Хэрэглэгчийн контентын зохицуулалтыг сулруулсан Meta компанийн шийдвэр нь хиймэл оюун ухаанаар бүтээсэн антисемит агуулгатай мэдээлэл түгэх таатай нөхцөлийг бүрдүүлж байна.
“Nexus Project”-ийн шинэ судалгаагаар Instagram платформ дээр еврейчүүдийн эсрэг чиглэсэн үзэн ядсан агуулга, хуйвалдааны онолууд эрчимтэй нэмэгдэж байгааг илрүүлжээ. Meta компани 2025 оны нэгдүгээр сард гуравдагч талын хяналтын байгууллагуудтай хамтран ажиллахаа зогсоож, хэрэглэгчдэд суурилсан зохицуулалтын загварт шилжсэн нь энэхүү сөрөг үзэгдлийг даамжруулсан гол хүчин зүйл болсон гэж үзэж байна. “Илүү их яриа, цөөн алдаа” гэх уриан дор хэрэгжүүлсэн энэхүү өөрчлөлт нь платформын аюулгүй байдлын хяналтыг сулруулж, хортой контентыг таслан зогсоох үүргийг хэрэглэгчдэд шилжүүлжээ.
Судалгааны багийнхан шинээр нээсэн бүртгэлүүд дээр хийсэн туршилтаар Instagram-ийн санал болгох алгоритм нь антисемит агуулгатай контентыг идэвхтэй дэмжиж байгааг тогтоосон байна. Тодруулбал, еврейчүүдтэй холбоотой түлхүүр үгсээр хайлт хийхэд хиймэл оюун ухаанаар бүтээсэн, үзэн ядалт төрүүлэхүйц дүрс бичлэгүүд хамгийн эхэнд гарч ирж, удалгүй хэрэглэгчийн нүүр хуудас бүхэлдээ ийм төрлийн контентоор дүүрчээ. Үүнд баталгаажсан бүртгэлүүдээс нийтэлсэн агуулга ч багтаж байгаа нь анхаарал татаж байна.
Эдгээр контентын ихэнх нь хиймэл оюун ухааны тусламжтайгаар хурдан бөгөөд хялбар бүтээгдэж, олон сая үзэлт авч байгаа нь мөнгө олох зорилготой байж болзошгүйг судалгаанд дурджээ. Meta компани эдгээр агуулгыг сурталчилгаа болон төлбөртэй үйлчилгээгээр дамжуулан ашиг олох боломжийг бүрдүүлж байна. Тус платформ нь хэрэглэгчдийн мэдээлсэн зүй бус контентыг устгахдаа хойрго хандаж байгаа нь асуудлыг улам хурцатгаж байна.
as a push often fails to take down reported by users at all associated with Meta Is Refusing to Remove AI-Generated Videos of Coaches Inappropriately Touching Young Athletes When Users Report Them
Дэлгэрэнгүйг эх сурвалжаас харах
↓Эх сурвалжийг нээх ↓
Antisemitism is running rampant on Instagram in the aftermath of Meta’s decision to limit content moderation on the social media site, according to a new report from the Nexus Project, an antisemitism research group.
According to the report, much of the antisemitic content on Instagram is AI-generated, resulting in a one-two punch: AI makes it extremely fast and easy to generate content that paints Jewish people in a nefarious light, while the lack of moderation allows for that overwhelming tide of AI-generated material, as well as anti-Jewish conspiracies and other hateful content, to flourish. Antisemitic content is also being promoted by Meta’s recommendation algorithms, per the Nexus Project, meaning that engaging with antisemitic conspiracies and propaganda will lead Instagram users to more and more of it.
“The scale and potential impact of this shift cannot be overstated,” reads the report. “While social media companies like Meta were always lax in their enforcement of hate content, recent changes have transformed the landscape.”
The report contends that Meta’s rollback of content moderation efforts back in January 2025, when it ended a partnership with third-party moderators and moved to a user-driven model, was an inflection point in the tide of antisemitic material. Billed as a push to promote “more speech and fewer mistakes,” Meta asserted that its moderation efforts had gone “too far” and resulted in too much censoring of “harmless content.” The sweeping change led to a far more hands-off approach to platform safety that put a heavy onus on users to detect and limit harmful content like hate speech, abuse material, and misinformation —and Meta often fails to take down abusive material reported by users at all.
Per the report, the Nexus Project started several brand-new Instagram accounts in order to test how the platform would respond to searches for content related to Jewish people. In one case, on a new account, the very first result for the term “synagogue” was an “AI rabbi welcoming the Antichrist, with over a million views.” After less than an hour, according to the report, nearly every single post on the new account’s homepage was antisemitic, with over half of that content coming from verified accounts.
“After less than an hour of watching a series of antisemitic Reels,” reads the report, “nearly our entire feed was devoted to antisemitic content with messages like ‘Jews started LGBTQ’ and a video telling viewers, ‘It’s time to wake up GOYIM.’” Goyim is a word for non-Jewish people associated with antisemitic messaging.
The Nexus Project also found that under antisemitic posts, Instagram suggested that users explore antisemitic conspiracies such as “goyim brainwashing” and “Dancing Israelis.” Under “trending” topics, the platform also recommended that they explore the Nazi marching song “Erika.” Thousands of Reels have used the song, according to the report.
The report also noted that many viral posts cut podcast clips and other video content to make it appear as though Jewish people have said or agreed with something they never did, for example the notion that the Jewish community secretly runs the world. Much of this content is also clearly geared at monetization, per the report, as accounts and creators capitalize on antisemitic content and conspiracy theories to push products such as supplements, e-books, courses, and subscriptions. Further, between pay-to-play checkmarks and ads, Meta stands to profit from the material as well.
In our testing, Nexus Project’s claims appeared to be accurate.
When we looked up “synagogue” on a decoy Instagram account, an AI-generated video of a large satanic-looking figure standing in a place of worship, covered in blood and being celebrated by Jewish men, was among the first results. The search term “Jew” returned a slew of despicable posts that presented cartoonish distortions of Jewish people that resembled Nazi propaganda from accounts devoted to conspiratorial and hateful content. When we typed “Jews” into the search bar, Instagram recommended that we search “jews controll [sic] everything.”
When we looked up the “Erika” song, the search returned troves of antisemitic content, including an AI-generated videos of students and fans at a high school football game doing the Sieg Heil salute to a rendition of the Nazi anthem. Some accounts promoting the song included links to download “Erika” ringtones in their bios.
The antisemitic videos that these searches returned collectively boasted millions of views. When we returned to the same account the next day, our feed was packed with even more hateful content.
More on Meta: Meta Is Refusing to Remove AI-Generated Videos of Coaches Inappropriately Touching Young Athletes When Users Report Them
The post Instagram’s Algorithm Is Promoting Antisemitic Content, Report Finds appeared first on Futurism.

