyani. Ngokomzekelo, Imigangatho Yoluntu yethu iyazalela iiakhawunti zobuxoki, ubuqhinga, kunye nendlela yokuziphatha engafanelekanga echuliweyo.
Njengoko okwenzeka kwi-intanethi nangaphandle kwayo kutshintsha yaye kuhambela phambili, siza kuqhubeka sihambela phambili nathi kwezi polisi. Iiakhawunti ezishera ngokuphindiweyo inkcazelo engeyonyani edweliswe ngezantsi zingathi, ukongezelela ekunyanzelisweni kwepolisi kwikhontenti yazo, zenziwe zingabonwa ngabantu abaninzi, zibekelwe izithintelo zokupapasha izibhengezo, okanye zisuswe kwiiplatfomu zethu. Inkcazelo engakumbi ngoko kwenzekayo xa uMeta esusa ikhontenti ungayifumana apha.
Izikhokelo
Inkcazelo engeyonyani esiyisusayo:
Siyazisusa ezi ntlobo zilandelayo zenkcazelo engeyonyani:
I. Ukonzakalisa emzimbeni okanye ubundlobongela
Siyayisusa inkcazelo engeyonyani okanye amarhe angaqinisekiswanga eziye iingcali zethu zagqiba kwelokuba maninzi amathuba okuba ifake isandla kumngcipheko wobundlobongela okanye ukwenzakalisa abantu emzimbeni okusele kuza kwenzeka. Inkcazelo engeyonyani siyichaza ngokuthi yikhontenti eyenza ibango elifunyaniswe libubuxoki ngomnye umntu onikwe igunya. Sichaza amarhe angaqinisekiswanga ngokuthi libango elinomthombo ongqinwe ziingcali njengokunzima kakhulu okanye ekungenakwenzeka ukuba ufunyanwe, elingenamithombo enegunya, apho kungekho kungekho nto inesihlahla enokuphikiswa njengobubuxoki, okanye apho ibango lingakholeleki kakhulu okanye lingavakali tu.
Siyazi ukuba ngamanye amaxesha inkcazelo engeyonyani engabonakala ingenabuzaza inako, kwimeko ethile, ukufaka isandla kumngcipheko wokoqobo, oquka izisongelo zobundlobongela ezingafaka isandla ekunyuseni umngcipheko wokufa, umenzakalo onzulu, okanye omnye umenzakalo wasemzimbeni. Sisebenza nenetwekhi esehlabathini jikelele yemibutho engeyokarhulumente (iiNGO), imibutho engenzelwanga ngeniso, imibutho yokunceda abahlelelekileyo, nemibutho yamazwe ngamazwe eziingcali kwezi meko zasekuhlaleni.
Kumazwe anomngcipheko ophezulu wobundlobongela ekuhlaleni, sisebenza nabasekuhlaleni ukuze siqonde kusengaphambili ukuba ngawaphi amabango angafaka isandla ngokungqalileyo kumngcipheko womonzakalo emzimbeni. Emva koko siye sisebenzele ukufumana nokususa ikhontenti eyenza loo mabango kwiplatfomu yethu. Ngokomzekelo, emva kokuthetha neengcali zasekuhlaleni, singayisusa imidiya engangeniyo kuloo meko ezenza ngathi ibonisa izenzo zobundlobongela, amaxhoba okanye abenzi bobundlobongela, izixhobo, okanye izinto zasemkhosini.
II. Inkcazelo engeyonyani eyonzakalisayo yezempilo
Sithethana nemibutho yezempilo ephambili ukuze sifumane inkcazelo engeyonyani yezempilo esenokufaka isandla ngokungqalileyo kumonakalo okhawulezileyo kwimpilo nakukhuseleko lukawonkewonke. Inkcazelo engeyonyani yezempilo eyingozi esiyisusayo iquka ezi zinto zilandelayo:
- Inkcazelo engeyonyani ngezitofu zokugonya. Siyayisusa inkcazelo engeyonyani ngokuyintloko ethetha ngezitofu zokugonya xa abasemagunyeni bezempilo begqibe kwelokuba loo nkcazelo ibubuxoki kwaye maninzi amathuba okuba ifake isandla ekubeni abantu bangavumi ukugonywa. Iquka ethi:
- Izitofu zokugonya zibangela iautism (Umzekelo: "Abantwana abaninzi beneautism nje zezi zitofu zokugonya zandileyo.")
- Izitofu zokugonya zidala Isigulo Sokusweleka Kwemveku Ngesiquphe (iSudden Infant Death Syndrome (Umzekelo: "Anazi na ukuba zizitofu zokugonya ezibangela iSIDS?"
- Izitofu zokugonya zibangela izifo ezenzelwe ukuba zikhusele kuzo, okanye zenza umntu ofumene eso sitofu sokugonya abe namathuba amaninzi okufumana isifo. (Umzekelo: "Eyona nto yenziwa kukugonywa kukunyusa nje amathuba okufumana izifo kuba zinentsholongwane yeso sifo ngaphakathi kwazo. Lumkani!")
- Izitofu okanye izinto ezenziwe ngazo ziyabulala, zinetyhefu, ziyenzakalisa, okanye ziyingozi (Umzekelo: "Kulungile ke, ungagonya, ukuba phofu ayikukhathazi into yokufaka ityhefu emzimbeni wakho.")
- Ukukwazi komzimba ukulwa iintsholongwane ngokwendalo kukhuseleke ngaphezu koko kubangelwa zizitofu zokugonya (Umzekelo: "Eyona nto ikhuselekileyo kukusifumana isifo kunokugonya.")
- Kuyingozi ukufumana izitofu eziliqela ngexeshana elifutshane, kwanokuba loo nto icetyiswa ngabezamayeza (Umzekelo: "Ungaze uhlabe isitofu sokungonya esingaphezu kwesinye ngexesha, yingozi loo nto - andikhathali nokuba uthini ugqirha wakho!")
- Izitofu zokugonya azizithinteli izifo ekuthiwa ziyazinqanda. Kodwa ke, kwiCovid-19, iflu, nesifo seengcongconi, asiwasusi amabango athi ezo zitofu zokugonya azimthinteli umntu ekosulelweni zezo ntsholongwane. (Imizekelo: Siyalisusa elithi – “Isitofu sokugonyela ipolio asikunqandi tu ukuba ukuba ungasifumani esi sifo"; Siyalisusa elithi – “Izitofu zokugonya azikuthinteli ekubeni ufumane izifo"; Siyalivumela elithi – “Izitofu zokugonya azikunqandi ukuba ungafumani iCOVID-19, yiloo nto kusafuneka umele kude nabanye abantu kwaye ufake imaski xa ukunye nabanye abantu.")
- Ukosulelwa yimasisi akubulali (kufuneka inkcazelo engakumbi kunye/okanye imeko) (Umzekelo: "Ungakhathazeki ngokosulelwa yimasisi, ayibulali.")
- UVithamin C usebenza kanye njengezitofu zokugonya ekuthinteleni izifo ezithintelwa zizifo ezinezitofu zokugonya.
- Inkcazelo engeyonyani engezempilo ebudeni bexesha bemeko yongxamiseko yezempilo echaphazela uwonkewonke. Siyayisusa inkcazelo engeyonyani ebudeni bemeko yongxamiseko yezempilo echaphazela uwonkewonke xa abasemagunyani bezempilo baseburhulumenteni begqiba kwelokuba loo nkcazelo ibubuxoki kwaye maninzi amathuba okufaka isandla kumenzakalo wasemzimbeni, kuquka ukufaka isandla kumngcipheko wokuba abantu bafumane okanye basasaze isifo esiyingozi okanye bangavumi ukunikwa isitofu sokugonya esinxulumene naso. Iimeko zongxamiseko ezibandakanya impilo kawonkewonke sizigqala sisebenzisana nabasemagunyeni bezempilo behlabathi jikelele nabasekuhlaleni.
- Ukupromowutha okanye ukuthethelela amayeza ommangaliso ayingozi ukuze kunyangwe iingxaki zempilo. La mayeza aquka unyango apho indlela yokulisebenzisa, kwimeko yezempilo, kungenzeka ukuba ifake isandla kumngcipheko womonzakalo onzulu okanye ekufeni, kwaye olo nyango lungenzelwanga injongo evakalayo yezempilo (Umzekelo: ibleach, idisinfectant, iblack salve, icaustic soda).
III. Ukuphazamisana Nabavoti Okanye Nokubalwa Kwabantu
Ngelizama ukukhuthaza ukuthembeka konyulo nokokubalwa kwabantu, siyayisusa inkcazelo engeyonyani apho emaninzi amathuba okuba ifake isandla kumngcipheko wokuphazamisana nokukwazi kwabantu ukuthatha inxaxheba kwezo nkqubo. Oku kuquka ezi zinto zilandelayo:
- Inkcazelo engeyonyani ngemihla, iindawo, amaxesha, neendlela zokuvota, ukubhalisela unyulo, okanye ukuthatha inxaxheba ekubalweni kwabantu.
- Inkcazelo engeyonyani malunga nokuba ngubani ongavota, ukuba zithini iimfaneleko zokuvota, ukuba ngaba ivoti iza kubalwa, nokuba yeyiphi inkcazelo okanye zeziphi izinto omele uze nazo ukuze uvote.
- Inkcazelo engeyonyani malunga nokuba umntu othile uzigqatsele ukunyulwa okanye akunjalo.
- Inkcazelo engachananga engokuba ngoobani abangathatha inxaxheba ekubalweni kwabantu nokuba yeyiphi inkcazelo okanye izinto amele umntu eze nazo ukuze athathe inxaxheba.
- Inkcazelo engachananga ngokubandakanyeka kukarhulumente ekubalweni kwabantu, kuquka, xa kusebenza oku, into yokuba inkcazelo yobalo yomntu iza kusherwa nenye iarhente karhulumente (engeyiyo eyobalo lwabantu).
- Amabango obuxoki nangaqinisekiswanga okuba i-Immigration and Customs Enforcement (ICE) yaseUnited States ikwindawo yonyulo.
- Amabango axoka ngokuphandle athi abantu baza kosulelwa ngeCOVID-19 (okanye ngesinye isifo esosulelayo) ukuba bathatha inxaxheba kwinkqubo yokuvota.
- Amabango obuxoki angeemeko yelo xesha yendawo ekuvotelwa kuyo eUnited States enokudala ukuba kungakwazeki tu ukuvota, aqinisekiswe ligunya lonyulo.
Sineepolisi ezingakumbi ezenzelwe ukugubungela ukuhlatywa kwekhwelo lokuba kwenziwe ubundlobongela, ukukhuthaza ukuthatha inxaxheba kwinto engekho mthethweni, nokuhlatywa kwekhwelo lokuba kusetyenziswane ekuphazamiseni unyulo, zinto ezo ezivelisiweyo kwamanye amacandelo eMigangatho Yoluntu yethu.
Kule khontenti elandelayo, siquka ileyibhuli enikela ngeenkcazelo:
Imidiya Esetyenziweyo
Imidiya iyakwazi ukuedithwa ngeendlela ezahlukeneyo. Kwiimeko ezininzi, olu tshintsho alunabuzaza, njengaxa ikhontenti isikiwe kwakho inxalenye engavelanga okanye yenziwe mfutshane ukuze ibe ntle okanye kufakelwe umculo. Kwezinye iimeko, kusenokungacaci ukuba isetyenziwe kwaye isengalahlekisa abantu.
- Ikhontenti Eyilwe Okanye Etshintshwe Ngedijithali Engalahlekisa Abantu. Kwikhontenti engayaphuliyo Imigangatho Yoluntu, singafaka ileyibhuli enenkcazelo phambi kwalo khontenti – okanye siyikhabe ikhontenti efakwe njengesibhengezo – xa loo khontenti ingumfanekiso okanye ividiyo yento yokwenene, iyiodiyo evakala iyeyokwenene, eyilwe okanye etshintshwe ngendlela yedijithali kwaye inyusa umngcipheko wokukhohlisa uwonkewonke kumba obalulekileyo kuwonkewonke.
8 uApr 2025
Policy Rationale
Misinformation is different from other types of speech addressed in our Community Standards because there is no way to articulate a comprehensive list of what is prohibited. With graphic violence or hateful conductspeech, for instance, our policies specify the contentspeech we prohibit, and even persons who disagree with those policies can follow them. With misinformation, however, we cannot provide such a line. The world is changing constantly, and what is true one minute may not be true the next minute. People also have different levels of information about the world around them, and may believe something is true when it is not. A policy that simply prohibits “misinformation” would not provide useful notice to the people who use our services and would be unenforceable, as we don’t have perfect access to information.
Instead, our policies articulate different categories of misinformation and try to provide clear guidance about how we treat that speech when we see it. For each category, our approach reflects our attempt to balance our values of expression, safety, dignity, authenticity, and privacy.
We remove misinformation where it is likely to directly contribute to the risk of imminent physical harm. We also remove content that is likely to directly contribute to interference with the functioning of political processes. In determining what constitutes misinformation in these categories, we partner with independent experts who possess knowledge and expertise to assess the truth of the content and whether it is likely to directly contribute to the risk of imminent harm. This includes, for instance, partnering with human rights organizations with a presence on the ground in a country to determine the truth of a rumor about civil conflict.
