Blab Politics
Metas Shift to Community Notes Sparks Surge in Antisemitic Content, Hits Jewish Creators
On Jan. 7 2025, Meta Platforms announced it would replace its proactive content‑moderation teams on Facebook, Instagram and Threads with a community‑based reporting tool called Community Notes. The move was presented as a trade‑off: fewer false‑positive removals at the cost of less immediate filtering of harmful posts.
The policy change coincided with a dramatic spike in antisemitic comments. Prior to the shift, Jewish members of Congress averaged 6.5 antisemitic remarks per day on their pages. By Feb. 4 2025, after the rollback, the average climbed to 29.9—an increase of more than thirteen times. On certain days researchers logged 701 toxic comments across 66 posts. In December 2025, Senators Ben Ray Luján, Martin Heinrich and Jacky Rosen sent Mark Zuckerberg a formal letter that highlighted a nearly five‑fold rise in antisemitic behavior and warned that 277 million posts previously removed would now remain live.
The Anti‑Defamation League (ADL) followed up with a detailed April 2026 report titled How Meta’s Content Moderation Changes Risk Turning Instagram into a Hub for Hate. The study found that only 7 % of reported extremist content was taken down after the policy shift. Instagram continued to surface antisemitic videos that amassed millions of likes and an estimated reach of 280 million users. One such video, which earned 191 000 likes and 184 000 shares, accused Jews of orchestrating conspiracies ranging from child‑sex trafficking to the 9/11 attacks.
A separate problem emerged in March when a Jewish creator discovered that a copyright claim filed under the name of The Walt Disney Company—using a Gmail address—triggered an automated enforcement action that removed her Facebook monetization. Her appeal was denied within minutes, and the platform labeled her account a “repeat offender” after a single post. The creator was then prevented from tagging the Disney CEO or the Lawfare Project, both active public accounts, and has reached out to both the Lawfare Project and the ADL for help.
Meta’s enforcement relies heavily on automation. Drop Site News reported in April 2025 that the company removed over 90 000 posts to comply with government takedown requests, with an average processing time of 30 seconds per post. The same automated systems have also been used to demonetize creators who produce content on Jewish history and counter‑extremism. The creator’s professional dashboard has been repeatedly disabled, and her subscriber feature shut down without notice.
The company outsources moderation and creator‑support operations to contractors in Africa, Asia and Eastern Europe. Investigations by the Bureau of Investigative Journalism and Foxglove documented conditions in these facilities, including surveillance and threats to workers who raise concerns. When a case exceeds a scripted response, agents transfer the user to a specialist team that often fails to resolve the issue.
Meta’s policy changes attracted shareholder scrutiny. In May 2025, a proposal demanding a report on how the company combats antisemitism received support from 46.8 % of independent shareholders—the highest level of backing for any human‑rights proposal that year. Zuckerberg, who holds 61.2 % of the company’s voting power, voted the proposal down.
Legal developments suggest Meta’s liability could widen. In March, a New Mexico jury found Meta liable for endangering children and ordered $375 million in civil penalties. A Los Angeles jury the following day held Meta 70 % responsible for harm caused to a minor addicted to social media. In August 2025, families of victims of the Hamas‑led Oct. 7, 2023 attacks filed a class‑action lawsuit in Tel Aviv seeking approximately $1.1 billion, alleging that Facebook and Instagram enabled livestreamed footage of murders.
The ADL’s 2023 platform study contrasted Meta with YouTube, showing that YouTube’s recommendation system does not amplify conspiracy‑adjacent content, whereas Meta’s does. The study implies that allowing algorithmic amplification of antisemitic content is a policy choice, not a technical limitation.
What remains unresolved is how Meta will respond to the surge in antisemitic content and the impact on creators who document Jewish history. The company has announced no plan to restore the previous level of proactive moderation. Until Meta revises its algorithmic recommendations, implements mandatory human review for account‑level enforcement, and publishes independent removal rates, Jewish creators and civil‑rights groups will likely continue to face automated suppression.
