AI-Generated Black Female Avatars Spark Racism Concerns Across Social Media

TikTok has deleted 20 accounts following a BBC investigation that revealed a troubling trend of AI-generated black female avatars being used to direct users to adult content. The social media platform acted after the BBC and researchers from the AI research outlet Riddance discovered dozens of accounts across Instagram and TikTok displaying highly sexualised digital characters with exaggerated body shapes and darkened skin tones. The accounts, which were unlabelled as AI-generated in apparent breach of platform policies, employed racial stereotypes in their naming practices and marketing copy. Whilst TikTok has responded quickly, Meta, Instagram’s parent company, said it was investigating but has failed to confirm taking comparable steps against the accounts active on its platform.

The Discovery: Hundreds of Fraudulent Accounts

The BBC’s inquiry, conducted in collaboration with Riddance researchers Jeremy Carrasco and Angel Nulani, uncovered a complex scheme spanning multiple platforms. The team discovered 60 accounts mainly found on Instagram that featured links and chains of links directing users to commercial adult content on external sites. Notably, whilst these third-party sites described the images as AI-generated, the Instagram accounts themselves offered no comparable clarification, establishing intentional misleading for unaware individuals. The research exposed an even larger ecosystem of comparable profiles on Instagram and TikTok alike that failed to direct users to paid content, suggesting the problem goes well past profit-driven abuse.

The accounts employed a calculated strategy to escape notice and expand reach quickly. Nearly all accounts were concentrated on Instagram, with approximately one-third also maintaining versions on TikTok. Account names deliberately invoked racial terminology, including terms such as “black”, “noir”, “dark” and “ebony”, alongside language promoting stereotypical depictions and sexualisation. Many accounts followed or engaged with each other, forming a coordinated web that increased their visibility and online presence. This organised strategy suggests organised activity rather than isolated incidents, prompting significant concerns about the scale and sophistication of the activity.

• AI avatars displayed distorted physical proportions and darkened skin tones
• Account names employed racial terminology and stereotypical language about white men
• Videos lacked labels as AI-generated, violating platform guidelines
• Many accounts followed each other, establishing coordinated systems for reach enhancement

Abuse Through Artificial Generation

Illicit Content and Online Alteration

The investigation exposed a especially concerning dimension to the scheme: the wholesale theft of material from authentic content creators. One profile that gathered three million followers within weeks of its December launch had methodically stolen footage from real people, most notably Malaysian model Riya Ulan. The individuals responsible overlaid the artificial avatar’s face—showing an darkened skin tone—onto Riya’s form, precisely duplicating her gestures, attire and backdrop. This brazen act of digital theft intensified the original exploitation, turning authentic creative work into material for deceptive accounts.

Riya’s experience illustrated the individual harm fundamental to such activities. When she found her videos were appropriated and redistributed, she communicated her frustration to the BBC: “I was upset. Of course my videos are all out there… It doesn’t mean that you can simply steal it and repurpose it and share it as your own.” Her distress reveals a critical gap in safeguarding systems, where creators miss sufficient defences against their identities being misused for fake profiles. The incident reveals how artificial intelligence amplifies current abuses, enabling bad actors to scale their exploitation across many creators at once.

The exploitation surpassed basic content piracy. Accounts deliberately constructed artificial personas employing exaggerated bodily features and synthetically modified skin tones to generate compelling, visually arresting material intended to drive user participation. These algorithmic creations were marketed as authentic influencers, complete with false narratives and character traits, at the same time perpetuating harmful racial stereotypes and objectifying narratives. The intricacy of the manipulation caused numerous people engaged with the posts thinking they were following real people, not programmed entities designed specifically to direct users to exploitative paid content.

• Avatar’s face overlaid onto pirated content from legitimate creators
• Artificially darkened complexions and pronounced characteristics intended to drive engagement
• Accounts marketed as authentic influencers with false backgrounds and personas

Racial Stereotypes and Prejudicial Labelling

The accounts discovered by the BBC and Riddance researchers exhibited a troubling pattern of discriminatory targeting through their intentional account names and content. Account identifiers incorporated terms such as “black”, “noir”, “dark” and “ebony”, whilst posts consistently displayed racialized commentary and fetishizing content. Many featured comments such as “loves white men” and “why I need a white guy in my life”, sustaining racist tropes and portraying Black women to sexualized objects. This linguistic framework rendered the profiles into stereotypical representations, sustaining historical racist tropes that objectify and strip of humanity Black femininity.