For all other misinformation, we focus on reducing its prevalence or creating an environment that fosters a productive dialogue. We know that people often use misinformation in harmless ways, such as to exaggerate a point (“This team has the worst record in the history of the sport!”) or in humor or satire (“My husband just won Husband of the Year.”) They also may share their experience through stories that contain inaccuracies. In some cases, people share deeply-held personal opinions that others consider false or share information that they believe to be true but others consider incomplete or misleading.
Recognizing how common such speech is, we focus on providingslowing the spread of hoaxes and viral misinformation, and directing users with helpfulto authoritative information when there is potentially misleading or confusing content. As part of that effort, outside of the United States we partner with third-party fact checking organizations in many parts of the world to review and rate the accuracy of the most viral content on our platforms (see here and here for our approach in the United States). Beginning in the United States, the Community Notes program lets people add more context to Facebook, Instagram and Threads posts that are potentially misleading or confusing (see here). We also provide resources to increase media and digital literacy so people can decide what to read, trust, and share themselves. We require people to disclose, using our AI-disclosure tool, whenever they post organic content with photorealistic video or realistic-sounding audio that was digitally created or altered, and we may apply penalties if they fail to do so. We may also add a label to certain digitally created or altered content that creates a particularly high risk of misleading people on a matter of public importance.
Finally, we prohibit content and behavior in other areas that often overlap with the spread of misinformation. For example, our Community Standards prohibit fake accounts, fraud, and coordinated inauthentic behavior.
As online and offline environments change and evolve, we will continue to evolve these policies. Accounts that repeatedly share the misinformation listed below may, in addition to having their content enforced on in accordance with this policy, receive decreased distribution, limitations on their ability to advertise, or be removed from our platforms. Additional information on what happens when Meta removes content can be found here.
Guidelines
Misinformation we remove:
We remove the following types of misinformation:
I. Physical Harm or Violence
We remove misinformation or unverifiable rumors that expert partners have determined are likely to directly contribute to a risk of imminent violence or physical harm to people. We define misinformation as content with a claim that is determined to be false by an authoritative third party. We define an unverifiable rumor as a claim whose source expert partners confirm is extremely hard or impossible to trace, for which authoritative sources are absent, where there is not enough specificity for the claim to be debunked, or where the claim is too incredulous or too irrational to be believed.
We know that sometimes misinformation that might appear benign could, in a specific context, contribute to a risk of offline harm, including threats of violence that could contribute to a heightened risk of death, serious injury, or other physical harm. We work with a global network of non-governmental organizations (NGOs), not-for-profit organizations, humanitarian organizations, and international organizations that have expertise in these local dynamics.
In countries experiencing a heightened risk of societal violence, we work proactively with local partners to understand which false claims may directly contribute to a risk of imminent physical harm. We then work to identify and remove content making those claims on our platform. For example, in consultation with local experts, we may remove out-of-context media falsely claiming to depict acts of violence, victims or perpetrators of violence, weapons, or military hardware.
II. Harmful Health Misinformation
We consult with leading health organizations to identify health misinformation likely to directly contribute to imminent harm to public health and safety. The harmful health misinformation that we remove includes the following:
- Misinformation about vaccines. We remove misinformation primarily about vaccines when public health authorities conclude that the information is false and likely to directly contribute to imminent vaccine refusals. They include:
- Vaccines cause autism (Ex: “Increased vaccinations are why so many kids have autism these days.”)
- Vaccines cause Sudden Infant Death Syndrome (Ex: “Don’t you know vaccines cause SIDS?”
- Vaccines cause the disease against which they are meant to protect, or cause the person receiving the vaccine to be more likely to get the disease (Ex: “Taking a vaccine actually makes you more likely to get the disease since there’s a strain of the disease inside. Beware!”)
- Vaccines or their ingredients are deadly, toxic, poisonous, harmful, or dangerous (Ex: “Sure, you can take vaccines, if you don’t mind putting poison in your body.”)
- Natural immunity is safer than vaccine-acquired immunity (Ex: “It’s safest to just get the disease rather than the vaccine.”)
- It is dangerous to get several vaccines in a short period of time, even if that timing is medically recommended (Ex: “Never take more than one vaccine at the same time, that is dangerous - I don’t care what your doctor tells you!”)
- Vaccines are not effective at preventing the disease against which they purport to protect. However, for the COVID-19, flu, and malaria vaccines, we do not remove claims that those vaccines are not effective in preventing someone from contracting those viruses. (Ex’s: Remove – “The polio vaccine doesn’t do anything to stop you from getting the disease”; Remove – “Vaccines actually don’t do anything to stop you from getting diseases”; Allow – “The vaccine doesn’t stop you from getting COVID-19, that’s why you still need to socially distance and wear a mask when you’re around others.”)
- Acquiring measles cannot cause death (requires additional information and/or context) (Ex: “Don’t worry about whether you get measles, it can’t be fatal.”)
- Vitamin C is as effective as vaccines in preventing diseases for which vaccines exist.
- Misinformation about health during public health emergencies. We remove misinformation during public health emergencies when public health authorities conclude that the information is false and likely to directly contribute to the risk of imminent physical harm, including by contributing to the risk of individuals getting or spreading a harmful disease or refusing an associated vaccine. We identify public health emergencies in partnership with global and local health authorities.
- Promoting or advocating for harmful miracle cures for health issues. These include treatments where the recommended application, in a health context, is likely to directly contribute to the risk of serious injury or death, and the treatment has no legitimate health use (ex: bleach, disinfectant, black salve, caustic soda).
III. Voter or Census Interference
In an effort to promote election and census integrity, we remove misinformation that is likely to directly contribute to a risk of interference with people’s ability to participate in those processes. This includes the following:
- Misinformation about the dates, locations, times, and methods for voting, voter registration, or census participation.
- Misinformation about who can vote, qualifications for voting, whether a vote will be counted, and what information or materials must be provided in order to vote.
- Misinformation about whether a candidate is running or not.
- Misinformation about who can participate in the census and what information or materials must be provided in order to participate.
- Misinformation about government involvement in the census, including, where applicable, that an individual's census information will be shared with another (non-census) government agency.
- False or unverified claims that the U.S. Immigration and Customs Enforcement (ICE) is at a voting location.
- Explicit false claims that people will be infected by COVID-19 (or another communicable disease) if they participate in the voting process.
- False claims about current conditions at a U.S. voting location that would make it impossible to vote, as verified by an election authority.
We have additional policies intended to cover calls for violence, the promotion of illegal participation, and calls for coordinated interference in elections, which are represented in other sections of our Community Standards.
For the following content, we include an informative label:
Manipulated Media
Media can be edited in a variety of ways. In many cases, these changes are benign, such as content being cropped or shortened for artistic reasons or music being added. In other cases, the manipulation is not apparent and could mislead.
- Content Digitally Created or Altered that May Mislead. For content that does not otherwise violate the Community Standards, we may place an informative label on the face of content – or reject content submitted as an advertisement – when the content is a photorealistic image or video, or realistic sounding audio, that was digitally created or altered and creates a particularly high risk of materially deceiving the public on a matter of public importance.
14 uMatshi 2025
Policy Rationale
Misinformation is different from other types of speech addressed in our Community Standards because there is no way to articulate a comprehensive list of what is prohibited. With graphic violence or hate speech, for instance, our policies specify the speech we prohibit, and even persons who disagree with those policies can follow them. With misinformation, however, we cannot provide such a line. The world is changing constantly, and what is true one minute may not be true the next minute. People also have different levels of information about the world around them, and may believe something is true when it is not. A policy that simply prohibits “misinformation” would not provide useful notice to the people who use our services and would be unenforceable, as we don’t have perfect access to information.
Instead, our policies articulate different categories of misinformation and try to provide clear guidance about how we treat that speech when we see it. For each category, our approach reflects our attempt to balance our values of expression, safety, dignity, authenticity, and privacy.
We remove misinformation where it is likely to directly contribute to the risk of imminent physical harm. We also remove content that is likely to directly contribute to interference with the functioning of political processes. In determining what constitutes misinformation in these categories, we partner with independent experts who possess knowledge and expertise to assess the truth of the content and whether it is likely to directly contribute to the risk of imminent harm. This includes, for instance, partnering with human rights organizations with a presence on the ground in a country to determine the truth of a rumor about civil conflict.
For all other misinformation, we focus on reducing its prevalence or creating an environment that fosters a productive dialogue. We know that people often use misinformation in harmless ways, such as to exaggerate a point (“This team has the worst record in the history of the sport!”) or in humor or satire (“My husband just won Husband of the Year.”) They also may share their experience through stories that contain inaccuracies. In some cases, people share deeply-held personal opinions that others consider false or share information that they believe to be true but others consider incomplete or misleading.
Recognizing how common such speech is, we focus on slowing the spread of hoaxes and viral misinformation, and directing users to authoritative information. As part of that effort, outside of the United States we partner with third-party fact checking organizations in many parts of the world to review and rate the accuracy of the most viral content on our platforms (see here and here for our approach in the United States). Beginning in the United States, the Community Notes program lets people add more context to Facebook, Instagram and Threads posts that are potentially misleading or confusing (see here). We also provide resources to increase media and digital literacy so people can decide what to read, trust, and share themselves. We require people to disclose, using our AI-disclosure tool, whenever they post organic content with photorealistic video or realistic-sounding audio that was digitally created or altered, and we may apply penalties if they fail to do so. We may also add a label to certain digitally created or altered content that creates a particularly high risk of misleading people on a matter of public importance.
Finally, we prohibit content and behavior in other areas that often overlap with the spread of misinformation. For example, our Community Standards prohibit fake accounts, fraud, and coordinated inauthentic behavior.
As online and offline environments change and evolve, we will continue to evolve these policies. Accounts that repeatedly share the misinformation listed below may, in addition to having their content enforced on in accordance with this policy, receive decreased distribution, limitations on their ability to advertise, or be removed from our platforms. Additional information on what happens when Meta removes content can be found here.
Guidelines
Misinformation we remove:
We remove the following types of misinformation:
I. Physical Harm or Violence
We remove misinformation or unverifiable rumors that expert partners have determined are likely to directly contribute to a risk of imminent violence or physical harm to people. We define misinformation as content with a claim that is determined to be false by an authoritative third party. We define an unverifiable rumor as a claim whose source expert partners confirm is extremely hard or impossible to trace, for which authoritative sources are absent, where there is not enough specificity for the claim to be debunked, or where the claim is too incredulous or too irrational to be believed.
We know that sometimes misinformation that might appear benign could, in a specific context, contribute to a risk of offline harm, including threats of violence that could contribute to a heightened risk of death, serious injury, or other physical harm. We work with a global network of non-governmental organizations (NGOs), not-for-profit organizations, humanitarian organizations, and international organizations that have expertise in these local dynamics.
In countries experiencing a heightened risk of societal violence, we work proactively with local partners to understand which false claims may directly contribute to a risk of imminent physical harm. We then work to identify and remove content making those claims on our platform. For example, in consultation with local experts, we may remove out-of-context media falsely claiming to depict acts of violence, victims or perpetrators of violence, weapons, or military hardware.
II. Harmful Health Misinformation
We consult with leading health organizations to identify health misinformation likely to directly contribute to imminent harm to public health and safety. The harmful health misinformation that we remove includes the following:
- Misinformation about vaccines. We remove misinformation primarily about vaccines when public health authorities conclude that the information is false and likely to directly contribute to imminent vaccine refusals. They include:
- Vaccines cause autism (Ex: “Increased vaccinations are why so many kids have autism these days.”)
- Vaccines cause Sudden Infant Death Syndrome (Ex: “Don’t you know vaccines cause SIDS?”
- Vaccines cause the disease against which they are meant to protect, or cause the person receiving the vaccine to be more likely to get the disease (Ex: “Taking a vaccine actually makes you more likely to get the disease since there’s a strain of the disease inside. Beware!”)
- Vaccines or their ingredients are deadly, toxic, poisonous, harmful, or dangerous (Ex: “Sure, you can take vaccines, if you don’t mind putting poison in your body.”)
- Natural immunity is safer than vaccine-acquired immunity (Ex: “It’s safest to just get the disease rather than the vaccine.”)
- It is dangerous to get several vaccines in a short period of time, even if that timing is medically recommended (Ex: “Never take more than one vaccine at the same time, that is dangerous - I don’t care what your doctor tells you!”)
- Vaccines are not effective at preventing the disease against which they purport to protect. However, for the COVID-19, flu, and malaria vaccines, we do not remove claims that those vaccines are not effective in preventing someone from contracting those viruses. (Ex’s: Remove – “The polio vaccine doesn’t do anything to stop you from getting the disease”; Remove – “Vaccines actually don’t do anything to stop you from getting diseases”; Allow – “The vaccine doesn’t stop you from getting COVID-19, that’s why you still need to socially distance and wear a mask when you’re around others.”)
- Acquiring measles cannot cause death (requires additional information and/or context) (Ex: “Don’t worry about whether you get measles, it can’t be fatal.”)
- Vitamin C is as effective as vaccines in preventing diseases for which vaccines exist.
- Misinformation about health during public health emergencies. We remove misinformation during public health emergencies when public health authorities conclude that the information is false and likely to directly contribute to the risk of imminent physical harm, including by contributing to the risk of individuals getting or spreading a harmful disease or refusing an associated vaccine. We identify public health emergencies in partnership with global and local health authorities.