At present, Meta’s community‑notes system is in place, antisemitic content remains largely unchecked, and creators who produce counter‑extremism content are subject to automated takedowns. The next steps will hinge on policy revisions, shareholder pressure and potential legal actions that could compel Meta to change its enforcement practices.
The policy change coincided with a dramatic spike in antisemitic comments. Prior to the shift, Jewish members of Congress averaged 6.5 antisemitic remarks per day on their pages. By Feb. 4 2025, after the rollback, the average climbed to 29.9—an increase of more than thirteen times. On certain days researchers logged 701 toxic comments across 66 posts. In December 2025, Senators Ben Ray Luján, Martin Heinrich and Jacky Rosen sent Mark Zuckerberg a formal letter that highlighted a nearly five‑fold rise in antisemitic behavior and warned that 277 million posts previously removed would now remain live.
The Anti‑Defamation League (ADL) followed up with a detailed April 2026 report titled How Meta’s Content Moderation Changes Risk Turning Instagram into a Hub for Hate. The study found that only 7 % of reported extremist content was taken down after the policy shift. Instagram continued to surface antisemitic videos that amassed millions of likes and an estimated reach of 280 million users. One such video, which earned 191 000 likes and 184 000 shares, accused Jews of orchestrating conspiracies ranging from child‑sex trafficking to the 9/11 attacks.
A separate problem emerged in March when a Jewish creator discovered that a copyright claim filed under the name of The Walt Disney Company—using a Gmail address—triggered an automated enforcement action that removed her Facebook monetization. Her appeal was denied within minutes, and the platform labeled her account a “repeat offender” after a single post. The creator was then prevented from tagging the Disney CEO or the Lawfare Project, both active public accounts, and has reached out to both the Lawfare Project and the ADL for help.
Meta’s enforcement relies heavily on automation. Drop Site News reported in April 2025 that the company removed over 90 000 posts to comply with government takedown requests, with an average processing time of 30 seconds per post. The same automated systems have also been used to demonetize creators who produce content on Jewish history and counter‑extremism. The creator’s professional dashboard has been repeatedly disabled, and her subscriber feature shut down without notice.
The company outsources moderation and creator‑support operations to contractors in Africa, Asia and Eastern Europe. Investigations by the Bureau of Investigative Journalism and Foxglove documented conditions in these facilities, including surveillance and threats to workers who raise concerns. When a case exceeds a scripted response, agents transfer the user to a specialist team that often fails to resolve the issue.
Meta’s policy changes attracted shareholder scrutiny. In May 2025, a proposal demanding a report on how the company combats antisemitism received support from 46.8 % of independent shareholders—the highest level of backing for any human‑rights proposal that year. Zuckerberg, who holds 61.2 % of the company’s voting power, voted the proposal down.
Legal developments suggest Meta’s liability could widen. In March, a New Mexico jury found Meta liable for endangering children and ordered $375 million in civil penalties. A Los Angeles jury the following day held Meta 70 % responsible for harm caused to a minor addicted to social media. In August 2025, families of victims of the Hamas‑led Oct. 7, 2023 attacks filed a class‑action lawsuit in Tel Aviv seeking approximately $1.1 billion, alleging that Facebook and Instagram enabled livestreamed footage of murders.
The ADL’s 2023 platform study contrasted Meta with YouTube, showing that YouTube’s recommendation system does not amplify conspiracy‑adjacent content, whereas Meta’s does. The study implies that allowing algorithmic amplification of antisemitic content is a policy choice, not a technical limitation.
What remains unresolved is how Meta will respond to the surge in antisemitic content and the impact on creators who document Jewish history. The company has announced no plan to restore the previous level of proactive moderation. Until Meta revises its algorithmic recommendations, implements mandatory human review for account‑level enforcement, and publishes independent removal rates, Jewish creators and civil‑rights groups will likely continue to face automated suppression.
At present, Meta’s community‑notes system is in place, antisemitic content remains largely unchecked, and creators who produce counter‑extremism content are subject to automated takedowns. The next steps will hinge on policy revisions, shareholder pressure and potential legal actions that could compel Meta to change its enforcement practices.
