` `

How Israeli Tech Groups Are Targeting Palestinian Content

Reem Abd Ulhamid Reem Abd Ulhamid
News
31st January 2024
How Israeli Tech Groups Are Targeting Palestinian Content
Palestinians report social media censorship without policy breaches (Getty)

Israeli requests to remove Palestinian content from social media platforms have augmented 10 folds according to Israeli state prosecutor office. Israel also succeeded in eliminating 94% of online Palestinian content from major social platforms. 

Several Israeli officials advocated shaping the international public opinion particularly on social media platforms as “the second battlefield” and “the information war.” Israel National Cyber Directorate official, Gaby Portnoy declared on several occasions that Cyber Dome will elevate national cybersecurity by implementing new mechanisms administered by cybersecurity experts from various departments, including the ministry of defence, the Israeli Forces, and Israeli intelligence agencies: Mossad and Shin Bet. Additionally, Itamar Ben Gvir, a far-right member of the new government in Israel established an anti-incitement team to monitor and remove what he described as inflammatory content on internet. 

Israeli Platforms and Incorporating Artificial Intelligence

Dozens of the dedicated websites and platforms emerged, after October 7 are led by tech executives, pro-Israel investors, as well as and Israeli government officials. Most of these online platforms, incorporate AI algorithms, data analysis to be treated by hundreds of tech volunteers around the globe and in several languages. 

A supporting image within the article body

The goal, as suggested by the platforms’ names, Iron Truth, Cyber Dome, Digital Dome, Word of Iron, is to intercept Palestinian content on social platforms just like Iron Dome military rocket interception system. Some like Canary Mission, a group that blacklists pro-Palestinian professors and students on U.S. college campuses, help train "AI models to classify antisemitic posts in social media platforms.” In their websites, they include organizations that “support or promote terrorist groups such as Hamas and Islamic Jihad.”

A supporting image within the article body

DigitalDome website is considered as part of hasbara (Israeli term for public diplomacy, pro-Israel public relations efforts, and collaborate work with Fakereporter (Israeli fact-checking website). It is funded by the government “J-Ventures Global Kibbutz Group,” a U.S.-Israeli investment fund also donated $19,531 to website to further develop intercepting Palestinian narrative on social media. 

A screenshot of a phone

Description automatically generated

According to their website, DigitalDome’s achievements include reporting and flagging an account that belongs to European Union legal body for posting videos with Hamas fighters.

A screenshot of a phone

Description automatically generated

Training Bots and AI Models

Platforms like “Iron Truth” and Israel Tech Guard use bots developed “to identify anti-Semitic and anti-Israeli content in social platforms.” Israeli Tech Guard explains to its volunteers the process, that includes 3 stages. The first is to monitor by scanning all social platforms including: Facebook, TikTok, Instagram, LinkedIn, X. The second is identifying “hate speech” where bot alerts relevant authorities within the social platform and finally the bot sends another alert for the employer of the person who posted what they classify as “hate speech.”

Key Words and Guidelines 

“Iron Truth” founded by Dani Kaganovitch, a Tel Aviv-based software engineer at Google and a group of volunteers that treat 700 reports on a daily base to remove hundreds of posts with a recent removal of 1000 posts tagged as antisemitic, or “pro-terrorist” across platforms such as X, YouTube, and TikTok.

Kaganovitch posted on his LinkedIn profile:

The fight for public opinion has never been more important. Our enemy’s fake news and incitement machine doesn’t rest for a second. It’s time to win on all fronts, including this one. For that purpose, we created the “Iron Truth” Bot. Now all we need is YOU – stumbled upon fake news? Inciting posts? Support of Hamas’ Terrorism? Report it to the “Iron Truth” Bot from which the content will be transferred to quick removal. Let’s win this fight. report here: 

A screenshot of a blue and white screen

Description automatically generated

Furthermore, tech volunteers receive training workshops on Zoom and other platforms as well as video tutorials, media archive on how to identify and tag/flag content including: text, images, videos and other aspects. Suggestions to lookout for include any content related to Israel’s cruel bombing of Gaza particularly hospitals; terms like “Israel bomb hospital,” “Israel bomb churches,” “Israel bomb humanitarian,” and “Israel committing genocide” “Israel war crimes” are often presented as disinformation and fake news keywords.