- Promoting or advocating for harmful miracle cures for health issues. These include treatments where the recommended application, in a health context, is likely to directly contribute to the risk of serious injury or death, and the treatment has no legitimate health use (ex: bleach, disinfectant, black salve, caustic soda).
III. Voter or Census Interference
In an effort to promote election and census integrity, we remove misinformation that is likely to directly contribute to a risk of interference with people’s ability to participate in those processes. This includes the following:
- Misinformation about the dates, locations, times, and methods for voting, voter registration, or census participation.
- Misinformation about who can vote, qualifications for voting, whether a vote will be counted, and what information or materials must be provided in order to vote.
- Misinformation about whether a candidate is running or not.
- Misinformation about who can participate in the census and what information or materials must be provided in order to participate.
- Misinformation about government involvement in the census, including, where applicable, that an individual's census information will be shared with another (non-census) government agency.
- False or unverified claims that the U.S. Immigration and Customs Enforcement (ICE) is at a voting location.
- Explicit false claims that people will be infected by COVID-19 (or another communicable disease) if they participate in the voting process.
- False claims about current conditions at a U.S. voting location that would make it impossible to vote, as verified by an election authority.
We have additional policies intended to cover calls for violence, the promotion of illegal participation, and calls for coordinated interference in elections, which are represented in other sections of our Community Standards.
For the following content, we include an informative label:
Manipulated Media
Media can be edited in a variety of ways. In many cases, these changes are benign, such as content being cropped or shortened for artistic reasons or music being added. In other cases, the manipulation is not apparent and could mislead.
- Content Digitally Created or Altered that May Mislead. For content that does not otherwise violate the Community Standards, we may place an informative label on the face of content – or reject content submitted as an advertisement – when the content is a photorealistic image or video, or realistic sounding audio, that was digitally created or altered and creates a particularly high risk of materially deceiving the public on a matter of public importance.
8 Januwari 2025
Policy Rationale
Misinformation is different from other types of speech addressed in our Community Standards because there is no way to articulate a comprehensive list of what is prohibited. With graphic violence or hate speech, for instance, our policies specify the speech we prohibit, and even persons who disagree with those policies can follow them. With misinformation, however, we cannot provide such a line. The world is changing constantly, and what is true one minute may not be true the next minute. People also have different levels of information about the world around them, and may believe something is true when it is not. A policy that simply prohibits “misinformation” would not provide useful notice to the people who use our services and would be unenforceable, as we don’t have perfect access to information.
Instead, our policies articulate different categories of misinformation and try to provide clear guidance about how we treat that speech when we see it. For each category, our approach reflects our attempt to balance our values of expression, safety, dignity, authenticity, and privacy.
We remove misinformation where it is likely to directly contribute to the risk of imminent physical harm. We also remove content that is likely to directly contribute to interference with the functioning of political processes. In determining what constitutes misinformation in these categories, we partner with independent experts who possess knowledge and expertise to assess the truth of the content and whether it is likely to directly contribute to the risk of imminent harm. This includes, for instance, partnering with human rights organizations with a presence on the ground in a country to determine the truth of a rumor about civil conflict.
For all other misinformation, we focus on reducing its prevalence or creating an environment that fosters a productive dialogue. We know that people often use misinformation in harmless ways, such as to exaggerate a point (“This team has the worst record in the history of the sport!”) or in humor or satire (“My husband just won Husband of the Year.”) They also may share their experience through stories that contain inaccuracies. In some cases, people share deeply-held personal opinions that others consider false or share information that they believe to be true but others consider incomplete or misleading.
Recognizing how common such speech is, we focus on slowing the spread of hoaxes and viral misinformation, and directing users to authoritative information. As part of that effort, we partner with third-party fact checking organizations in many parts of the world to review and rate the accuracy of the most viral content on our platforms (see hereto learn more about how our fact-checking program works)and here for our approach in the United States). We also provide resources to increase media and digital literacy so people can decide what to read, trust, and share themselves. We require people to disclose, using our AI-disclosure tool, whenever they post organic content with photorealistic video or realistic-sounding audio that was digitally created or altered, and we may apply penalties if they fail to do so. We may also add a label to certain digitally created or altered content that creates a particularly high risk of misleading people on a matter of public importance.
Finally, we prohibit content and behavior in other areas that often overlap with the spread of misinformation. For example, our Community Standards prohibit fake accounts, fraud, and coordinated inauthentic behavior.
As online and offline environments change and evolve, we will continue to evolve these policies. Accounts that repeatedly share the misinformation listed below may, in addition to having their content enforced on in accordance with this policy, receive decreased distribution, limitations on their ability to advertise, or be removed from our platforms. Additional information on what happens when Meta removes content can be found here.
Guidelines
Misinformation we remove:
We remove the following types of misinformation:
I. Physical Harm or Violence
We remove misinformation or unverifiable rumors that expert partners have determined are likely to directly contribute to a risk of imminent violence or physical harm to people. We define misinformation as content with a claim that is determined to be false by an authoritative third party. We define an unverifiable rumor as a claim whose source expert partners confirm is extremely hard or impossible to trace, for which authoritative sources are absent, where there is not enough specificity for the claim to be debunked, or where the claim is too incredulous or too irrational to be believed.
We know that sometimes misinformation that might appear benign could, in a specific context, contribute to a risk of offline harm, including threats of violence that could contribute to a heightened risk of death, serious injury, or other physical harm. We work with a global network of non-governmental organizations (NGOs), not-for-profit organizations, humanitarian organizations, and international organizations that have expertise in these local dynamics.
In countries experiencing a heightened risk of societal violence, we work proactively with local partners to understand which false claims may directly contribute to a risk of imminent physical harm. We then work to identify and remove content making those claims on our platform. For example, in consultation with local experts, we may remove out-of-context media falsely claiming to depict acts of violence, victims or perpetrators of violence, weapons, or military hardware.
II. Harmful Health Misinformation
We consult with leading health organizations to identify health misinformation likely to directly contribute to imminent harm to public health and safety. The harmful health misinformation that we remove includes the following:
- Misinformation about vaccines. We remove misinformation primarily about vaccines when public health authorities conclude that the information is false and likely to directly contribute to imminent vaccine refusals. They include:
- Vaccines cause autism (Ex: “Increased vaccinations are why so many kids have autism these days.”)
- Vaccines cause Sudden Infant Death Syndrome (Ex: “Don’t you know vaccines cause SIDS?”
- Vaccines cause the disease against which they are meant to protect, or cause the person receiving the vaccine to be more likely to get the disease (Ex: “Taking a vaccine actually makes you more likely to get the disease since there’s a strain of the disease inside. Beware!”)
- Vaccines or their ingredients are deadly, toxic, poisonous, harmful, or dangerous (Ex: “Sure, you can take vaccines, if you don’t mind putting poison in your body.”)
- Natural immunity is safer than vaccine-acquired immunity (Ex: “It’s safest to just get the disease rather than the vaccine.”)
- It is dangerous to get several vaccines in a short period of time, even if that timing is medically recommended (Ex: “Never take more than one vaccine at the same time, that is dangerous - I don’t care what your doctor tells you!”)
- Vaccines are not effective at preventing the disease against which they purport to protect. However, for the COVID-19, flu, and malaria vaccines, we do not remove claims that those vaccines are not effective in preventing someone from contracting those viruses. (Ex’s: Remove – “The polio vaccine doesn’t do anything to stop you from getting the disease”; Remove – “Vaccines actually don’t do anything to stop you from getting diseases”; Allow – “The vaccine doesn’t stop you from getting COVID-19, that’s why you still need to socially distance and wear a mask when you’re around others.”)
- Acquiring measles cannot cause death (requires additional information and/or context) (Ex: “Don’t worry about whether you get measles, it can’t be fatal.”)
- Vitamin C is as effective as vaccines in preventing diseases for which vaccines exist.
- Misinformation about health during public health emergencies. We remove misinformation during public health emergencies when public health authorities conclude that the information is false and likely to directly contribute to the risk of imminent physical harm, including by contributing to the risk of individuals getting or spreading a harmful disease or refusing an associated vaccine. We identify public health emergencies in partnership with global and local health authorities.
- Promoting or advocating for harmful miracle cures for health issues. These include treatments where the recommended application, in a health context, is likely to directly contribute to the risk of serious injury or death, and the treatment has no legitimate health use (ex: bleach, disinfectant, black salve, caustic soda).
III. Voter or Census Interference
In an effort to promote election and census integrity, we remove misinformation that is likely to directly contribute to a risk of interference with people’s ability to participate in those processes. This includes the following:
- Misinformation about the dates, locations, times, and methods for voting, voter registration, or census participation.
- Misinformation about who can vote, qualifications for voting, whether a vote will be counted, and what information or materials must be provided in order to vote.
- Misinformation about whether a candidate is running or not.
- Misinformation about who can participate in the census and what information or materials must be provided in order to participate.
- Misinformation about government involvement in the census, including, where applicable, that an individual's census information will be shared with another (non-census) government agency.
- False or unverified claims that the U.S. Immigration and Customs Enforcement (ICE) is at a voting location.
- Explicit false claims that people will be infected by COVID-19 (or another communicable disease) if they participate in the voting process.
- False claims about current conditions at a U.S. voting location that would make it impossible to vote, as verified by an election authority.
We have additional policies intended to cover calls for violence, the promotion of illegal participation, and calls for coordinated interference in elections, which are represented in other sections of our Community Standards.
For the following content, we include an informative label:
Manipulated Media
Media can be edited in a variety of ways. In many cases, these changes are benign, such as content being cropped or shortened for artistic reasons or music being added. In other cases, the manipulation is not apparent and could mislead.
- Content Digitally Created or Altered that May Mislead. For content that does not otherwise violate the Community Standards, we may place an informative label on the face of content – or reject content submitted as an advertisement – when the content is a photorealistic image or video, or realistic sounding audio, that was digitally created or altered and creates a particularly high risk of materially deceiving the public on a matter of public importance.
19 Julayi 2024
Policy Rationale
Misinformation is different from other types of speech addressed in our Community Standards because there is no way to articulate a comprehensive list of what is prohibited. With graphic violence or hate speech, for instance, our policies specify the speech we prohibit, and even persons who disagree with those policies can follow them. With misinformation, however, we cannot provide such a line. The world is changing constantly, and what is true one minute may not be true the next minute. People also have different levels of information about the world around them, and may believe something is true when it is not. A policy that simply prohibits “misinformation” would not provide useful notice to the people who use our services and would be unenforceable, as we don’t have perfect access to information.
Instead, our policies articulate different categories of misinformation and try to provide clear guidance about how we treat that speech when we see it. For each category, our approach reflects our attempt to balance our values of expression, safety, dignity, authenticity, and privacy.
We remove misinformation where it is likely to directly contribute to the risk of imminent physical harm. We also remove content that is likely to directly contribute to interference with the functioning of political processes. In determining what constitutes misinformation in these categories, we partner with independent experts who possess knowledge and expertise to assess the truth of the content and whether it is likely to directly contribute to the risk of imminent harm. This includes, for instance, partnering with human rights organizations with a presence on the ground in a country to determine the truth of a rumor about civil conflict, and partnering with health organizations during the global COVID-19 pandemic.
For all other misinformation, we focus on reducing its prevalence or creating an environment that fosters a productive dialogue. We know that people often use misinformation in harmless ways, such as to exaggerate a point (“This team has the worst record in the history of the sport!”) or in humor or satire (“My husband just won Husband of the Year.”) They also may share their experience through stories that contain inaccuracies. In some cases, people share deeply-held personal opinions that others consider false or share information that they believe to be true but others consider incomplete or misleading.
Recognizing how common such speech is, we focus on slowing the spread of hoaxes and viral misinformation, and directing users to authoritative information. As part of that effort, we partner with third-party fact checking organizations to review and rate the accuracy of the most viral content on our platforms (see here to learn more about how our fact-checking program works). We also provide resources to increase media and digital literacy so people can decide what to read, trust, and share themselves. We require people to disclose, using our AI-disclosure tool, whenever they post organic content with photorealistic video or realistic-sounding audio that was digitally created or altered, and we may apply penalties if they fail to do so. We may also add a label to certain digitally created or altered content that creates a particularly high risk of misleading people on a matter of public importance.
Finally, we prohibit content and behavior in other areas that often overlap with the spread of misinformation. For example, our Community Standards prohibit fake accounts, fraud, and coordinated inauthentic behavior.
As online and offline environments change and evolve, we will continue to evolve these policies. Pages, Groups, Profiles, and Instagram accountsAccounts that repeatedly share the misinformation listed below may, in addition to having their content removedenforced on in accordance with this policy, receive decreased distribution, limitations on their ability to advertise, or be removed from our platforms. Additional information on what happens when FacebookMeta removes content can be found here.
Guidelines
Misinformation we remove:
We remove the following types of misinformation:
I. Physical Harm or Violence
We remove misinformation or unverifiable rumors that expert partners have determined are likely to directly contribute to a risk of imminent violence or physical harm to people. We define misinformation as content with a claim that is determined to be false by an authoritative third party. We define an unverifiable rumor as a claim whose source expert partners confirm is extremely hard or impossible to trace, for which authoritative sources are absent, where there is not enough specificity for the claim to be debunked, or where the claim is too incredulous or too irrational to be believed.
