Хиймэл оюун ухаанаар үүсгэсэн, хүүхдийг баривчлах дүр зураг бүхий хуурамч дүрс бичлэгүүд Facebook болон Instagram сүлжээнд түгээмэл болж, хэрэглэгчдийн сэтгэл хөдлөлийг ашиглан “лайк” болон дагагч цуглуулах хэрэгсэл болж байна.
Meta платформ дээр “Emotional Tots”, “Tiny Miracle” зэрэг хаягууд хиймэл оюун ухаан ашиглан хүүхэд дэлгүүрээс талх хулгайлсны улмаас цагдаад баривчлагдаж буй хуурамч бичлэгүүдийг идэвхтэй түгээж байна. Эдгээр бичлэгт дүрслэгдсэн цагдаагийн ажилтан хүүхдийг “хангалттай лайк, дагагчгүй” гэсэн шалтгаанаар баривчилж буйгаа хэлж, үзэгчдийг туслахыг уриалдаг. Энэхүү аргачлал нь хэрэглэгчдийн өрөвдөх сэтгэлийг ашиглан хандалт авах зорилготой бөгөөд олон хэрэглэгч уг бичлэгийг хиймэл оюун ухаанаар бүтээгдсэнийг ялгаж чадахгүй байна.
Энэ асуудлаар тодруулга авахад Meta компанийн хэвлэлийн төлөөлөгч бодлого зөрчсөн тохиолдолд хиймэл оюун ухаан эсвэл хүний бүтээсэн эсэхээс үл хамааран контентыг устгана гэж мэдэгдсэн юм. Гэсэн хэдий ч тус компани эдгээр бичлэг нь компанийн бодлогыг зөрчиж байгаа эсэх талаар тодорхой хариулт өгөхөөс татгалзжээ. Одоогоор Meta компани зарим хаягийг устгасан боловч ижил төстэй контент бүхий бусад хаягууд нь платформууд дээр хэвээр байна.
Шинжээчид Meta компанийг өмнө нь хүний эрхийн зөрчил болон арьс өнгөөр ялгаварлан гадуурхсан агуулга бүхий контентуудад хангалттай арга хэмжээ аваагүй гэж шүүмжилдэг. Иймд хүүхдийн дүр ашиглан сэтгэл зүйгээр манипуляц хийж буй энэ төрлийн “клик-ферм”-үүдийг зогсоох тал дээр тодорхой ахиц гарах эсэх нь эргэлзээтэй хэвээр байна.
Дэлгэрэнгүйг эх сурвалжаас харах
↓Эх сурвалжийг нээх ↓
In the age of AI, Meta platforms like Facebook and Instagram have become breeding grounds for content farms peddling in emotionally charged AI slop.
Some click mills traffic in racist incitement against minorities, others in violent depictions of child abuse, but all of them converge on a common quality: bizarre and disturbing attempts at attention-hacking users on a platform that’s all but given up on proactive moderation.
A new case in point is the rise of AI-generated clips depicting seemingly homeless children being arrested after stealing morsels of food from a store.
Shared by accounts with names like “Emotional Tots” and “Tiny Miracle,” these AI Reels follow a simple formula: a child runs in or out of a store, grabbing some basic grocery staple, often a loaf of bread. They’re then accosted by an authority figure, like a police officer or security guard, who declares that the kid doesn’t have enough likes and followers — and if viewers will just help out, they’ll be free to go.
“You are under arrest,” an AI-generated officer in one viral example declares to a crying child hugging a loaf of bread. “Why? I didn’t do anything,” the kid responds.
“You don’t have enough followers and shares on your videos,” the officer explains. “If the person watching this video follows and shares, then you will be free.”
“Please give me one more chance,” the child sobs. “I don’t want to go to jail… Please share and follow.”
Another clip shows a toddler running off with a sack of fruit before being accosted by a cop.
“It’s for my grandma,” the AI-generated child stammers. “Please don’t take it, my grandma is hungry and my mom is gone.”
The officer turns to face the viewer: “Should I let him take these to his grandma? Follow, like, and comment below.”
A Meta spokesperson repeatedly asked for more time to respond to questions about this story after we provided the company with links to several specific videos. After several days the company pulled down two accounts we flagged, but left another online, complete with many more similar videos.
“So far, we’ve removed two of the three links you shared,” a Meta spokesperson told us, days after Futurism approached them for comment. “We will remove content, regardless of whether it is created by AI or a person, if it violates our policies.”
The spokesperson declined to say whether videos of sobbing children begging for likes and follows under the threat of arrest violate the company’s policies.
In showing police apprehend starving street children, the videos unintentionally play on a long-standing critique in Western countries: that police exist less to help people — in this case emaciated children — and more to protect private property, for example by recovering stolen bread. As a result, comments under each video authored by humans and bots alike interject to demand justice on the toddlers’ behalf, in many cases seemingly unaware that the entire video is a work of AI fiction.
“Let baby have the bread,” one comment with 45 likes demanded. “He didn’t do any thing, give him chance; no jail no jail no jail please please, he is a a baby,” another user entreated.
Others didn’t seem to understand what they were looking at, either metaphorically speaking, or even just literally: “Big boys I do follow because the officer is a good guy,” one seemingly human account created in 2019 replied to a similar Reel. “Amen.”
Considering the breadth of human rights abuses and flagrantly racist content Meta allows to proliferate on its platforms, AI clip farms depicting fake starving children aren’t exactly the most egregious thing —but they certainly rank among the most emotionally manipulative. Given Mark Zuckerberg’s company’s inaction in the former cases, it seems highly unlikely that anything will be done about the starving child click-farms, allowing this soul-sucking content to spread on Meta, Instagram, and beyond.
More on Meta: Meta’s Muse Is Spying on All Your Friends and Family
The post Meta Won’t Say If It Violates Its Policies to Upload AI-Generated Videos of Sobbing Little Children Begging for Likes and Follows So They Don’t Get Arrested by Police appeared first on Futurism.
