How Bollywood Became the World’s Biggest Live Experiment in AI Cinema, and What That Means for Hollywood
While Hollywood has been locked in debate over whether artificial intelligence belongs on a film set at all, India’s entertainment industry has already moved past the question entirely.
With no empowered unions pressing for caution and no national legislation introducing meaningful guardrails, Bollywood has become the most consequential real-world test of what happens when one of the planet’s most prolific film industries embraces AI without a safety net, according to a report from The Hollywood Reporter.
India produces more films than any other country, and the pressures accelerating the shift are stark. The number of moviegoers fell to 832 million in the most recent year measured, down from over a billion in 2019, while box-office revenue has remained choppy and reliant on a handful of hits and pricier tickets.
In response, Indian studios are deploying AI at a scale that surpasses much of Hollywood, using it to generate entire films, dub content into multiple languages, and even recut the endings of older titles.
The most provocative example of that last practice involved the romantic drama ‘Raanjhanaa’. Eros Media World re-released the beloved film with an AI-altered conclusion, replacing a tragic ending where the protagonist dies with a happier finale in which he opens his eyes to his lover’s tearful smile.
Lead actor Dhanush publicly condemned the move, stating the AI remake had stripped the film of its soul and set a deeply troubling precedent for art and artists. Under a SAG-AFTRA agreement in the United States, studios cannot digitally alter an actor’s performance or create a digital replica without the performer’s informed consent. No equivalent protection exists in India.
The commercial results, however, were enough to embolden the studio. India’s largest cinema chain PVR Inox reported that ticket sales for the AI-altered Tamil version outpaced the 2025 average by 12 percentage points. Eros is now reviewing its 3,000-title catalog for additional candidates.
The economics driving all of this are hard to dismiss. One AI studio head noted that production costs have been cut to roughly one-fifth of traditional filmmaking in genres like mythology and fantasy, with timelines compressed to a quarter of what they once were.
The disruption extends well beyond directors and actors. India’s dubbing industry, an ecosystem of roughly 20,000 freelance voice artists servicing a film market that spans more than ten major languages, is confronting an existential threat. A tool called VisualDub, developed by Bengaluru-based startup NeuralGarage, now adjusts an actor’s lip movements and facial expressions in real time so that dialogue appears to be spoken naturally in a second language. Veteran voice artist Ghazal Khanna estimates that around 70 to 80 percent of brand voices for major Indian TV and video commercials have already been replaced by AI.
The contrast with Hollywood remains sharp, though not simply because American studios are more principled. The WGA and SAG-AFTRA strikes of a few years ago were fought in significant part to establish contractual limits around AI use, and those protections continue to be expanded. In India, filmmakers working without that institutional backstop are left to navigate the tension individually.
Speaking to The Hollywood Reporter, director Shakun Batra argued that resistance from established Hollywood names comes from “the top of a very established industry” and that a young filmmaker in a rural state without access to studios or budgets “is not going to wait ten years for permission to make their first film.”
Director Anurag Kashyap, however, pushed back on that framing, saying that AI comes with both a human and an environmental cost that he simply cannot ignore, and that all a filmmaker truly needs is a camera. The debate within Bollywood mirrors the one consuming Hollywood, except that in India, the experiment is already well underway, and the results, for better or worse, may be writing the script for everywhere else.
Have something to add? Let us know in the comments.

