Social media companies must urgently address the fanning of online hate and racism against Palestinian and Jewish communities, Amnesty International said today, as the conflict in Israel and the Occupied Palestinian Territories escalates further.
The organization has found an alarming rise in advocacy of hatred that constitutes incitement to violence, hostility, and discrimination on social media platforms, which is prohibited under international human rights law, and other harmful content against Palestinian and Jewish people. Amnesty International also documented concerning reports that content posted by Palestinians and advocates of Palestinian rights are being subject to potentially discriminatory content moderation by different social media platforms.
Social media can play a vital role in times of crisis for communication––as such, major social media companies need to urgently step-up measures to protect human rights.
Rasha Abdul-Rahim, Director of Amnesty Tech
“Social media can play a vital role in times of crisis for communication––as such, major social media companies need to urgently step-up measures to protect human rights. The companies must ensure their platforms are not broadcasting messages of hatred and violence, or they risk contributing to gross human rights violations and serious violations of international humanitarian law. This includes war crimes, which have marked this conflict, and caused unprecedented levels of civilian casualties and a catastrophic humanitarian crisis,” said Rasha Abdul-Rahim, Director of Amnesty Tech.
Amnesty International found that a significant number of social media posts glorify Israel’s attacks on civilians in Gaza, support the destruction of Gaza and advocate violence against Palestinians. Many posts use dehumanizing and racist language against Palestinians, a number of which also draw on language used by the Israeli authorities.
The organization also documented a number of antisemitic posts, many of them advocating hatred and violence against Jewish people. Previous research by the Center for Countering Digital Hate has highlighted a proliferation of antisemitic content on X in recent months.
Palestinian NGO 7amleh’s Violence Indicator detected over 493,000 cases (and increasing) of advocacy of hatred against Palestinians and advocates of Palestinian rights in Hebrew language content on social media platforms, since 7 October 2023.
Israeli government and military officials have also used dehumanizing language and anti-Palestinian racism. On 16 October, the Prime Minister of Israel’s X account (@IsraeliPM) posted a discriminatory post using dehumanizing language, saying: “There is a struggle between the children of light and children of darkness, between humanity and law of the jungle.”
Similarly, the Minister of National Security, Itamar Ben Gvir on 17 October 2023 stated: “As long as Hamas does not release the hostages in its hands – the only thing that needs to enter Gaza are hundreds of tons of explosives from the Air Force, not an ounce of humanitarian aid.”
Outright blackout and “shadowbanning” of Palestinian content
Amnesty International also received concerning reports indicating an over-broad censorship of content from Palestinian accounts and advocates of Palestinian rights on several social media platforms.
Palestinians in the occupied Gaza Strip are trapped in an increasing communication blackoutlimiting their ability to seek, receive and impart information. Unequal content moderation by social media platforms risks further undermining the ability of Palestinians in and outside of Gaza to exercise their rights to freedom of expression, association and peaceful assembly.
“As Israel intensifies its unprecedented bombardment of Gaza that has killed over 7,000 people, most of whom are civilians, we are extremely concerned by reports of partial blocking and removal, known as ‘shadowbanning’, of content from advocates of Palestinian rights,” said Rasha Abdul-Rahim.
As Israel intensifies its unprecedented bombardment of Gaza that has killed over 7,000 people, most of whom are civilians, we are extremely concerned by reports of partial blocking and removal, known as ‘shadowbanning’, of content from advocates of Palestinian rights.
Rasha Abdul-Rahim, Director of Amnesty Tech
“Content moderation must be conducted in a manner ensuring respect for the right to freedom of expression, while addressing rampant advocacy of hatred. Social media companies need to invest adequate resources in human oversight of AI (Artificial Intelligence)-driven content moderation systems to ensure all users, including Palestinians, can equally exercise their rights online, regardless of language and political views. Several platforms have undermined their ability to do so by laying off the staff responsible for human rights compliance.”
Research has shown that AI systems often reproduce pre-existing societal biases under a veil of neutrality. As recently as 19 October 2023, Meta apologized for inserting the word “terrorist” to Instagram profile translations containing the words “Palestinian” and “Alhamdullilah” (which means Praise be to God), and the Palestinian flag emoji. The company also reduced the threshold of certainty required for ‘hiding’ hostile content from 80% to 25%, for content originating in large parts of the Middle East. This was an attempt to stem the flow of hostile speech but could result in overbroad restrictions on content.
“As Israel intensifies its unprecedented bombardment of Gaza that has killed over 7,000 people, most of whom are civilians, we are extremely concerned by reports of partial blocking and removal, known as ‘shadowbanning’, of content from advocates of Palestinian rights,” said Rasha Abdul-Rahim.
In May 2021, a Business for Social Responsibility report revealed that Arabic-language content had greater “over-enforcement” on Meta’s platforms than other languages, including Hebrew. “False strikes” contributed to the reduced visibility and engagement of Arabic posts.
Even before the current crisis, Amnesty International’s research found that Israel’s system of oppression and domination of Palestinians amounts to the crime against humanity of apartheid. Amnesty International’s research has also shown that social media algorithms are designed to maximize engagement, resulting in often disproportionately amplified harmful and inflammatory content, including advocacy of hatred inciting violence, hostility and discrimination.
Against this backdrop, it is imperative that Big Tech companies address the actual and potential human rights impacts of their operations to ensure they do not cause or contribute to human rights abuses, or enable hate, racism and misinformation to proliferate.
On 7 October, Hamas and other Palestinian armed groups flagrantly violated international law and displayed a chilling disregard for human life by carrying out cruel and brutal crimes including mass summary killings, hostage-taking, and launching indiscriminate rocket attacks into Israel. Israeli forces have since then launched relentless air bombardments across the whole of the Gaza Strip with cataclysmic effects. The grave humanitarian situation stemming from Israel’s 16-year-long illegal blockade has been further exacerbated also by Israel’s Minister for Defense order to impose a complete siege on 8 October, which cuts off Palestinians in Gaza from access to food, water, fuel, electricity and communication and amounts to the war crime of collective punishment.
Background:
Mounting restrictions are also being placed by several authorities in the West Bank and Europe, among other places, on protests defending Palestinian human rights.
Amnesty International is continuing to monitor instances of the wrongful blocking or removal of social media posts, as well as shadowbanning of accounts in the context of the current conflict. If you or anyone you know has experienced this, please share links and/or screenshots of the content that was flagged or removed, as well as screenshots of the original content with [email protected].