Jerusalem-based journalist Faten Elwan has more than 300,000 followers on her Instagram account, a page which regularly covers events in Palestine and documents the Israeli occupation, but now her posts typically get just a few hundred views.

In April 2026, Elwan was on a hike when she discovered her account had been unexpectedly reactivated after a two-month suspension, something she believes was due to her pro-Palestine content.

Her first story back on the platform was of her walk along the sea and reached 30,000 views. Within an hour, the number for later stories had dropped to a few hundred.

She is one of many pro-Palestinian account holders who believe they have been “shadow-banned” by Meta, a technique allegedly used by social media companies to restrict the visibility of certain accounts, consequently limiting the number of people who can see their content.

“People tell me: if we don’t specifically search for you and type your name, ‘Faten Elwan’, together, we can’t find you. I no longer exist,” she told Al Jazeera.

Elwan’s account was suspended for the first time in late 2023, shortly after the start of Israel’s genocidal assault on Gaza, and her account has been taken down eight times during the war. In May 2025, Elwan managed to meet representatives of Meta, the company that owns Facebook and Instagram, to flag her silencing on their platforms, but she said their response to her complaint appeared dismissive and noncommitted.

“During my meeting with Meta, they heard my statement, and there was no reaction whatsoever. I gave my statement and provided examples of what happened to me, how I’m technically just a shadow on social media,” she said.

“We all used to have a school principal. The principal would punish the bad student and shut him up. Meta is simply playing that same role, but on a global scale.”

Shadow banning

Elwan’s case is one of many documented in a report investigating how Meta has allegedly suppressed Palestinian voices over the past five years, including during successive Israeli assaults on Gaza.

The Platformicide of Palestine 2021–2025, a report published by 7amleh, the Arab Centre for the Advancement of Social Media, in partnership with an academic research team at Utrecht University, analysed more than 3,500 verified cases of digital rights violations logged by users through 7amleh’s own observatory, 7or.

The report coined the term “platformicide” to describe “the structural and intentional erasure of Palestinian digital presence, visibility, and participation across social media platforms”.

Fabio Cristiano, who led the report’s research and analysis at Utrecht University, said the report’s findings clearly show that the trend cannot be dismissed as a series of errors.

“This is an approach, a structural and intentional approach, which goes beyond content moderation alone. Meta’s erasure of Palestine operates through a broader architecture of exceptional governance, combining overbroad policy application, disproportionate enforcement, ad hoc policy revisions, algorithmic suppression, delayed restoration of content, and opaque or absent communication with users,” Cristiano told Al Jazeera.

“Platformicide is not simply about removing content. It is about denying a population access to the infrastructures through which they can claim their existence online.”

‘Dangerous Organizations and Individuals’

While the report does not assert that Meta is deliberately censoring users, it argues that the patterns of restrictions and deactivations of pro-Palestine accounts cannot be explained by error alone, and are structural – therefore meeting the authors’ definition of platformicide.

Of the cases documented between January 2021 and December 2025, only 32.5 percent had an identifiable policy violation attached to them. The majority of Palestinian users whose content was removed, restricted, or suppressed received no explanation they could point to.

The reasons given to users for their accounts being suspended follow a similar pattern: violation of Meta’s Dangerous Organizations and Individuals policy, known internally as DOI, which was applied in 57.2 percent of identifiable cases.

That figure rose to 71.1 percent among journalists. The policy, designed to prevent the promotion of terrorist and other violent organisations and individuals, was being used to moderate media outlets, NGOs, and reporters covering Israel’s genocidal war on Gaza, activists and experts claim.

“In this sense, the overbroad application of DOI seems to criminalise Palestinian society as a whole,” said Cristiano.

Punishing advocacy

Meta’s suspension of pro-Palestine accounts, removal of content and restriction of livestreaming capabilities have had a profound impact on journalists working in Gaza whose primary mode of communication with the outside world is through their social media platforms. The accounts are also an essential medium for getting news on the genocide out of Gaza, particularly given the ambivalence among some sections of the mainstream media about carrying pro-Palestine voices.

“Meta and platforms in general have a responsibility with regard to the public debate,” Cristiano said.

Elwan was banned three years ago from livestreaming on Instagram for 360 consecutive days with automatic renewals of the restriction since then.

The bans are ostensibly due to breaking community guidelines, although which specific rules she had allegedly breached have not been made clear by Meta, according to Elwan.