We know that sometimes misinformation that might appear benign could, in a specific context, contribute to a risk of offline harm, including threats of violence that could contribute to a heightened risk of death, serious injury, or other physical harm. We work with a global network of non-governmental organizations (NGOs), not-for-profit organizations, humanitarian organizations, and international organizations that have expertise in these local dynamics.
In countries experiencing a heightened risk of societal violence, we work proactively with local partners to understand which false claims may directly contribute to a risk of imminent physical harm. We then work to identify and remove content making those claims on our platform. For example, in consultation with local experts, we may remove out-of-context media falsely claiming to depict acts of violence, victims or perpetrators of violence, weapons, or military hardware.
II. Harmful Health Misinformation
We consult with leading health organizations to identify health misinformation likely to directly contribute to imminent harm to public health and safety. The harmful health misinformation that we remove includes the following:
- Misinformation about vaccines. We remove misinformation primarily about vaccines when public health authorities conclude that the information is false and likely to directly contribute to imminent vaccine refusals. They include:
- Vaccines cause autism (Ex: “Increased vaccinations are why so many kids have autism these days.”)
- Vaccines cause Sudden Infant Death Syndrome (Ex: “Don’t you know vaccines cause SIDS?”
- Vaccines cause the disease against which they are meant to protect, or cause the person receiving the vaccine to be more likely to get the disease (Ex: “Taking a vaccine actually makes you more likely to get the disease since there’s a strain of the disease inside. Beware!”)
- Vaccines or their ingredients are deadly, toxic, poisonous, harmful, or dangerous (Ex: “Sure, you can take vaccines, if you don’t mind putting poison in your body.”)
- Natural immunity is safer than vaccine-acquired immunity (Ex: “It’s safest to just get the disease rather than the vaccine.”)
- It is dangerous to get several vaccines in a short period of time, even if that timing is medically recommended (Ex: “Never take more than one vaccine at the same time, that is dangerous - I don’t care what your doctor tells you!”)
- Vaccines are not effective at preventing the disease against which they purport to protect. However, for the COVID-19, flu, and malaria vaccines, we do not remove claims that those vaccines are not effective in preventing someone from contracting those viruses. (Ex’s: Remove – “The polio vaccine doesn’t do anything to stop you from getting the disease”; Remove – “Vaccines actually don’t do anything to stop you from getting diseases”; Allow – “The vaccine doesn’t stop you from getting COVID-19, that’s why you still need to socially distance and wear a mask when you’re around others.”)
- Acquiring measles cannot cause death (requires additional information and/or context) (Ex: “Don’t worry about whether you get measles, it can’t be fatal.”)
- Vitamin C is as effective as vaccines in preventing diseases for which vaccines exist.
- Misinformation about health during public health emergencies. We remove misinformation during public health emergencies when public health authorities conclude that the information is false and likely to directly contribute to the risk of imminent physical harm, including by contributing to the risk of individuals getting or spreading a harmful disease or refusing an associated vaccine. We identify public health emergencies in partnership with global and local health authorities.
- Promoting or advocating for harmful miracle cures for health issues. These include treatments where the recommended application, in a health context, is likely to directly contribute to the risk of serious injury or death, and the treatment has no legitimate health use (ex: bleach, disinfectant, black salve, caustic soda).
III. Voter or Census Interference
In an effort to promote election and census integrity, we remove misinformation that is likely to directly contribute to a risk of interference with people’s ability to participate in those processes. This includes the following:
- Misinformation about the dates, locations, times, and methods for voting, voter registration, or census participation.
- Misinformation about who can vote, qualifications for voting, whether a vote will be counted, and what information or materials must be provided in order to vote.
- Misinformation about whether a candidate is running or not.
- Misinformation about who can participate in the census and what information or materials must be provided in order to participate.
- Misinformation about government involvement in the census, including, where applicable, that an individual's census information will be shared with another (non-census) government agency.
- False or unverified claims that the U.S. Immigration and Customs Enforcement (ICE) is at a voting location.
- Explicit false claims that people will be infected by COVID-19 (or another communicable disease) if they participate in the voting process.
- False claims about current conditions at a U.S. voting location that would make it impossible to vote, as verified by an election authority.
We have additional policies intended to cover calls for violence, the promotion of illegal participation, and calls for coordinated interference in elections, which are represented in other sections of our Community Standards.
For the following content, we include an informative label:
Manipulated Media
Media can be edited in a variety of ways. In many cases, these changes are benign, such as content being cropped or shortened for artistic reasons or music being added. In other cases, the manipulation is not apparent and could mislead.
- Content Digitally Created or Altered that May Mislead. For content that does not otherwise violate the Community Standards, we may place an informative label on the face of content – or reject content submitted as an advertisement – when the content is a photorealistic image or video, or realistic sounding audio, that was digitally created or altered and creates a particularly high risk of materially deceiving the public on a matter of public importance.
10 Julayi 2024
Policy Rationale
Misinformation is different from other types of speech addressed in our Community Standards because there is no way to articulate a comprehensive list of what is prohibited. With graphic violence or hate speech, for instance, our policies specify the speech we prohibit, and even persons who disagree with those policies can follow them. With misinformation, however, we cannot provide such a line. The world is changing constantly, and what is true one minute may not be true the next minute. People also have different levels of information about the world around them, and may believe something is true when it is not. A policy that simply prohibits “misinformation” would not provide useful notice to the people who use our services and would be unenforceable, as we don’t have perfect access to information.
Instead, our policies articulate different categories of misinformation and try to provide clear guidance about how we treat that speech when we see it. For each category, our approach reflects our attempt to balance our values of expression, safety, dignity, authenticity, and privacy.
We remove misinformation where it is likely to directly contribute to the risk of imminent physical harm. We also remove content that is likely to directly contribute to interference with the functioning of political processes and certain highly deceptive manipulated media. In determining what constitutes misinformation in these categories, we partner with independent experts who possess knowledge and expertise to assess the truth of the content and whether it is likely to directly contribute to the risk of imminent harm. This includes, for instance, partnering with human rights organizations with a presence on the ground in a country to determine the truth of a rumor about civil conflict, and partnering with health organizations during the global COVID-19 pandemic.
For all other misinformation, we focus on reducing its prevalence or creating an environment that fosters a productive dialogue. We know that people often use misinformation in harmless ways, such as to exaggerate a point (“This team has the worst record in the history of the sport!”) or in humor or satire (“My husband just won Husband of the Year.”) They also may share their experience through stories that contain inaccuracies. In some cases, people share deeply-held personal opinions that others consider false or share information that they believe to be true but others consider incomplete or misleading.
Recognizing how common such speech is, we focus on slowing the spread of hoaxes and viral misinformation, and directing users to authoritative information. As part of that effort, we partner with third-party fact checking organizations to review and rate the accuracy of the most viral content on our platforms (see here to learn more about how our fact-checking program works). We also provide resources to increase media and digital literacy so people can decide what to read, trust, and share themselves. We require people to disclose, using our AI-disclosure tool, whenever they post organic content with photorealistic video or realistic-sounding audio that was digitally created or altered, and we may apply penalties if they fail to do so. We may also add a label to certain digitally created or altered content that creates a particularly high risk of misleading people on a matter of public importance.
Finally, we prohibit content and behavior in other areas that often overlap with the spread of misinformation. For example, our Community Standards prohibit fake accounts, fraud, and coordinated inauthentic behavior.
As online and offline environments change and evolve, we will continue to evolve these policies. Pages, Groups, Profiles, and Instagram accounts that repeatedly share the misinformation listed below may, in addition to having their content removed, receive decreased distribution, limitations on their ability to advertise, or be removed from our platforms. Additional information on what happens when Facebook removes content can be found here.
Misinformation we remove:
We remove the following types of misinformation:
I. Physical Harm or Violence
We remove misinformation or unverifiable rumors that expert partners have determined are likely to directly contribute to a risk of imminent violence or physical harm to people. We define misinformation as content with a claim that is determined to be false by an authoritative third party. We define an unverifiable rumor as a claim whose source expert partners confirm is extremely hard or impossible to trace, for which authoritative sources are absent, where there is not enough specificity for the claim to be debunked, or where the claim is too incredulous or too irrational to be believed.
We know that sometimes misinformation that might appear benign could, in a specific context, contribute to a risk of offline harm, including threats of violence that could contribute to a heightened risk of death, serious injury, or other physical harm. We work with a global network of non-governmental organizations (NGOs), not-for-profit organizations, humanitarian organizations, and international organizations that have expertise in these local dynamics.
In countries experiencing a heightened risk of societal violence, we work proactively with local partners to understand which false claims may directly contribute to a risk of imminent physical harm. We then work to identify and remove content making those claims on our platform. For example, in consultation with local experts, we may remove out-of-context media falsely claiming to depict acts of violence, victims or perpetrators of violence, weapons, or military hardware.
II. Harmful Health Misinformation
We consult with leading health organizations to identify health misinformation likely to directly contribute to imminent harm to public health and safety. The harmful health misinformation that we remove includes the following:
- Misinformation about vaccines. We remove misinformation primarily about vaccines when public health authorities conclude that the information is false and likely to directly contribute to imminent vaccine refusals. They include:
- Vaccines cause autism (Ex: “Increased vaccinations are why so many kids have autism these days.”)
- Vaccines cause Sudden Infant Death Syndrome (Ex: “Don’t you know vaccines cause SIDS?”
- Vaccines cause the disease against which they are meant to protect, or cause the person receiving the vaccine to be more likely to get the disease (Ex: “Taking a vaccine actually makes you more likely to get the disease since there’s a strain of the disease inside. Beware!”)
- Vaccines or their ingredients are deadly, toxic, poisonous, harmful, or dangerous (Ex: “Sure, you can take vaccines, if you don’t mind putting poison in your body.”)
- Natural immunity is safer than vaccine-acquired immunity (Ex: “It’s safest to just get the disease rather than the vaccine.”)
- It is dangerous to get several vaccines in a short period of time, even if that timing is medically recommended (Ex: “Never take more than one vaccine at the same time, that is dangerous - I don’t care what your doctor tells you!”)
- Vaccines are not effective at preventing the disease against which they purport to protect. However, for the COVID-19, flu, and malaria vaccines, we do not remove claims that those vaccines are not effective in preventing someone from contracting those viruses. (Ex’s: Remove – “The polio vaccine doesn’t do anything to stop you from getting the disease”; Remove – “Vaccines actually don’t do anything to stop you from getting diseases”; Allow – “The vaccine doesn’t stop you from getting COVID-19, that’s why you still need to socially distance and wear a mask when you’re around others.”)
- Acquiring measles cannot cause death (requires additional information and/or context) (Ex: “Don’t worry about whether you get measles, it can’t be fatal.”)
- Vitamin C is as effective as vaccines in preventing diseases for which vaccines exist.
- Misinformation about health during public health emergencies. We remove misinformation during public health emergencies when public health authorities conclude that the information is false and likely to directly contribute to the risk of imminent physical harm, including by contributing to the risk of individuals getting or spreading a harmful disease or refusing an associated vaccine. We identify public health emergencies in partnership with global and local health authorities.
- Promoting or advocating for harmful miracle cures for health issues. These include treatments where the recommended application, in a health context, is likely to directly contribute to the risk of serious injury or death, and the treatment has no legitimate health use (ex: bleach, disinfectant, black salve, caustic soda).
III. Voter or Census Interference
In an effort to promote election and census integrity, we remove misinformation that is likely to directly contribute to a risk of interference with people’s ability to participate in those processes. This includes the following:
- Misinformation about the dates, locations, times, and methods for voting, voter registration, or census participation.
- Misinformation about who can vote, qualifications for voting, whether a vote will be counted, and what information or materials must be provided in order to vote.
- Misinformation about whether a candidate is running or not.
- Misinformation about who can participate in the census and what information or materials must be provided in order to participate.
- Misinformation about government involvement in the census, including, where applicable, that an individual's census information will be shared with another (non-census) government agency.
- False or unverified claims that the U.S. Immigration and Customs Enforcement (ICE) is at a voting location.
- Explicit false claims that people will be infected by COVID-19 (or another communicable disease) if they participate in the voting process.
- False claims about current conditions at a U.S. voting location that would make it impossible to vote, as verified by an election authority.
We have additional policies intended to cover calls for violence, the promotion of illegal participation, and calls for coordinated interference in elections, which are represented in other sections of our Community Standards.
For the following content, we include an informative label:
Manipulated Media
Media can be edited in a variety of ways. In many cases, these changes are benign, such as content being cropped or shortened for artistic reasons or music being added. In other cases, the manipulation is not apparent and could mislead.
- Content Digitally Created or Altered that May Mislead. For content that does not otherwise violate the Community Standards, we may place an informative label on the face of content – or reject content submitted as an advertisement – when the content is a photorealistic image or video, or realistic sounding audio, that was digitally created or altered and creates a particularly high risk of materially deceiving the public on a matter of public importance.
