In China, a new marketplace is emerging where human faces are treated as digital assets, much like stock photographs. Platforms such as ActID and New Claw allow individuals to license their likeness to artificial intelligence systems, enabling AI creators to generate realistic characters for short dramas, advertisements, and other content. According to Rest of World, contributors have earned between $15 and $700 through these platforms. For studios, this model provides a steady supply of virtual actors without the logistical challenges of traditional shoots. Contributors receive defined payment terms, but once the facial data leaves the platform, the agreement cannot guarantee that every buyer will remain compliant with usage restrictions.

How Does Renting a Face Work?

ActID, one of the leading platforms in this space, reportedly has around 800 registered users, including roughly 300 people who have licensed their likenesses. On this platform, creators pay between 99 and 500 yuan per episode to use a licensed face for AI-generated performances. ActID takes a 10% commission on each transaction. New Claw, another platform, allows contributors to upload their own images and set a minimum licensing price of 500 yuan per image. Buyers can then use those likenesses in AI projects, provided they adhere to the contractual agreement.

The fundamental difference between a stock photo and a licensed AI likeness is flexibility. A stock photo captures a single expression in one setting, whereas a licensed AI likeness can be manipulated to speak, move, and appear as different characters across multiple episodes or campaigns. This gives buyers far more creative freedom but also creates more opportunities for misuse. Contributors must track where their digital faces appear, which can be challenging given the scale of AI-generated content.

Why Does AI Need So Many Faces?

China's appetite for AI-generated entertainment has grown rapidly, creating demand for a vast supply of realistic virtual actors. According to figures from the China Netcasting Services Association, AI was used in more than 95% of the 128,000 microdramas released during the first quarter of 2026. Studios producing content at this scale need characters they can license and deploy quickly, often without the time or budget for traditional casting. One photo session can lead to an unusually long acting career, minus the trailers, press tours, and other overhead of real-world production.

The rise of AI-generated microdramas is part of a broader trend in Chinese entertainment. Short-form video platforms like Douyin have popularized bite-sized storytelling, and AI production tools have lowered the barrier to creating episodes. Licensing faces through these platforms allows creators to bypass the complexities of hiring actors, such as scheduling, contracts, and travel. The result is a streamlined pipeline from concept to finished product, which is critical in a market where thousands of new microdramas are released each week.

Beyond microdramas, licensed faces are also used in branded content, live-streaming avatars, and interactive virtual hosts. These applications are particularly common in e-commerce, where AI-generated presenters can promote products around the clock without fatigue. The scalability of AI actors makes them attractive for companies that need consistent, cost-effective spokespeople.

Can You Get Your Face Back?

Chinese law provides some protections for individuals who license their biometric data. The Personal Information Protection Law (PIPL) requires explicit consent before biometric data can be processed, and it prohibits fabricating someone's likeness without permission. Face-licensing platforms add contracts and technical restrictions intended to keep buyers within agreed uses. However, enforcement is far from perfect. The Guangzhou Internet Court has handled around 700 AI-related disputes over three years, including cases involving face-swapping and copyright violations. These cases highlight the gap between legal theory and practical enforcement.

One specific challenge is the difficulty of tracking unauthorized use once facial data has been distributed. Digital watermarking and blockchain-based tracking are being explored, but these technologies are not yet widely adopted. Contributors often rely on reverse image searches or manual monitoring to find misuse, which is time-consuming and unreliable. Even when misuse is discovered, legal recourse can be expensive and slow.

Another concern is the permanence of digital likenesses. Unlike a traditional acting career, which ends when the actor retires, a licensed face can persist indefinitely in AI-generated content. The original contributor may have no control over future applications, especially if the platform changes ownership or goes offline. Contracts can define permission boundaries, but they cannot erase every rogue version of a face that has been uploaded to the internet.

To mitigate these risks, experts recommend that individuals carefully review platform policies before licensing their faces. Contributors should look for platforms that offer clear tracking mechanisms, robust legal protections, and transparent reporting of usage. Some platforms also provide opt-out clauses that allow contributors to revoke a license under certain conditions, though these clauses can be difficult to enforce internationally.

The economic incentives are clear: for many people, especially students and gig workers, the chance to earn $15 to $700 from a single photo session is appealing. But the long-term implications of surrendering one's facial identity are less obvious. As AI technology advances, the line between authentic and synthetic faces will continue to blur, raising ethical questions about consent, privacy, and identity theft.

The Guangzhou Internet Court's caseload demonstrates that disputes over AI-generated faces are already common. In one notable case, a plaintiff sued a platform after their face was used in a sexually suggestive advertisement without permission. The court ruled in favor of the plaintiff, awarding damages and ordering the platform to remove the content. However, such victories are rare, and many victims never discover the misuse of their likeness in the first place.

Internationally, similar issues are emerging. In the United States, the recent use of AI-generated voices and faces in political advertisements has prompted calls for regulation, though no comprehensive federal law exists. The European Union's General Data Protection Regulation (GDPR) offers strong protections for biometric data, but enforcement varies by member state. China's approach, combining the PIPL with court decisions, provides a template that other countries may follow.

For now, the face-licensing market in China continues to grow. Platforms like ActID and New Claw are refining their business models, adding features such as revenue sharing and dynamic pricing. Some contributors have become repeat sellers, uploading multiple sessions to build a portfolio of faces. Studios, meanwhile, are experimenting with hybrid models that combine licensed faces with real actors, creating content that is part human, part machine.

As the technology matures, the need for standardized contracts and industry best practices will become more pressing. Trade associations and regulators have begun to discuss guidelines for AI-generated content, though progress is slow. Contributors who want to license their faces should stay informed about these developments and consider consulting legal experts before signing agreements.

Ultimately, the decision to license one's face to AI is a personal one, balancing immediate financial gain against potential long-term risks. The market offers opportunities for those who are cautious, but it also presents dangers for those who are not. As one contributor told Rest of World, "It's not a gold rush yet, but the door is open."

Source: Digital Trends News