May 22, 2024
A message to META investors: Vote FOR on Proposal Eight Regarding Report on Human Rights Risks in Non-US Markets
Our proposal requests a report to shareholders on the effectiveness of measures Meta is taking to prevent and mitigate human rights risks in its five largest non-US markets (based on number of users) relating to the proliferation of hate speech, disinformation, and incitement to violence enabled by its Instagram and Facebook platforms. The company has responded by noting its policy positions but also has noted in its 2021 Civil Rights Audit Update that:
Though implementing these 16 recommendations is important to improving our content moderation policies, needless to say, there is much that still needs to be done and policies are constantly evolving to meet the needs of our users.1
Our request for a report on the content moderation performance in the five largest non-US markets asks Meta’s management to provide investors with transparent data on these issues in markets of immense significance to the growth and success of our company’s platforms. A failure to moderate content effectively will lead to further regulatory efforts that may curb profits and growth.
We are particularly concerned given that In just the past few weeks, we have seen evidence of a lack of progress in Meta’s largest market, India, and we are now hearing from Brazil that disinformation content on Meta platforms is hindering response efforts to the severe flooding in Rio Grande do Sul.
In April and May a series of investigations published by Ekō during India’s ongoing election have revealed how Meta has facilitated2 far-right networks reaching millions of Indian voters with Islamophobic hate ads.
_____________________________
1 https://about.fb.com/wp-content/uploads/2021/11/Metas-Progress-on-Civil-Rights-Audit-Commitments.pdf
2 https://aks3.eko.org/pdf/Slander_Lies_and_Incitement_Report.pdf
The far-right’s utilization of Meta as an incubator of hate speech has been extensively documented for nearly a decade. The new report provides clear evidence to show how Meta’s “Indian election preparation”3 statement is already failing due to their piece-meal actions against a tide of systemic issues embedded in their harmful business model. The report underscores the urgent need for mechanisms to curb the influence of paid advertisements by shadow actors on social media platforms and safeguard India’s democratic ethos.
Key findings include:
| ● | 1M USD spent by 22 far-right shadow advertisers over 90 days on Meta, accounting for nearly 22% of the total sum of "issues, elections or politics" advertisements in India across the same time period. |
| ● | 36 ads potentially breaking Indian election laws pushing hate speech, Islamophobia, communal violence, and misinformation amassing between 65-66M impressions. The ads include: |
| ○ | Islamophobic tropes and disinformation targeting Muslims: depicting them as sexually violent invaders, “instigating communal riots”, and “patrons of terror” |
| ○ | Hindu supremacist narratives: ads promoting narratives to recreate India into a Hindu nation and claiming “Hindustan is a country for Hindus” as well as hashtags tied to Hindu supremacist groups |
| ○ | Targeting women lawmakers, journalists, and activists as “anti-India” |
| ○ | Aggressive and violent rhetoric against the BJP opposition: referring to opponents as a "virus" that needs to be eradicated and that a BJP candidate will "break their [DMK] spine.” As well as insults such as "demon," and "poisonous snake" |
| ● | Researchers uncovered a coordinated network led by a public page named Ulta Chashma, which has nine other pages sharing content, hashtags, and ad payments. This network amassed 10.53M interactions and 34.64M video views within 90 days, with a loophole allowing early political ad postings, indicating a breach of Meta's ad transparency policy |
| ● | 23.13M interactions on 22 far-right publisher pages over 90 days |
| ● | None of the far-right publishers analyzed in the report responded on the phone numbers provided as disclaimers in Meta’s Ad library |
This week a new report4 also documents how Meta approved AI-manipulated political ads seeking to spread disinformation and incite religious violence during India's elections. Between May 8th and May 13th, Meta approved 14 highly inflammatory ads. These ads called for violent uprisings targeting Muslim minorities, disseminated blatant disinformation exploiting communal or religious conspiracy theories prevalent in India's political landscape, and incited violence through Hindu supremacist narratives.
_____________________________
3 https://about.fb.com/news/2024/03/how-meta-is-preparing-for-indian-general-elections-2024/
4 https://aks3.eko.org/pdf/Meta_AI_ads_investigation.pdf
Accompanying each ad text were manipulated images generated by AI image tools, proving how quickly and easily this new technology can be deployed to amplify harmful content. Meta’s systems did not block researchers posting political and incendiary ads during the election “silence period”. The process of setting up the Facebook accounts was extremely simple and researchers were able to post these ads from outside of India. Researchers pulled the ads before they were actually published.
As we write, Brazil is experiencing devastating floods. Our proposal comes amidst the rescue response to this historic flooding, which has been significantly impeded by disinformation and conspiracy theories pushed by the far-right and pro-Bolsonaro allies. The floods have already killed 161 people and displaced over 580,000.
Brazilian fact-checkers5 are working around the clock to bust the infodemic, such as:
| ● | Allegations that a local government is repackaging public donations with official logos to appear as if they are from federal agencies. |
| ● | Claims that Lula’s government denied donations from certain countries. |
| ● | False claims that “Starlink”, Musk's satellite internet, is the only available internet service in the state; at least two other operators are fully functional. |
| ● | An old audio clip from 2019 regarding Israeli military rescue support, circulated as if it were current. |
| ● | Inflated numbers of hospital deaths. |
The most damaging narratives discredit federal authorities, suggesting they are blocking or denying rescue efforts and donations to victims. Other disinformation includes false claims related to climate change, international celebrities criticizing the government, and the use of unrelated images and videos to exaggerate the tragedy’s death toll.
In light of these current examples of Meta’s lack of effective content moderation in two of its largest non-US markets, the proposal’s request for a report to shareholders on the effectiveness of measures it is taking to prevent and mitigate human rights risks in its five largest non-US markets (based on number of users) is compelling. As investors, it is essential we have a transparent and accurate view of content moderation capacity and competence in such key markets.
_____________________________
5 https://www.estadao.com.br/estadao-verifica/fake-news-enchente-chuvas-rio-grande-sul/?j=1268203&sfmc_sub=1001900989&l=9451_HTML&u=36879072&mid=534001280&jb=7011&utm_medium=email&utm_source=salesforce&utm_campaign=sqeng-estadao-verifica-fake-news-rs&utm_term=20240515&utm_content=ler-agora-digital
