If artificial intelligence could stand in for human actors, what would remain of warmth?
En China, la frontera entre la creatividad humana y la automatización se ha vuelto un campo de disputa concreta: iQiyi lanzó NaDou Pro, una base de datos que promete conectar creadores con actores dispuestos a participar en proyectos generados por inteligencia artificial, pero varios de esos actores negaron públicamente haber dado su consentimiento. Lo que comenzó como una herramienta de mercado se convirtió en un espejo de una ansiedad más profunda: si las máquinas pueden interpretar, ¿qué queda del arte como acto humano? La industria del entretenimiento chino no enfrenta solo una controversia tecnológica, sino una pregunta sobre el valor de la presencia y la identidad en la era de la simulación.
- iQiyi afirmó tener más de 100 actores inscritos en su plataforma de IA, pero en menos de 24 horas varios artistas y sus representantes desmintieron públicamente haber firmado acuerdo alguno.
- En Weibo, la controversia detonó un debate masivo sobre si la inteligencia artificial debería reemplazar a los actores o simplemente asistirlos, revelando una fractura entre la industria tecnológica y los trabajadores creativos.
- El Comité de Actores de la Federación China de Sociedades de Radio y Televisión ya había advertido en marzo sobre riesgos concretos: deepfakes, clonación de voz y uso no autorizado de imágenes para entrenar modelos de aprendizaje automático.
- iQiyi intentó calmar las aguas aclarando que la inscripción en NaDou Pro no implica autorización automática, sino solo disposición a negociar proyecto por proyecto, aunque la credibilidad de esa distinción quedó en entredicho.
- La tensión no se ha resuelto: China enfrenta ahora la presión de regular la IA en el entretenimiento antes de que la brecha entre innovación tecnológica y protección laboral se vuelva irreparable.
El lunes, iQiyi anunció el lanzamiento de NaDou Pro, una base de datos pensada para conectar creadores de contenido con actores dispuestos a participar en proyectos generados por inteligencia artificial. La plataforma aseguró que más de cien intérpretes ya se habían sumado. Al día siguiente, esa afirmación comenzó a desmoronarse: estudios y clubes de fans de varios actores publicaron desmentidos, negando haber autorizado el uso de sus imágenes o voces para proyectos de IA. La contradicción encendió una conversación que iba mucho más allá del malentendido contractual.
El vicepresidente senior de iQiyi había presentado la iniciativa como un mercado transparente: los creadores podrían explorar un catálogo de actores disponibles, negociar condiciones y obtener autorización explícita para cada proyecto. Pero la distancia entre lo que la empresa anunció y lo que los actores dijeron haber acordado generó una fricción inmediata. En Weibo, los usuarios se preguntaron si el arte podía seguir siendo arte cuando la actuación es generada por una máquina, mientras otros argumentaban que la IA debería servir a la creatividad humana, no sustituirla.
El malestar no era nuevo. En marzo, el Comité de Actores de la Federación China de Sociedades de Radio y Televisión había emitido un comunicado formal señalando preocupaciones precisas: el uso de deepfakes para suplantar rostros, la clonación no autorizada de voces y la recopilación de imágenes y audio para entrenar modelos de inteligencia artificial. Eran daños concretos, no hipotéticos: la pérdida de control sobre la propia identidad, el riesgo de suplantación, la erosión del consentimiento.
Ante las críticas, iQiyi aclaró que inscribirse en NaDou Pro solo indicaba disposición general a considerar proyectos de IA, no una autorización en blanco. Cada papel, cada producción, requeriría negociación y aprobación por separado. Si esa distinción resultaría suficiente para restaurar la confianza era una pregunta sin respuesta. Lo que la controversia dejó en claro es que la tecnología ya existe y funciona, pero la industria y su público no están listos para aceptarla como sustituto de la actuación humana. La forma en que China resuelva esta tensión en los próximos meses podría definir cómo se regula y despliega la IA en toda la economía creativa.