11 Juni 2024
Policy Rationale
Misinformation is different from other types of speech addressed in our Community Standards because there is no way to articulate a comprehensive list of what is prohibited. With graphic violence or hate speech, for instance, our policies specify the speech we prohibit, and even persons who disagree with those policies can follow them. With misinformation, however, we cannot provide such a line. The world is changing constantly, and what is true one minute may not be true the next minute. People also have different levels of information about the world around them, and may believe something is true when it is not. A policy that simply prohibits “misinformation” would not provide useful notice to the people who use our services and would be unenforceable, as we don’t have perfect access to information.
Instead, our policies articulate different categories of misinformation and try to provide clear guidance about how we treat that speech when we see it. For each category, our approach reflects our attempt to balance our values of expression, safety, dignity, authenticity, and privacy.
We remove misinformation where it is likely to directly contribute to the risk of imminent physical harm. We also remove content that is likely to directly contribute to interference with the functioning of political processes and certain highly deceptive manipulated media. In determining what constitutes misinformation in these categories, we partner with independent experts who possess knowledge and expertise to assess the truth of the content and whether it is likely to directly contribute to the risk of imminent harm. This includes, for instance, partnering with human rights organizations with a presence on the ground in a country to determine the truth of a rumor about civil conflict, and partnering with health organizations during the global COVID-19 pandemic.
For all other misinformation, we focus on reducing its prevalence or creating an environment that fosters a productive dialogue. We know that people often use misinformation in harmless ways, such as to exaggerate a point (“This team has the worst record in the history of the sport!”) or in humor or satire (“My husband just won Husband of the Year.”) They also may share their experience through stories that contain inaccuracies. In some cases, people share deeply-held personal opinions that others consider false or share information that they believe to be true but others consider incomplete or misleading.
Recognizing how common such speech is, we focus on slowing the spread of hoaxes and viral misinformation, and directing users to authoritative information. As part of that effort, we partner with third-party fact checking organizations to review and rate the accuracy of the most viral content on our platforms (see here to learn more about how our fact-checking program works). We also provide resources to increase media and digital literacy so people can decide what to read, trust, and share themselves. We require people to disclose, using our AI-disclosure tool, whenever they post organic content with photorealistic video or realistic-sounding audio that was digitally created or altered, and we may apply penalties if they fail to do so.We may also add a label to certain digitally created or altered content that creates a particularly high risk of misleading people on a matter of public importance.
Finally, we prohibit content and behavior in other areas that often overlap with the spread of misinformation. For example, our Community Standards prohibit fake accounts, fraud, and coordinated inauthentic behavior.
As online and offline environments change and evolve, we will continue to evolve these policies. Pages, Groups, Profiles, and Instagram accounts that repeatedly share the misinformation listed below may, in addition to having their content removed, receive decreased distribution, limitations on their ability to advertise, or be removed from our platforms. Additional information on what happens when Facebook removes content can be found here.
Misinformation we remove:
We remove the following types of misinformation:
I. Physical Harm or Violence
We remove misinformation or unverifiable rumors that expert partners have determined are likely to directly contribute to a risk of imminent violence or physical harm to people. We define misinformation as content with a claim that is determined to be false by an authoritative third party. We define an unverifiable rumor as a claim whose source expert partners confirm is extremely hard or impossible to trace, for which authoritative sources are absent, where there is not enough specificity for the claim to be debunked, or where the claim is too incredulous or too irrational to be believed.
We know that sometimes misinformation that might appear benign could, in a specific context, contribute to a risk of offline harm, including threats of violence that could contribute to a heightened risk of death, serious injury, or other physical harm. We work with a global network of non-governmental organizations (NGOs), not-for-profit organizations, humanitarian organizations, and international organizations that have expertise in these local dynamics.
In countries experiencing a heightened risk of societal violence, we work proactively with local partners to understand which false claims may directly contribute to a risk of imminent physical harm. We then work to identify and remove content making those claims on our platform. For example, in consultation with local experts, we may remove out-of-context media falsely claiming to depict acts of violence, victims or perpetrators of violence, weapons, or military hardware.
II. Harmful Health Misinformation
We consult with leading health organizations to identify health misinformation likely to directly contribute to imminent harm to public health and safety. The harmful health misinformation that we remove includes the following:
- Misinformation about vaccines. We remove misinformation primarily about vaccines when public health authorities conclude that the information is false and likely to directly contribute to imminent vaccine refusals. They include:
- Vaccines cause autism (Ex: “Increased vaccinations are why so many kids have autism these days.”)
- Vaccines cause Sudden Infant Death Syndrome (Ex: “Don’t you know vaccines cause SIDS?”
- Vaccines cause the disease against which they are meant to protect, or cause the person receiving the vaccine to be more likely to get the disease (Ex: “Taking a vaccine actually makes you more likely to get the disease since there’s a strain of the disease inside. Beware!”)
- Vaccines or their ingredients are deadly, toxic, poisonous, harmful, or dangerous (Ex: “Sure, you can take vaccines, if you don’t mind putting poison in your body.”)
- Natural immunity is safer than vaccine-acquired immunity (Ex: “It’s safest to just get the disease rather than the vaccine.”)
- It is dangerous to get several vaccines in a short period of time, even if that timing is medically recommended (Ex: “Never take more than one vaccine at the same time, that is dangerous - I don’t care what your doctor tells you!”)
- Vaccines are not effective at preventing the disease against which they purport to protect. However, for the COVID-19, flu, and malaria vaccines, we do not remove claims that those vaccines are not effective in preventing someone from contracting those viruses. (Ex’s: Remove – “The polio vaccine doesn’t do anything to stop you from getting the disease”; Remove – “Vaccines actually don’t do anything to stop you from getting diseases”; Allow – “The vaccine doesn’t stop you from getting COVID-19, that’s why you still need to socially distance and wear a mask when you’re around others.”)
- Acquiring measles cannot cause death (requires additional information and/or context) (Ex: “Don’t worry about whether you get measles, it can’t be fatal.”)
- Vitamin C is as effective as vaccines in preventing diseases for which vaccines exist.
- Misinformation about health during public health emergencies. We remove misinformation during public health emergencies when public health authorities conclude that the information is false and likely to directly contribute to the risk of imminent physical harm, including by contributing to the risk of individuals getting or spreading a harmful disease or refusing an associated vaccine. We identify public health emergencies in partnership with global and local health authorities. This currently includes certain false claims related to COVID-19 in countries still experiencing a publicly announced state of emergency regarding the virus. Click here for a complete set of rules regarding what misinformation we do not allow regarding COVID-19 and vaccines.
- Promoting or advocating for harmful miracle cures for health issues. These include treatments where the recommended application, in a health context, is likely to directly contribute to the risk of serious injury or death, and the treatment has no legitimate health use (ex: bleach, disinfectant, black salve, caustic soda).
III. Voter or Census Interference
In an effort to promote election and census integrity, we remove misinformation that is likely to directly contribute to a risk of interference with people’s ability to participate in those processes. This includes the following:
- Misinformation about the dates, locations, times, and methods for voting, voter registration, or census participation.
- Misinformation about who can vote, qualifications for voting, whether a vote will be counted, and what information or materials must be provided in order to vote.
- Misinformation about whether a candidate is running or not.
- Misinformation about who can participate in the census and what information or materials must be provided in order to participate.
- Misinformation about government involvement in the census, including, where applicable, that an individual's census information will be shared with another (non-census) government agency.
- False or unverified claims that the U.S. Immigration and Customs Enforcement (ICE) is at a voting location.
- Explicit false claims that people will be infected by COVID-19 (or another communicable disease) if they participate in the voting process.
- False claims about current conditions at a U.S. voting location that would make it impossible to vote, as verified by an election authority.
We have additional policies intended to cover calls for violence, the promotion of illegal participation, and calls for coordinated interference in elections, which are represented in other sections of our Community Standards.
IV. Manipulated Media
Media can be edited in a variety of ways. In many cases, these changes are benign, such as content being cropped or shortened for artistic reasons or music being added. In other cases, the manipulation is not apparent and could mislead, particularly in the case of video content. We remove this content because it can go viral quickly and experts advise that false beliefs regarding manipulated media often cannot be corrected through further discourse.
We remove videos under this policy if specific criteria are met: (1) the video has been edited or synthesized, beyond adjustments for clarity or quality, in ways that are not apparent to an average person, and would likely mislead an average person to believe a subject of the video said words that they did not say; and (2) the video is the product of artificial intelligence or machine learning, including deep learning techniques (e.g., a technical deepfake), that merges, combines, replaces, and/or superimposes content onto a video, creating a video that appears authentic.
For the following content, we include an informative label:
Content Digitally Created or Altered that May Mislead
For content that does not otherwise violate the Community Standards, we may place an informative label on the face of content – or reject content submitted as an advertisement – when the content is a photorealistic image or video, or realistic sounding audio, that was digitally created or altered and creates a particularly high risk of materially deceiving the public on a matter of public importance.
21 Meyi 2024
Policy Rationale
Misinformation is different from other types of speech addressed in our Community Standards because there is no way to articulate a comprehensive list of what is prohibited. With graphic violence or hate speech, for instance, our policies specify the speech we prohibit, and even persons who disagree with those policies can follow them. With misinformation, however, we cannot provide such a line. The world is changing constantly, and what is true one minute may not be true the next minute. People also have different levels of information about the world around them, and may believe something is true when it is not. A policy that simply prohibits “misinformation” would not provide useful notice to the people who use our services and would be unenforceable, as we don’t have perfect access to information.
Instead, our policies articulate different categories of misinformation and try to provide clear guidance about how we treat that speech when we see it. For each category, our approach reflects our attempt to balance our values of expression, safety, dignity, authenticity, and privacy.
We remove misinformation where it is likely to directly contribute to the risk of imminent physical harm. We also remove content that is likely to directly contribute to interference with the functioning of political processes and certain highly deceptive manipulated media. In determining what constitutes misinformation in these categories, we partner with independent experts who possess knowledge and expertise to assess the truth of the content and whether it is likely to directly contribute to the risk of imminent harm. This includes, for instance, partnering with human rights organizations with a presence on the ground in a country to determine the truth of a rumor about civil conflict, and partnering with health organizations during the global COVID-19 pandemic.
For all other misinformation, we focus on reducing its prevalence or creating an environment that fosters a productive dialogue. We know that people often use misinformation in harmless ways, such as to exaggerate a point (“This team has the worst record in the history of the sport!”) or in humor or satire (“My husband just won Husband of the Year.”) They also may share their experience through stories that contain inaccuracies. In some cases, people share deeply-held personal opinions that others consider false or share information that they believe to be true but others consider incomplete or misleading.
Recognizing how common such speech is, we focus on slowing the spread of hoaxes and viral misinformation, and directing users to authoritative information. As part of that effort, we partner with third-party fact checking organizations to review and rate the accuracy of the most viral content on our platforms (see here to learn more about how our fact-checking program works). We also provide resources to increase media and digital literacy so people can decide what to read, trust, and share themselves. We require people to disclose, using our AI-disclosure tool, whenever they post organic content with photorealistic video or realistic-sounding audio that was digitally created or altered, and we may apply penalties if they fail to do so.We may also add a label to certain digitally created or altered content that creates a particularly high risk of misleading people on a matter of public importance.
Finally, we prohibit content and behavior in other areas that often overlap with the spread of misinformation. For example, our Community Standards prohibit fake accounts, fraud, and coordinated inauthentic behavior.
As online and offline environments change and evolve, we will continue to evolve these policies. Pages, Groups, Profiles, and Instagram accounts that repeatedly share the misinformation listed below may, in addition to having their content removed, receive decreased distribution, limitations on their ability to advertise, or be removed from our platforms. Additional information on what happens when Facebook removes content can be found here.
Misinformation we remove:
We remove the following types of misinformation:
I. Physical Harm or Violence
We remove misinformation or unverifiable rumors that expert partners have determined are likely to directly contribute to a risk of imminent violence or physical harm to people. We define misinformation as content with a claim that is determined to be false by an authoritative third party. We define an unverifiable rumor as a claim whose source expert partners confirm is extremely hard or impossible to trace, for which authoritative sources are absent, where there is not enough specificity for the claim to be debunked, or where the claim is too incredulous or too irrational to be believed.
We know that sometimes misinformation that might appear benign could, in a specific context, contribute to a risk of offline harm, including threats of violence that could contribute to a heightened risk of death, serious injury, or other physical harm. We work with a global network of non-governmental organizations (NGOs), not-for-profit organizations, humanitarian organizations, and international organizations that have expertise in these local dynamics.
In countries experiencing a heightened risk of societal violence, we work proactively with local partners to understand which false claims may directly contribute to a risk of imminent physical harm. We then work to identify and remove content making those claims on our platform. For example, in consultation with local experts, we may remove out-of-context media falsely claiming to depict acts of violence, victims or perpetrators of violence, weapons, or military hardware.
II. Harmful Health Misinformation
We consult with leading health organizations to identify health misinformation likely to directly contribute to imminent harm to public health and safety. The harmful health misinformation that we remove includes the following:
- Misinformation about vaccines. We remove misinformation primarily about vaccines when public health authorities conclude that the information is false and likely to directly contribute to imminent vaccine refusals. They include:
- Vaccines cause autism (Ex: “Increased vaccinations are why so many kids have autism these days.”)
- Vaccines cause Sudden Infant Death Syndrome (Ex: “Don’t you know vaccines cause SIDS?”