The visual presentation intensified these linguistic harms. The avatars were consistently depicted in revealing swimwear and skimpy clothing, their bodies digitally manipulated into exaggerated proportions designed to maximise interaction via sexual imagery. Complexions were artificially darkened to produce a fabricated, near monstrous appearance that bore little resemblance to genuine human diversity. This mix of exaggerated features, exposed clothing and synthetically altered traits produced algorithmic stereotypes that functioned as contemporary racist performance, transforming racist imagery into algorithmic content optimised for viral spread and monetisation.

Why AI Amplifies the Problem

Artificial intelligence technology has substantially altered the scale and sophistication of racist exploitation online. Previously, such harmful content required significant resources and manual effort to produce; AI generation facilitates the development of racist caricatures, enabling bad actors to produce large quantities of believable fake accounts at minimal cost. The technology’s capability for producing photorealistic imagery imparts unwarranted credibility to racist tropes, making them appear credible to unsuspecting users. This technological advantage transforms what might have been fringe exploitation into a scalable, profitable enterprise.

The algorithmic structure of social media platforms intensifies these harms exponentially. AI-generated content tailored for user interaction—particularly content that leverages racial and sexual stereotypes—proliferates quickly through recommendation systems designed to prioritise user interaction. Platforms face challenges moderating the sheer volume of AI-generated content, whilst the lack of mandatory disclosure labels means users cannot distinguish authentic creators from algorithmic constructs. This creates an environment where racist stereotypes proliferate unchecked, obscured by the veneer of digital advancement and framed as entertainment rather than exploitation.

Platform Responsibility and Accountability

TikTok’s rapid action in taking down 20 accounts constitutes a infrequent example of platform enforcement, yet it underscores the reactive nature of content moderation in the social media era. The bans came only after sustained media scrutiny and public demand, indicating that absent external scrutiny, these accounts would have remained active unchecked. Notably, TikTok’s response was restricted to a single platform, whilst many of the offending accounts kept active profiles on Instagram, where they kept collecting followers and sending users to exploitative third-party sites. This piecemeal strategy demonstrates the insufficiency of isolated platform measures to structural issues.

Meta’s claim to be investigating, paired with its inability to declare concrete action, demonstrates the divide between corporate responsibility messaging and genuine enforcement. The company’s sluggish reply stands in stark contrast to the pressing need to tackle swiftly spreading damaging material. Both platforms have created rules stipulating the announcement of machine-generated visual content, yet these rules remained unenforced across thousands of accounts. This implementation shortfall indicates that without legislative action and legal accountability, platforms will keep favouring traffic statistics and commercial returns over platform security and the protection of marginalised communities from systematic exploitation and racist imagery.

Extended Implications for Digital Integrity

The rapid growth of unlabelled AI-generated material on major social media platforms presents fundamental challenges about online integrity and audience confidence. As artificial intelligence becomes more advanced, distinguishing between genuine human creators and artificial personas becomes significantly harder for everyday people. This erosion of authenticity threatens the foundational premise upon which social media communities are built—the belief that accounts represent real people posting authentic content. When vast numbers of people unknowingly engage with fabricated personas created to manipulate them, the networks’ trustworthiness as trustworthy spaces for interaction declines sharply. The absence of mandatory disclosure mechanisms effectively deceives audiences and compromises informed consent.

Beyond personal deception, the widespread deployment of artificially created black female avatars constitutes a especially pernicious exploitation of racial identity in digital spaces. These artificial characters weaponise damaging generalisations and hypersexualised racial tropes whilst concurrently appropriating real creators’ content and labour. The practice reinforces the commercialisation of black female identity within algorithmic systems designed to maximise engagement and profit. This technological racism operates extensively, affecting vast populations whilst staying largely undetected to content moderation departments. Without robust regulatory structures requiring openness, verification protocols, and meaningful consequences for violations, digital spaces will keep enabling systematic harm against marginalised communities.

• AI disclosure requirements must be legally enforceable on every platform worldwide
• Authentication systems must confirm creator authenticity ahead of earning potential and audience expansion
• Racial stereotyping in synthetic content requires direct restrictions and active detection
• Platforms need to deploy real-time moderation with consequences matching content severity