On Monday, iQiyi, one of China's largest streaming platforms, announced the launch of NaDou Pro, a database designed to match content creators with actors willing to participate in artificial intelligence-generated projects. The company claimed more than 100 performers had already signed on. By Tuesday, the claim had begun to unravel. Studios and fan clubs representing multiple actors posted statements denying they had ever agreed to license their images or voices for AI work. The contradiction sparked a broader conversation online about whether machines should be making art at all.
The initiative, as iQiyi's senior vice president Liu Wenfeng described it at the company's annual conference, was meant to be a straightforward marketplace. Creators interested in AI-generated content could browse a roster of willing performers, negotiate terms, and secure independent authorization for each project. The platform framed it as a tool for connection, not replacement. But the gap between what the company announced and what actors said they had actually agreed to created immediate friction.
The backlash was swift and pointed. On Weibo, China's dominant social media platform, users raised questions that went beyond the immediate controversy. One comment captured the unease: if artificial intelligence could stand in for human actors, what would remain of the warmth and humanity that art is supposed to carry? Others pushed back, arguing that AI's true purpose was to serve human creativity, not displace it. The debate revealed a deeper anxiety about the technology's role in an industry built on human performance and presence.
China's entertainment sector has been grappling with AI's implications for months. In early March, the Actors Committee of the China Federation of Radio and Television Societies issued a formal statement raising specific concerns: the use of deepfake technology to swap faces, the unauthorized cloning of voices, and the harvesting of actors' images and audio to train machine learning models. These weren't abstract worries. They pointed to concrete harms—the loss of control over one's own likeness, the potential for impersonation, the erosion of the boundary between consent and exploitation.
IQiyi responded to the criticism by clarifying its position. Joining the NaDou Pro database, the company said, simply indicated a performer's willingness to consider AI projects. It did not constitute blanket permission. Each specific role, each individual project, would still require separate negotiation and explicit approval. The company was attempting to distinguish between a general interest in the space and actual participation in any given work. Whether that distinction would hold in practice remained unclear.
What emerged from the controversy was a fundamental tension in Chinese entertainment. The technology exists. It works. It can generate convincing performances at a fraction of the cost and time of traditional production. But the industry's workers—and the public watching them—were not yet ready to accept it as a replacement for human actors. The question of whether AI should augment human creativity or substitute for it had moved from the realm of speculation into the realm of immediate, practical conflict. The answer China's entertainment sector arrives at in the coming months will likely shape how the technology is regulated and deployed not just in streaming, but across the entire creative economy.
Citas Notables
If the actors are substituted by AI, what warmth will remain in artistic works?— Weibo user, quoted by The Paper
The original intention of AI is to serve humanity better, not replace it— Weibo commenters defending the technology
La Conversación del Hearth Otra perspectiva de la historia
Why would iQiyi announce something like this if they hadn't actually secured agreements from the actors they claimed were on board?
That's the question everyone's asking. The most charitable reading is that the company conflated interest from agents or studios with actual signed contracts. The less charitable reading is that they announced first and hoped to negotiate consent afterward.
So actors are saying they never agreed to this at all?
Not all of them. Some studios and fan clubs posted denials, which suggests at least some of the 100 names on that list didn't authorize their inclusion. Whether it's all of them or just a few, the damage to trust is already done.
What's the actual fear here? Is it that AI will replace actors, or something else?
Both. There's the existential worry—will machines do this job cheaper and faster, leaving actors without work? But there's also something more immediate: the loss of control over your own face and voice. If your likeness can be trained into a model without your permission, you've lost something fundamental.
The company says each project still needs separate approval. Does that solve the problem?
It helps, but only if it's enforced. The real issue is that once your image is in a training database, the genie is out of the bottle. Separate approvals for individual projects might protect you going forward, but it doesn't undo the initial violation of consent.
What does the actors' committee actually want?
They want clear rules. No deepfakes without consent. No voice cloning for model training. No harvesting of images and audio without explicit permission. Basically, they want the technology to respect the same boundaries that have always protected performers' rights.
Is China likely to regulate this?
Almost certainly. The government has shown it's willing to step in when tech companies move too fast. But the real question is whether the rules will protect workers or just manage the technology's growth.