- Vaccines cause the disease against which they are meant to protect, or cause the person receiving the vaccine to be more likely to get the disease (Ex: “Taking a vaccine actually makes you more likely to get the disease since there’s a strain of the disease inside. Beware!”)
- Vaccines or their ingredients are deadly, toxic, poisonous, harmful, or dangerous (Ex: “Sure, you can take vaccines, if you don’t mind putting poison in your body.”)
- Natural immunity is safer than vaccine-acquired immunity (Ex: “It’s safest to just get the disease rather than the vaccine.”)
- It is dangerous to get several vaccines in a short period of time, even if that timing is medically recommended (Ex: “Never take more than one vaccine at the same time, that is dangerous - I don’t care what your doctor tells you!”)
- Vaccines are not effective at preventing the disease against which they purport to protect. However, for the COVID-19, flu, and malaria vaccines, we do not remove claims that those vaccines are not effective in preventing someone from contracting those viruses. (Ex’s: Remove – “The polio vaccine doesn’t do anything to stop you from getting the disease”; Remove – “Vaccines actually don’t do anything to stop you from getting diseases”; Allow – “The vaccine doesn’t stop you from getting COVID-19, that’s why you still need to socially distance and wear a mask when you’re around others.”)
- Acquiring measles cannot cause death (requires additional information and/or context) (Ex: “Don’t worry about whether you get measles, it can’t be fatal.”)
- Vitamin C is as effective as vaccines in preventing diseases for which vaccines exist.
- Misinformation about health during public health emergencies. We remove misinformation during public health emergencies when public health authorities conclude that the information is false and likely to directly contribute to the risk of imminent physical harm, including by contributing to the risk of individuals getting or spreading a harmful disease or refusing an associated vaccine. We identify public health emergencies in partnership with global and local health authorities. This currently includes certain false claims related to COVID-19 in countries still experiencing a publicly announced state of emergency regarding the virus. Click here for a complete set of rules regarding what misinformation we do not allow regarding COVID-19 and vaccines.
- Promoting or advocating for harmful miracle cures for health issues. These include treatments where the recommended application, in a health context, is likely to directly contribute to the risk of serious injury or death, and the treatment has no legitimate health use (ex: bleach, disinfectant, black salve, caustic soda).
III. Voter or Census Interference
In an effort to promote election and census integrity, we remove misinformation that is likely to directly contribute to a risk of interference with people’s ability to participate in those processes. This includes the following:
- Misinformation about the dates, locations, times, and methods for voting, voter registration, or census participation.
- Misinformation about who can vote, qualifications for voting, whether a vote will be counted, and what information or materials must be provided in order to vote.
- Misinformation about whether a candidate is running or not.
- Misinformation about who can participate in the census and what information or materials must be provided in order to participate.
- Misinformation about government involvement in the census, including, where applicable, that an individual's census information will be shared with another (non-census) government agency.
- False or unverified claims that the U.S. Immigration and Customs Enforcement (ICE) is at a voting location.
- Explicit false claims that people will be infected by COVID-19 (or another communicable disease) if they participate in the voting process.
- False claims about current conditions at a U.S. voting location that would make it impossible to vote, as verified by an election authority.
We have additional policies intended to cover calls for violence, the promotion of illegal participation, and calls for coordinated interference in elections, which are represented in other sections of our Community Standards.
IV. Manipulated Media
Media can be edited in a variety of ways. In many cases, these changes are benign, such as content being cropped or shortened for artistic reasons or music being added. In other cases, the manipulation is not apparent and could mislead, particularly in the case of video content. We remove this content because it can go viral quickly and experts advise that false beliefs regarding manipulated media often cannot be corrected through further discourse.
We remove videos under this policy if specific criteria are met: (1) the video has been edited or synthesized, beyond adjustments for clarity or quality, in ways that are not apparent to an average person, and would likely mislead an average person to believe a subject of the video said words that they did not say; and (2) the video is the product of artificial intelligence or machine learning, including deep learning techniques (e.g., a technical deepfake), that merges, combines, replaces, and/or superimposes content onto a video, creating a video that appears authentic.
For the following content, we include an informative label:
Content Digitally Created or Altered that May Mislead
For content that does not otherwise violate the Community Standards, we may place an informative label on the face of content – or reject content submitted as an advertisement – when the content is a photorealistic image or video, or realistic sounding audio, that was digitally created or altered and creates a particularly high risk of materially deceiving the public on a matter of public importance.
7 Februwari 2024
Policy Rationale
Misinformation is different from other types of speech addressed in our Community Standards because there is no way to articulate a comprehensive list of what is prohibited. With graphic violence or hate speech, for instance, our policies specify the speech we prohibit, and even persons who disagree with those policies can follow them. With misinformation, however, we cannot provide such a line. The world is changing constantly, and what is true one minute may not be true the next minute. People also have different levels of information about the world around them, and may believe something is true when it is not. A policy that simply prohibits “misinformation” would not provide useful notice to the people who use our services and would be unenforceable, as we don’t have perfect access to information.
Instead, our policies articulate different categories of misinformation and try to provide clear guidance about how we treat that speech when we see it. For each category, our approach reflects our attempt to balance our values of expression, safety, dignity, authenticity, and privacy.
We remove misinformation where it is likely to directly contribute to the risk of imminent physical harm. We also remove content that is likely to directly contribute to interference with the functioning of political processes and certain highly deceptive manipulated media. In determining what constitutes misinformation in these categories, we partner with independent experts who possess knowledge and expertise to assess the truth of the content and whether it is likely to directly contribute to the risk of imminent harm. This includes, for instance, partnering with human rights organizations with a presence on the ground in a country to determine the truth of a rumor about civil conflict, and partnering with health organizations during the global COVID-19 pandemic.
For all other misinformation, we focus on reducing its prevalence or creating an environment that fosters a productive dialogue. We know that people often use misinformation in harmless ways, such as to exaggerate a point (“This team has the worst record in the history of the sport!”) or in humor or satire (“My husband just won Husband of the Year.”) They also may share their experience through stories that contain inaccuracies. In some cases, people share deeply-held personal opinions that others consider false or share information that they believe to be true but others consider incomplete or misleading.
Recognizing how common such speech is, we focus on slowing the spread of hoaxes and viral misinformation, and directing users to authoritative information. As part of that effort, we partner with third-party fact checking organizations to review and rate the accuracy of the most viral content on our platforms (see here to learn more about how our fact-checking program works). We also provide resources to increase media and digital literacy so people can decide what to read, trust, and share themselves. We may also add a label to certain digitally created or altered content that creates a particularly high risk of misleading people on a matter of public importance.
Finally, we prohibit content and behavior in other areas that often overlap with the spread of misinformation. For example, our Community Standards prohibit fake accounts, fraud, and coordinated inauthentic behavior.
As online and offline environments change and evolve, we will continue to evolve these policies. Pages, Groups, Profiles, and Instagram accounts that repeatedly share the misinformation listed below may, in addition to having their content removed, receive decreased distribution, limitations on their ability to advertise, or be removed from our platforms. Additional information on what happens when Facebook removes content can be found here.
Misinformation we remove:
We remove the following types of misinformation:
I. Physical Harm or Violence
We remove misinformation or unverifiable rumors that expert partners have determined are likely to directly contribute to a risk of imminent violence or physical harm to people. We define misinformation as content with a claim that is determined to be false by an authoritative third party. We define an unverifiable rumor as a claim whose source expert partners confirm is extremely hard or impossible to trace, for which authoritative sources are absent, where there is not enough specificity for the claim to be debunked, or where the claim is too incredulous or too irrational to be believed.
We know that sometimes misinformation that might appear benign could, in a specific context, contribute to a risk of offline harm, including threats of violence that could contribute to a heightened risk of death, serious injury, or other physical harm. We work with a global network of non-governmental organizations (NGOs), not-for-profit organizations, humanitarian organizations, and international organizations that have expertise in these local dynamics.
In countries experiencing a heightened risk of societal violence, we work proactively with local partners to understand which false claims may directly contribute to a risk of imminent physical harm. We then work to identify and remove content making those claims on our platform. For example, in consultation with local experts, we may remove out-of-context media falsely claiming to depict acts of violence, victims or perpetrators of violence, weapons, or military hardware.
II. Harmful Health Misinformation
We consult with leading health organizations to identify health misinformation likely to directly contribute to imminent harm to public health and safety. The harmful health misinformation that we remove includes the following:
- Misinformation about vaccines. We remove misinformation primarily about vaccines when public health authorities conclude that the information is false and likely to directly contribute to imminent vaccine refusals. They include:
- Vaccines cause autism (Ex: “Increased vaccinations are why so many kids have autism these days.”)
- Vaccines cause Sudden Infant Death Syndrome (Ex: “Don’t you know vaccines cause SIDS?”
- Vaccines cause the disease against which they are meant to protect, or cause the person receiving the vaccine to be more likely to get the disease (Ex: “Taking a vaccine actually makes you more likely to get the disease since there’s a strain of the disease inside. Beware!”)
- Vaccines or their ingredients are deadly, toxic, poisonous, harmful, or dangerous (Ex: “Sure, you can take vaccines, if you don’t mind putting poison in your body.”)
- Natural immunity is safer than vaccine-acquired immunity (Ex: “It’s safest to just get the disease rather than the vaccine.”)
- It is dangerous to get several vaccines in a short period of time, even if that timing is medically recommended (Ex: “Never take more than one vaccine at the same time, that is dangerous - I don’t care what your doctor tells you!”)
- Vaccines are not effective at preventing the disease against which they purport to protect. However, for the COVID-19, flu, and malaria vaccines, we do not remove claims that those vaccines are not effective in preventing someone from contracting those viruses. (Ex’s: Remove – “The polio vaccine doesn’t do anything to stop you from getting the disease”; Remove – “Vaccines actually don’t do anything to stop you from getting diseases”; Allow – “The vaccine doesn’t stop you from getting COVID-19, that’s why you still need to socially distance and wear a mask when you’re around others.”)
- Acquiring measles cannot cause death (requires additional information and/or context) (Ex: “Don’t worry about whether you get measles, it can’t be fatal.”)
- Vitamin C is as effective as vaccines in preventing diseases for which vaccines exist.
- Misinformation about health during public health emergencies. We remove misinformation during public health emergencies when public health authorities conclude that the information is false and likely to directly contribute to the risk of imminent physical harm, including by contributing to the risk of individuals getting or spreading a harmful disease or refusing an associated vaccine. We identify public health emergencies in partnership with global and local health authorities. This currently includes certain false claims related to COVID-19 in countries still experiencing a publicly announced state of emergency regarding the virus. Click here for a complete set of rules regarding what misinformation we do not allow regarding COVID-19 and vaccines.
- Promoting or advocating for harmful miracle cures for health issues. These include treatments where the recommended application, in a health context, is likely to directly contribute to the risk of serious injury or death, and the treatment has no legitimate health use (ex: bleach, disinfectant, black salve, caustic soda).
III. Voter or Census Interference
In an effort to promote election and census integrity, we remove misinformation that is likely to directly contribute to a risk of interference with people’s ability to participate in those processes. This includes the following:
- Misinformation about the dates, locations, times, and methods for voting, voter registration, or census participation.
- Misinformation about who can vote, qualifications for voting, whether a vote will be counted, and what information or materials must be provided in order to vote.
- Misinformation about whether a candidate is running or not.
- Misinformation about who can participate in the census and what information or materials must be provided in order to participate.
- Misinformation about government involvement in the census, including, where applicable, that an individual's census information will be shared with another (non-census) government agency.
- False or unverified claims that the U.S. Immigration and Customs Enforcement (ICE) is at a voting location.
- Explicit false claims that people will be infected by COVID-19 (or another communicable disease) if they participate in the voting process.
- False claims about current conditions at a U.S. voting location that would make it impossible to vote, as verified by an election authority.
We have additional policies intended to cover calls for violence, the promotion of illegal participation, and calls for coordinated interference in elections, which are represented in other sections of our Community Standards.
IV. Manipulated Media
Media can be edited in a variety of ways. In many cases, these changes are benign, such as content being cropped or shortened for artistic reasons or music being added. In other cases, the manipulation is not apparent and could mislead, particularly in the case of video content. We remove this content because it can go viral quickly and experts advise that false beliefs regarding manipulated media often cannot be corrected through further discourse.
We remove videos under this policy if specific criteria are met: (1) the video has been edited or synthesized, beyond adjustments for clarity or quality, in ways that are not apparent to an average person, and would likely mislead an average person to believe a subject of the video said words that they did not say; and (2) the video is the product of artificial intelligence or machine learning, including deep learning techniques (e.g., a technical deepfake), that merges, combines, replaces, and/or superimposes content onto a video, creating a video that appears authentic.
For the following content, we include an informative label:
Content Digitally Created or Altered that May Mislead
For content that does not otherwise violate the Community Standards, we may place an informative label on the face of content – or reject content submitted as an advertisement – when the content is a photorealistic image or video, or realistic sounding audio, that was digitally created or altered and creates a particularly high risk of materially deceiving the public on a matter of public importance.
6 Disemba 2023
Policy Rationale
Misinformation is different from other types of speech addressed in our Community Standards because there is no way to articulate a comprehensive list of what is prohibited. With graphic violence or hate speech, for instance, our policies specify the speech we prohibit, and even persons who disagree with those policies can follow them. With misinformation, however, we cannot provide such a line. The world is changing constantly, and what is true one minute may not be true the next minute. People also have different levels of information about the world around them, and may believe something is true when it is not. A policy that simply prohibits “misinformation” would not provide useful notice to the people who use our services and would be unenforceable, as we don’t have perfect access to information.
