The annual iQiyi World Conference always aims to make waves. In 2025, the company faced widespread online criticism for its "Skip & Watch" feature. In 2026, it staged a dramatic sequence of events: a high-profile announcement, a swift rebuttal, and an overnight attempt at damage control. This time, the controversy revolves around AI.
On April 20, iQiyi founder Gong Yu took the stage with great ambition, painting a picture of a utopian film and television industry reshaped by AI. He announced the official launch of an AI Actor Library and the commercial opening of the Nadou Pro platform, claiming hundreds of performers had already "joined." In his vision, creators could freely utilize digital replicas of stars, drastically reducing the barriers and costs of production.
However, immediately after this announcement, a digital wall displaying performer photos at the conference ignited a firestorm. Fans were the first to react vehemently in online communities, questioning whether their favorite stars had sold their digital rights. Social media platforms were quickly flooded with skepticism. The situation escalated when actors themselves began to publicly distance themselves. Stars including Zhang Ruoyun, Yu Hewei, and Wang Churan issued statements through their studios, denying having signed any AI authorization agreements—directly contradicting iQiyi's claims.
In the early hours of April 21, iQiyi was forced into emergency public relations, awkwardly redefining the term "joined" to mean merely "having expressed interest in discussions." Overnight, the narrative shifted from a presented fait accompli to what appeared to be wishful thinking.
The phrase "iQiyi has gone crazy" trended online. Gong Yu's comment that "live-action filming might become intangible cultural heritage in the future" was interpreted by netizens as "actors will be replaced by AI." Comment sections exploded with reactions like, "If actors can be AI, then let AI audiences pay for memberships." Gong Yu posted three consecutive Weibo messages emphasizing that "technology is never meant to replace people," but all three had comment sections selectively filtered—a move that spoke volumes.
Why did a technological solution ostensibly aimed at "reducing costs and increasing efficiency" provoke such a strong backlash? What specific missteps did iQiyi's high-stakes AI gamble take?
The core issue touches upon performers' most vital assets. An actor's primary value lies not merely in their fee but in their unique image, performance style, and the public recognition built over years. The essence of an AI Actor Library is to transform these irreplaceable human qualities into infinitely replicable, combinable, and distributable digital assets. Critical questions arose: Who holds the copyright to AI-replicated performances? Could biometric data be vulnerable to theft by third parties? What happens to a digital likeness after an authorization expires? iQiyi's announcement provided no answers.
Gong Yu's remarks added to the anxiety. He suggested that by authorizing AI collaborations, actors could work more regular hours, potentially taking on more projects annually, albeit for lower pay per project. For top-tier actors accustomed to high fees, "earning a bit less" is not an attractive proposition. A deeper concern is that if AI matures sufficiently, platforms could bypass expensive A-list talent altogether, collaborating instead with mid-tier actors or even amateurs and using AI to enhance their performances. Some industry figures have openly stated that avoiding costly top actors leads to "purer" content. Certain productions already use unknown actors for initial performances, which are then processed by AI, with directors claiming the results are "no worse than an average actor's performance." If this model proves successful, the bargaining power of top actors could be severely diminished.
To understand iQiyi's aggressive push into AI, one must examine its precarious position. In 2025, iQiyi's revenue was 27.29 billion yuan, marking a second consecutive year of decline. More alarming was the profit situation: Non-GAAP operating profit plummeted by over 70% year-over-year to 620 million yuan, while GAAP operating profit fell 87.3% to 229 million yuan. The company swung to a net loss of 206.3 million yuan, a stark contrast to the profit of 764.1 million yuan in 2024. All major business lines—membership, online advertising, content distribution, and others—weakened. Its market capitalization has evaporated over 95% since its 2018 IPO, now standing below $1.4 billion.
The traditional long-form content market is shrinking. Meanwhile, short-form video platforms are booming, with competitors like ByteDance's Hongguo Short Drama reaching hundreds of millions of monthly active users. Furthermore, viewers can often get the gist of a series through "sliced" clips on short-video platforms, reducing the incentive to pay for long-form content on sites like iQiyi.
Producing hit shows is capital-intensive and high-risk. While iQiyi has had past successes, creating a phenomenon like "The Knockout" is rare. Gong Yu's "Media 112 Principle" envisions content costs reduced to one-tenth or even one-twentieth, with a tenfold increase in creators and a hundredfold increase in output. Given iQiyi's 2025 content costs of 15.45 billion yuan—consuming over half its revenue—the potential savings from AI are undeniably compelling for a company struggling near the break-even point.
Despite the backlash, iQiyi's stance remains "All in AI." Essentially, it has few other compelling narratives left. The company is part of the Baidu ecosystem, whose founder, Robin Li, has championed aggressive AI integration. With limited external support, iQiyi must forge its own path.
iQiyi's decision to heavily bet on AI is not inherently wrong. The misstep lay in the overly rough handling of information disclosure and an overconfident communication style. The rapid transition from a highlight announcement to emergency clarification exposed not just a public relations weakness but also an industry grappling with how to balance commercial interests against human dignity in the face of technological change.
The wave of AI in entertainment is rising, and no one can stop it. However, the direction of this wave depends on the choices made by every participant. iQiyi's gamble is not just on its own future but also on the industry's collective judgment regarding the value of human contribution.
Comments