Back-up accounts set up by Elwan were subsequently suspended hours after being created.

“What’s happening is that they’re completely silencing journalists on the ground,” she said. “Many human rights organisations, the [United Nations], the [European Union], they used to collaborate with me to amplify voices or criticise certain issues. Now my content doesn’t reach anyone at all.”

The report found that account suspensions were the most common enforcement mechanism, affecting about a third of all documented cases. Restrictions on features such as going live were the second most common action, followed by content takedowns. After October 7, 2023 – the date of Palestinian group Hamas’s attack on Israel and the start of Israel’s war on Gaza – shadowbanning, the covert reduction of a user’s reach without notification, rose from near-zero to 3.5 percent of enforcement cases.

Maisoon Zoubi, a journalism teacher and media specialist who works with I’lam, a Palestinian media centre, described a different but related pressure: the chilling effect that visible enforcement creates on journalists who have not yet been targeted.

“A state of tension has been prevalent ever since we learned about cases of investigations, arrests, and page bans against journalists,” she said. “This makes me strengthen my own self-censorship on myself and on the interviews I conduct.”

During the war, Zoubi said, her team refrained from publishing events, images or reports that could trigger a complaint. The threat came not only from Meta’s automated systems but also from coordinated user reporting, which she described as “civilian censorship by ordinary people who monitor journalists and file digital complaints against them”.

Yousef Abu Watfa, the Gaza-based editor-in-chief of the Quds News Network (QNN), said Meta has systematically restricted his outlet’s content since 2015, with enforcement escalating sharply after the Gaza war began.

Abu Watfa said QNN’s Facebook page was taken down during the war. It had faced two earlier attempts to block it beforehand, which he attributed to an Israeli incitement campaign, and now nothing remains on Facebook in either Arabic or English.

On Instagram, only the Arabic-language account has survived, while the English-language version was removed from both platforms.

“This pushed us towards other platforms – X to some extent, Telegram, and WhatsApp to a lesser degree – though we’ve continued to face restrictions there as well,” Abu Watfa told Al Jazeera.

“All of this has certainly affected the Palestinian narrative. The war and the media coverage it has received have helped the images from Gaza reach the world, but what’s actually been conveyed doesn’t exceed 20 to 30 percent at best of the real picture of what Gaza has lived through.”

In response to a request for comment, a Meta spokesperson pointed towards independent due diligence reports on Israel and Palestine published in 2022, and implementation updates in its 2022–2024 human rights reports.

The company said it had introduced a system that routes flagged content to moderators most likely to understand the relevant Arabic dialect, and revised its Dangerous Organisations and Individuals policy to allow more social and political discourse in certain cases.

Lack of recourse

7amleh submitted 2,828 appeals to Meta on behalf of affected users over a five-year period but received no response from the social media giant for nearly half of those requests. 7amleh is a member of Meta’s Trusted Partner programme, which is designed to provide civil society organisations with preferential access to escalate cases. The report’s data challenges the value of that status.

Lama Nazeeh, advocacy manager at 7amleh, the Palestinian digital rights group that commissioned the report told Al Jazeera: “Independent reviews show the partners who get real responses are those with private contacts inside the company, not those using the official channel.”

“We hold our participation to one test: Does it protect Palestinian users, or does it protect Meta’s reputation?”

Meta is a notoriously difficult entity to contact despite the EU’s Digital Services Act and other international legal obligations requiring social media platforms to promptly communicate with users.

For better accountability, Utrecht University’s Cristiano said that Meta should not be treated as a remote private actor but as a platform accountable to its users.

“Private companies also have responsibilities. We do not need less content moderation, we need better content moderation, one that does not exceptionalise and criminalise Palestinian users and Palestinian content,” he said.

The removal of content from Gaza, such as videos of air strikes or the testimonies of survivors, affects evidence gathering for future accountability proceedings.

The Berkeley Protocol, adopted by the UN to guide the use of digital open-source information in international investigations, treats such material as potential evidence of war crimes. When that content is deleted, this evidence may become inaccessible, raising concerns for future accountability proceedings, and so it becomes an issue of international law.

“We preserve the fragments we can reach. Meta holds the entire archive of evidence and treats it as disposable,” Nazeeh said.

“A people’s right to document its own existence during a genocide is too important to leave to the discretion of a single company. The remedy is a change of framework: from voluntary commitments to binding obligations.”