Instead, our policies articulate different categories of misinformation and try to provide clear guidance about how we treat that speech when we see it. For each category, our approach reflects our attempt to balance our values of expression, safety, dignity, authenticity, and privacy.
We remove misinformation where it is likely to directly contribute to the risk of imminent physical harm. We also remove content that is likely to directly contribute to interference with the functioning of political processes. In determining what constitutes misinformation in these categories, we partner with independent experts who possess knowledge and expertise to assess the truth of the content and whether it is likely to directly contribute to the risk of imminent harm. This includes, for instance, partnering with human rights organizations with a presence on the ground in a country to determine the truth of a rumor about civil conflict.
For all other misinformation, we focus on reducing its prevalence or creating an environment that fosters a productive dialogue. We know that people often use misinformation in harmless ways, such as to exaggerate a point (“This team has the worst record in the history of the sport!”) or in humor or satire (“My husband just won Husband of the Year.”) They also may share their experience through stories that contain inaccuracies. In some cases, people share deeply-held personal opinions that others consider false or share information that they believe to be true but others consider incomplete or misleading.
Recognizing how common such speech is, we focus on slowing the spread of hoaxes and viral misinformation, and directing users to authoritative information. As part of that effort, we partner with third-party fact checking organizations to review and rate the accuracy of the most viral content on our platforms (see here to learn more about how our fact-checking program works). We also provide resources to increase media and digital literacy so people can decide what to read, trust, and share themselves. We require people to disclose, using our AI-disclosure tool, whenever they post organic content with photorealistic video or realistic-sounding audio that was digitally created or altered, and we may apply penalties if they fail to do so. We may also add a label to certain digitally created or altered content that creates a particularly high risk of misleading people on a matter of public importance.
Finally, we prohibit content and behavior in other areas that often overlap with the spread of misinformation. For example, our Community Standards prohibit fake accounts, fraud, and coordinated inauthentic behavior.
As online and offline environments change and evolve, we will continue to evolve these policies. Accounts that repeatedly share the misinformation listed below may, in addition to having their content enforced on in accordance with this policy, receive decreased distribution, limitations on their ability to advertise, or be removed from our platforms. Additional information on what happens when Meta removes content can be found here.
Misinformation we remove:
We remove the following types of misinformation:
I. Physical Harm or Violence
We remove misinformation or unverifiable rumors that expert partners have determined are likely to directly contribute to a risk of imminent violence or physical harm to people. We define misinformation as content with a claim that is determined to be false by an authoritative third party. We define an unverifiable rumor as a claim whose source expert partners confirm is extremely hard or impossible to trace, for which authoritative sources are absent, where there is not enough specificity for the claim to be debunked, or where the claim is too incredulous or too irrational to be believed.
We know that sometimes misinformation that might appear benign could, in a specific context, contribute to a risk of offline harm, including threats of violence that could contribute to a heightened risk of death, serious injury, or other physical harm. We work with a global network of non-governmental organizations (NGOs), not-for-profit organizations, humanitarian organizations, and international organizations that have expertise in these local dynamics.
In countries experiencing a heightened risk of societal violence, we work proactively with local partners to understand which false claims may directly contribute to a risk of imminent physical harm. We then work to identify and remove content making those claims on our platform. For example, in consultation with local experts, we may remove out-of-context media falsely claiming to depict acts of violence, victims or perpetrators of violence, weapons, or military hardware.
II. Harmful Health Misinformation
We consult with leading health organizations to identify health misinformation likely to directly contribute to imminent harm to public health and safety. The harmful health misinformation that we remove includes the following:
- Misinformation about vaccines. We remove misinformation primarily about vaccines when public health authorities conclude that the information is false and likely to directly contribute to imminent vaccine refusals. They include:
- Vaccines cause autism (Ex: “Increased vaccinations are why so many kids have autism these days.”)
- Vaccines cause Sudden Infant Death Syndrome (Ex: “Don’t you know vaccines cause SIDS?”
- Vaccines cause the disease against which they are meant to protect, or cause the person receiving the vaccine to be more likely to get the disease (Ex: “Taking a vaccine actually makes you more likely to get the disease since there’s a strain of the disease inside. Beware!”)
- Vaccines or their ingredients are deadly, toxic, poisonous, harmful, or dangerous (Ex: “Sure, you can take vaccines, if you don’t mind putting poison in your body.”)
- Natural immunity is safer than vaccine-acquired immunity (Ex: “It’s safest to just get the disease rather than the vaccine.”)
- It is dangerous to get several vaccines in a short period of time, even if that timing is medically recommended (Ex: “Never take more than one vaccine at the same time, that is dangerous - I don’t care what your doctor tells you!”)
- Vaccines are not effective at preventing the disease against which they purport to protect. However, for the COVID-19, flu, and malaria vaccines, we do not remove claims that those vaccines are not effective in preventing someone from contracting those viruses. (Ex’s: Remove – “The polio vaccine doesn’t do anything to stop you from getting the disease”; Remove – “Vaccines actually don’t do anything to stop you from getting diseases”; Allow – “The vaccine doesn’t stop you from getting COVID-19, that’s why you still need to socially distance and wear a mask when you’re around others.”)
- Acquiring measles cannot cause death (requires additional information and/or context) (Ex: “Don’t worry about whether you get measles, it can’t be fatal.”)
- Vitamin C is as effective as vaccines in preventing diseases for which vaccines exist.
- Misinformation about health during public health emergencies. We remove misinformation during public health emergencies when public health authorities conclude that the information is false and likely to directly contribute to the risk of imminent physical harm, including by contributing to the risk of individuals getting or spreading a harmful disease or refusing an associated vaccine. We identify public health emergencies in partnership with global and local health authorities. This currently includes certain false claims related to COVID-19 in countries still experiencing a publicly announced state of emergency regarding the virus. Click here for a complete set of rules regarding what misinformation we do not allow regarding COVID-19 and vaccines.
- Promoting or advocating for harmful miracle cures for health issues. These include treatments where the recommended application, in a health context, is likely to directly contribute to the risk of serious injury or death, and the treatment has no legitimate health use (ex: bleach, disinfectant, black salve, caustic soda).
III. Voter or Census Interference
In an effort to promote election and census integrity, we remove misinformation that is likely to directly contribute to a risk of interference with people’s ability to participate in those processes. This includes the following:
- Misinformation about the dates, locations, times, and methods for voting, voter registration, or census participation.
- Misinformation about who can vote, qualifications for voting, whether a vote will be counted, and what information or materials must be provided in order to vote.
- Misinformation about whether a candidate is running or not.
- Misinformation about who can participate in the census and what information or materials must be provided in order to participate.
- Misinformation about government involvement in the census, including, where applicable, that an individual's census information will be shared with another (non-census) government agency.
Content falsely claimingFalse or unverified claimsthat the U.S. Immigration and Customs Enforcement (ICE) is at a voting location.- Explicit false claims that people will be infected by COVID-19 (or another communicable disease) if they participate in the voting process.
- False claims about current conditions at a U.S. voting location that would make it impossible to vote, as verified by an election authority.
We have additional policies intended to cover calls for violence, the promotion of illegal participation, and calls for coordinated interference in elections, which are represented in other sections of our Community Standards.
IV. Manipulated Media
Media can be edited in a variety of ways. In many cases, these changes are benign, such as content being cropped or shortened for artistic reasons or music being added. In other cases, the manipulation is not apparent and could mislead, particularly in the case of video content. We remove this content because it can go viral quickly and experts advise that false beliefs regarding manipulated media often cannot be corrected through further discourse.
We remove videos under this policy if specific criteria are met: (1) the video has been edited or synthesized, beyond adjustments for clarity or quality, in ways that are not apparent to an average person, and would likely mislead an average person to believe a subject of the video said words that they did not say; and (2) the video is the product of artificial intelligence or machine learning, including deep learning techniques (e.g., a technical deepfake), that merges, combines, replaces, and/or superimposes content onto a video, creating a video that appears authentic.
30 Septemba 2023
Policy Rationale
Misinformation is different from other types of speech addressed in our Community Standards because there is no way to articulate a comprehensive list of what is prohibited. With graphic violence or hate speech, for instance, our policies specify the speech we prohibit, and even persons who disagree with those policies can follow them. With misinformation, however, we cannot provide such a line. The world is changing constantly, and what is true one minute may not be true the next minute. People also have different levels of information about the world around them, and may believe something is true when it is not. A policy that simply prohibits “misinformation” would not provide useful notice to the people who use our services and would be unenforceable, as we don’t have perfect access to information.
Instead, our policies articulate different categories of misinformation and try to provide clear guidance about how we treat that speech when we see it. For each category, our approach reflects our attempt to balance our values of expression, safety, dignity, authenticity, and privacy.
We remove misinformation where it is likely to directly contribute to the risk of imminent physical harm. We also remove content that is likely to directly contribute to interference with the functioning of political processes. In determining what constitutes misinformation in these categories, we partner with independent experts who possess knowledge and expertise to assess the truth of the content and whether it is likely to directly contribute to the risk of imminent harm. This includes, for instance, partnering with human rights organizations with a presence on the ground in a country to determine the truth of a rumor about civil conflict.
For all other misinformation, we focus on reducing its prevalence or creating an environment that fosters a productive dialogue. We know that people often use misinformation in harmless ways, such as to exaggerate a point (“This team has the worst record in the history of the sport!”) or in humor or satire (“My husband just won Husband of the Year.”) They also may share their experience through stories that contain inaccuracies. In some cases, people share deeply-held personal opinions that others consider false or share information that they believe to be true but others consider incomplete or misleading.
Recognizing how common such speech is, we focus on slowing the spread of hoaxes and viral misinformation, and directing users to authoritative information. As part of that effort, we partner with third-party fact checking organizations to review and rate the accuracy of the most viral content on our platforms (see here to learn more about how our fact-checking program works). We also provide resources to increase media and digital literacy so people can decide what to read, trust, and share themselves. We require people to disclose, using our AI-disclosure tool, whenever they post organic content with photorealistic video or realistic-sounding audio that was digitally created or altered, and we may apply penalties if they fail to do so. We may also add a label to certain digitally created or altered content that creates a particularly high risk of misleading people on a matter of public importance.
Finally, we prohibit content and behavior in other areas that often overlap with the spread of misinformation. For example, our Community Standards prohibit fake accounts, fraud, and coordinated inauthentic behavior.
As online and offline environments change and evolve, we will continue to evolve these policies. Accounts that repeatedly share the misinformation listed below may, in addition to having their content enforced on in accordance with this policy, receive decreased distribution, limitations on their ability to advertise, or be removed from our platforms. Additional information on what happens when Meta removes content can be found here.
Misinformation we remove:
We remove the following types of misinformation:
I. Physical Harm or Violence
We remove misinformation or unverifiable rumors that expert partners have determined are likely to directly contribute to a risk of imminent violence or physical harm to people. We define misinformation as content with a claim that is determined to be false by an authoritative third party. We define an unverifiable rumor as a claim whose source expert partners confirm is extremely hard or impossible to trace, for which authoritative sources are absent, where there is not enough specificity for the claim to be debunked, or where the claim is too incredulous or too irrational to be believed.
We know that sometimes misinformation that might appear benign could, in a specific context, contribute to a risk of offline harm, including threats of violence that could contribute to a heightened risk of death, serious injury, or other physical harm. We work with a global network of non-governmental organizations (NGOs), not-for-profit organizations, humanitarian organizations, and international organizations that have expertise in these local dynamics.
In countries experiencing a heightened risk of societal violence, we work proactively with local partners to understand which false claims may directly contribute to a risk of imminent physical harm. We then work to identify and remove content making those claims on our platform. For example, in consultation with local experts, we may remove out-of-context media falsely claiming to depict acts of violence, victims or perpetrators of violence, weapons, or military hardware.
II. Harmful Health Misinformation
We consult with leading health organizations to identify health misinformation likely to directly contribute to imminent harm to public health and safety. The harmful health misinformation that we remove includes the following:
- Misinformation about vaccines. We remove misinformation primarily about vaccines when public health authorities conclude that the information is false and likely to directly contribute to imminent vaccine refusals. They include:
- Vaccines cause autism (Ex: “Increased vaccinations are why so many kids have autism these days.”)
- Vaccines cause Sudden Infant Death Syndrome (Ex: “Don’t you know vaccines cause SIDS?”
- Vaccines cause the disease against which they are meant to protect, or cause the person receiving the vaccine to be more likely to get the disease (Ex: “Taking a vaccine actually makes you more likely to get the disease since there’s a strain of the disease inside. Beware!”)
- Vaccines or their ingredients are deadly, toxic, poisonous, harmful, or dangerous (Ex: “Sure, you can take vaccines, if you don’t mind putting poison in your body.”)
- Natural immunity is safer than vaccine-acquired immunity (Ex: “It’s safest to just get the disease rather than the vaccine.”)
- It is dangerous to get several vaccines in a short period of time, even if that timing is medically recommended (Ex: “Never take more than one vaccine at the same time, that is dangerous - I don’t care what your doctor tells you!”)