A screenshot of a phone

Description automatically generated

Boosting the Israeli Narrative on Social Platforms

Reporting and flagging content is not the only mission, several platforms like Stand4Israel, Word of Iron, fighthamas and others offer an alternative Israeli narrative to follow once the volunteer flags the content. Stand4Israel provides its volunteers with hundreds of “streaming” videos and content ready to share in 11 different languages. The user is also provided with a media collection that includes videoclips and interviews on Palestinian children taught to hate Jews, the truth behind the blast in Gazan hospitals, Silent victims: animals affected by war, and other videos that reveal “evidence of Hamas using civilians as human shields” and others. 

A supporting image within the article body

Words of Iron, invites users to boost Israel's Voice on social media with a Single Click, the tutorial video illustrates how volunteers can choose different automatised responses in the chosen language.

A screenshot of a phone

Description automatically generated

Fighthamas is another platform that provide and train its volunteers to respond with relevant automated responses for example posts.

A screenshot of a social media message

Description automatically generated

Oct7.io platform, which is also an App that can be downloaded on iPhone and Android, provides its volunteers with “the most critical posts.” The App proudly informs its users that it handled 16.630 posts until now. 

A blue screen with white text

Description automatically generated

hashtags like #HamasIsISIS, #IStandwithIsrael, and #IsraelUnderAttack are usually automated for most of the suggested responses. 

Ramifications: Shadow Banning and Restricting Palestinian Users

Palestinian social media users, particularly activists, journalists and human rights defenders have complained that content has been censored on social platforms without any clear violation of a given company’s policies.

At the beginning of Israeli war on Gaza, Instagram apologized for inserting the word “terrorist” to Palestinian profiles because of a bug, WhatsApp also auto generated with the word Palestine illustrations of gun wielding children. 

7amleh, The Arab Center for Social Media Advancement documented 4,400 violations varied between removing and restricting content, hacking accounts, all the way to hate speech, violence, and incitement in their recent report. The report added that 69% of these violations happened after October 7, 2023. The organization also noted that that content that supported Palestine in Arabic was more censored on social platforms because of the “massive investment in Arabic content classifier” whereas hostile content in Hebrew language was not censored due to “lack of sufficient data to deal with offensive content in Hebrew.”

According to dozens of Palestinian users Meta, parent company of Facebook and Instagram deleted or suspended users’ accounts, limited users’ engagement, or “shadow banned” them, making their posts less visible without informing them which is considered against their policies. Shadow banning is a mechanism used to reduce visibility of content. Consequently, last October, Meta updated comment methods programming as well as changed the publishing system, which resulted in reducing accessibilities for instance the default status for publishing became for ‘friends’ only instead of public. Interaction with posts has changed as well when a user wants to share its default would be to “send a private message.”

Censoring Palestinian and pro-Palestinian content on Meta is not recent. In 2018 the Israeli ministry of justice claimed 90 percent compliance rate across social media platforms. 

Human Rights Watch report described Meta’s censorship of Palestinian content as systemic and criticized suppression and removal of “peaceful expression that supports Palestine and public debate about Palestinian human rights.” 

In 2021, Business for Social Responsibility found out, in their independent investigation, that Meta had violated Palestinian human rights by heavily censoring content on its platforms, particularly “freedom of speech, political participation, and non-discrimination, thus on the ability of Palestinians to share information and insights about their experiences as they occurred.”

In a humanitarian catastrophic where around 30 thousand Palestinians were killed, according to the Palestinian ministry of health, including the killing of 120 journalists in Gaza, it is crucial that Meta and other social media platforms avoid censoring content, which suppresses the Palestinian narrative online. 

Read More

Investigating the Link Between Disinformation of the FDD and Its Leaders’ Support of Israel

The Washington Institute for Near East Policy: The Academic Legitimacy of Israeli Discourse

Most Read