- Vaccines are not effective at preventing the disease against which they purport to protect. However, for the COVID-19, flu, and malaria vaccines, we do not remove claims that those vaccines are not effective in preventing someone from contracting those viruses. (Ex’s: Remove – “The polio vaccine doesn’t do anything to stop you from getting the disease”; Remove – “Vaccines actually don’t do anything to stop you from getting diseases”; Allow – “The vaccine doesn’t stop you from getting COVID-19, that’s why you still need to socially distance and wear a mask when you’re around others.”)
- Acquiring measles cannot cause death (requires additional information and/or context) (Ex: “Don’t worry about whether you get measles, it can’t be fatal.”)
- Vitamin C is as effective as vaccines in preventing diseases for which vaccines exist.
- Misinformation about health during public health emergencies. We remove misinformation during public health emergencies when public health authorities conclude that the information is false and likely to directly contribute to the risk of imminent physical harm, including by contributing to the risk of individuals getting or spreading a harmful disease or refusing an associated vaccine. We identify public health emergencies in partnership with global and local health authorities. This currently includes certain false claims related to COVID-19 in countries still experiencing a publicly announced state of emergency regarding the virus. Click here for a complete set of rules regarding what misinformation we do not allow regarding COVID-19 and vaccines.
- Promoting or advocating for harmful miracle cures for health issues. These include treatments where the recommended application, in a health context, is likely to directly contribute to the risk of serious injury or death, and the treatment has no legitimate health use (ex: bleach, disinfectant, black salve, caustic soda).
III. Voter or Census Interference
In an effort to promote election and census integrity, we remove misinformation that is likely to directly contribute to a risk of interference with people’s ability to participate in those processes. This includes the following:
- Misinformation about the dates, locations, times, and methods for voting, voter registration, or census participation.
- Misinformation about who can vote, qualifications for voting, whether a vote will be counted, and what information or materials must be provided in order to vote.
- Misinformation about whether a candidate is running or not.
- Misinformation about who can participate in the census and what information or materials must be provided in order to participate.
- Misinformation about government involvement in the census, including, where applicable, that an individual's census information will be shared with another (non-census) government agency.
- Content falsely claiming that the U.S. Immigration and Customs Enforcement (ICE) is at a voting location.
- Explicit false claims that people will be infected by COVID-19 (or another communicable disease) if they participate in the voting process.
We have additional policies intended to cover calls for violence, the promotion of illegal participation, and calls for coordinated interference in elections, which are represented in other sections of our Community Standards.
IV. Manipulated Media
Media can be edited in a variety of ways. In many cases, these changes are benign, such as content being cropped or shortened for artistic reasons or music being added. In other cases, the manipulation is not apparent and could mislead, particularly in the case of video content. We remove this content because it can go viral quickly and experts advise that false beliefs regarding manipulated media often cannot be corrected through further discourse.
We remove videos under this policy if specific criteria are met: (1) the video has been edited or synthesized, beyond adjustments for clarity or quality, in ways that are not apparent to an average person, and would likely mislead an average person to believe a subject of the video said words that they did not say; and (2) the video is the product of artificial intelligence or machine learning, including deep learning techniques (e.g., a technical deepfake), that merges, combines, replaces, and/or superimposes content onto a video, creating a video that appears authentic.
To provide transparency that in countries experiencing a heightened risk of societal violence, we identify and remove content making harmful misinformation claims.
15 Julayi 2023
Policy Rationale
Misinformation is different from other types of speech addressed in our Community Standards because there is no way to articulate a comprehensive list of what is prohibited. With graphic violence or hate speech, for instance, our policies specify the speech we prohibit, and even persons who disagree with those policies can follow them. With misinformation, however, we cannot provide such a line. The world is changing constantly, and what is true one minute may not be true the next minute. People also have different levels of information about the world around them, and may believe something is true when it is not. A policy that simply prohibits “misinformation” would not provide useful notice to the people who use our services and would be unenforceable, as we don’t have perfect access to information.
Instead, our policies articulate different categories of misinformation and try to provide clear guidance about how we treat that speech when we see it. For each category, our approach reflects our attempt to balance our values of expression, safety, dignity, authenticity, and privacy.
We remove misinformation where it is likely to directly contribute to the risk of imminent physical harm. We also remove content that is likely to directly contribute to interference with the functioning of political processes. In determining what constitutes misinformation in these categories, we partner with independent experts who possess knowledge and expertise to assess the truth of the content and whether it is likely to directly contribute to the risk of imminent harm. This includes, for instance, partnering with human rights organizations with a presence on the ground in a country to determine the truth of a rumor about civil conflict.
For all other misinformation, we focus on reducing its prevalence or creating an environment that fosters a productive dialogue. We know that people often use misinformation in harmless ways, such as to exaggerate a point (“This team has the worst record in the history of the sport!”) or in humor or satire (“My husband just won Husband of the Year.”) They also may share their experience through stories that contain inaccuracies. In some cases, people share deeply-held personal opinions that others consider false or share information that they believe to be true but others consider incomplete or misleading.
Recognizing how common such speech is, we focus on slowing the spread of hoaxes and viral misinformation, and directing users to authoritative information. As part of that effort, we partner with third-party fact checking organizations to review and rate the accuracy of the most viral content on our platforms (see here to learn more about how our fact-checking program works). We also provide resources to increase media and digital literacy so people can decide what to read, trust, and share themselves. We require people to disclose, using our AI-disclosure tool, whenever they post organic content with photorealistic video or realistic-sounding audio that was digitally created or altered, and we may apply penalties if they fail to do so. We may also add a label to certain digitally created or altered content that creates a particularly high risk of misleading people on a matter of public importance.
Finally, we prohibit content and behavior in other areas that often overlap with the spread of misinformation. For example, our Community Standards prohibit fake accounts, fraud, and coordinated inauthentic behavior.
As online and offline environments change and evolve, we will continue to evolve these policies. Accounts that repeatedly share the misinformation listed below may, in addition to having their content enforced on in accordance with this policy, receive decreased distribution, limitations on their ability to advertise, or be removed from our platforms. Additional information on what happens when Meta removes content can be found here.
Misinformation we remove:
We remove the following types of misinformation:
I. Physical Harm or Violence
We remove misinformation or unverifiable rumors that expert partners have determined are likely to directly contribute to a risk of imminent violence or physical harm to people. We define misinformation as content with a claim that is determined to be false by an authoritative third party. We define an unverifiable rumor as a claim whose source expert partners confirm is extremely hard or impossible to trace, for which authoritative sources are absent, where there is not enough specificity for the claim to be debunked, or where the claim is too incredulous or too irrational to be believed.
We know that sometimes misinformation that might appear benign could, in a specific context, contribute to a risk of offline harm, including threats of violence that could contribute to a heightened risk of death, serious injury, or other physical harm. We work with a global network of non-governmental organizations (NGOs), not-for-profit organizations, humanitarian organizations, and international organizations that have expertise in these local dynamics.
II. Harmful Health Misinformation
We consult with leading health organizations to identify health misinformation likely to directly contribute to imminent harm to public health and safety. The harmful health misinformation that we remove includes the following:
- Misinformation about vaccines. We remove misinformation primarily about vaccines when public health authorities conclude that the information is false and likely to directly contribute to imminent vaccine refusals. They include:
- Vaccines cause autism (Ex: “Increased vaccinations are why so many kids have autism these days.”)
- Vaccines cause Sudden Infant Death Syndrome (Ex: “Don’t you know vaccines cause SIDS?”
- Vaccines cause the disease against which they are meant to protect, or cause the person receiving the vaccine to be more likely to get the disease (Ex: “Taking a vaccine actually makes you more likely to get the disease since there’s a strain of the disease inside. Beware!”)
- Vaccines or their ingredients are deadly, toxic, poisonous, harmful, or dangerous (Ex: “Sure, you can take vaccines, if you don’t mind putting poison in your body.”)
- Natural immunity is safer than vaccine-acquired immunity (Ex: “It’s safest to just get the disease rather than the vaccine.”)
- It is dangerous to get several vaccines in a short period of time, even if that timing is medically recommended (Ex: “Never take more than one vaccine at the same time, that is dangerous - I don’t care what your doctor tells you!”)
- Vaccines are not effective at preventing the disease against which they purport to protect. However, for the COVID-19, flu, and malaria vaccines, we do not remove claims that those vaccines are not effective in preventing someone from contracting those viruses. (Ex’s: Remove – “The polio vaccine doesn’t do anything to stop you from getting the disease”; Remove – “Vaccines actually don’t do anything to stop you from getting diseases”; Allow – “The vaccine doesn’t stop you from getting COVID-19, that’s why you still need to socially distance and wear a mask when you’re around others.”)
- Acquiring measles cannot cause death (requires additional information and/or context) (Ex: “Don’t worry about whether you get measles, it can’t be fatal.”)
- Vitamin C is as effective as vaccines in preventing diseases for which vaccines exist.
- Misinformation about health during public health emergencies. We remove misinformation during public health emergencies when public health authorities conclude that the information is false and likely to directly contribute to the risk of imminent physical harm, including by contributing to the risk of individuals getting or spreading a harmful disease or refusing an associated vaccine. We identify public health emergencies in partnership with global and local health authorities. This currently includes certain false claims related to COVID-19 in countries still experiencing a publicly announced state of emergency regarding the virus.
This currently includes false claims related to COVID-19 that are verified by expert health authorities, about the existence or severity of the virus, how to cure or prevent it, how the virus is transmitted or who is immune, and false claims which discourage good health practices related to COVID-19 (such as getting tested, social distancing, wearing a face mask, and getting a vaccine for COVID-19).Click here for a complete set of rules regarding what misinformation we do not allow regarding COVID-19 and vaccines.
- Promoting or advocating for harmful miracle cures for health issues. These include treatments where the recommended application, in a health context, is likely to directly contribute to the risk of serious injury or death, and the treatment has no legitimate health use (ex: bleach, disinfectant, black salve, caustic soda).
III. Voter or Census Interference
In an effort to promote election and census integrity, we remove misinformation that is likely to directly contribute to a risk of interference with people’s ability to participate in those processes. This includes the following:
- Misinformation about the dates, locations, times, and methods for voting, voter registration, or census participation.
- Misinformation about who can vote, qualifications for voting, whether a vote will be counted, and what information or materials must be provided in order to vote.
- Misinformation about whether a candidate is running or not.
- Misinformation about who can participate in the census and what information or materials must be provided in order to participate.
- Misinformation about government involvement in the census, including, where applicable, that an individual's census information will be shared with another (non-census) government agency.
- Content falsely claiming that the U.S. Immigration and Customs Enforcement (ICE) is at a voting location.
- Explicit false claims that people will be infected by COVID-19 (or another communicable disease) if they participate in the voting process.
We have additional policies intended to cover calls for violence, the promotion of illegal participation, and calls for coordinated interference in elections, which are represented in other sections of our Community Standards.
IV. Manipulated Media
Media can be edited in a variety of ways. In many cases, these changes are benign, such as content being cropped or shortened for artistic reasons or music being added. In other cases, the manipulation is not apparent and could mislead, particularly in the case of video content. We remove this content because it can go viral quickly and experts advise that false beliefs regarding manipulated media often cannot be corrected through further discourse.
We remove videos under this policy if specific criteria are met: (1) the video has been edited or synthesized, beyond adjustments for clarity or quality, in ways that are not apparent to an average person, and would likely mislead an average person to believe a subject of the video said words that they did not say; and (2) the video is the product of artificial intelligence or machine learning, including deep learning techniques (e.g., a technical deepfake), that merges, combines, replaces, and/or superimposes content onto a video, creating a video that appears authentic.
Okubonwa yiuser
Jonga eminye imizekelo ebonisa ukuba lunjani unyazeliso lweepolisi kubantu abakuFacebook, njengokuthi: kuba njani ukuxela into ocinga ukuba ayimele ibe kuFacebook, ukuxelelwa ukuba unyhashe Imigangatho Yoluntu yethu kunye nokubona isikrini esilumkisayo kwikhontenti ethile.
Phawula: Sisoloko siphucula, ngoko into oyibona apha isengaba seyibuphelelwa lixesha xa ithelekiswa nento esiyisebenzisayo ngoku sithethayo.
OKUBONWA YIUSER
Ukuxela
OKUBONWA YIUSER
Unxibelelwano lwasemva kokuba kuxeliwe
AMAVA EUSER
Amava okukhutshwa
AMAVA EUSER
Izikrini ezilumkisayo
Unyazeliso lweepolisi
Sineepolisi ezifanayo ehlabathini lonke, ukwenzela wonke ubani okuFacebook.
Amaqela ohlolo
Iqela lethu elisehlabathini lonke elinabahloli abangaphezu kwama15000 lisebenza yonke imihla ukugcina abantu abakuFacebook bekhuselekile.
Ukuthethana nababandakanyekayo
Iingcali zangaphandle, izifundiswa, iiNGO nabaqulunqi beepolisi bayanceda ukufaka igalelo kwiMigangatho Yoluntu kaFacebook.
Fumana uncedo malunga nenkcazelo engeyonyani
Funda oko unokukwenza ukuba ubona into kuFacebook engqubana neMigangatho Yoluntu yethu.
Ndwendwela Iziko Loncedo lethu