In one episode of one of India's most watched new mythological series, the warrior princess Gandhari covers her eyes, deciding to be blind like her blind husband whom she has been married to. The moment is arresting, the fabric just falls the right way, the face beneath it has a kind of tension that suggests it's been done for months and millions of dollars on a real set, shimmering and falling, but it's not, it's just one moment. There was not a single camera called on that production. Nobody shouted action. The sequence was typed by a team of engineers in Bengaluru into an AI interface, checked, tweaked and sent on.

That's where the worldwide discussion about artificial intelligence in entertainment goes from theoretical to money chasing.

In October last year, the joint venture between Mukesh Ambani's Reliance Industries and Walt Disney debuted a 100 episode series called JioStar that was created by generative AI and is a complete retelling of Mahabharat. According to the company, the series had 6.5 million views the day of the launch and was more than double the platform's average. To put that in perspective, 200 million people watched the same epic when it was made into a television series back in 1988-1990. That's not exactly a compliment, but the money sense it exposes is precisely the idea.

India has the largest number of film production in the world. It isn't a cottage industry, it's a $32 billion machine. However, changing consumer habits, such as streaming, are putting pressure on production spending and attendance dropped to 832 million in 2025. As studios battle a drop in attendance, and as streaming platforms desire more and more content, they have come to a logical conclusion: the only way to feed the beast is to make content cheaper and faster. This is not about some philosophical decision about AI. It's an economic imperative wearing technological camouflage.

The statistics coming out of early adopters are pretty astronomical. According to Collective's AI studio head Rahul Regulapati, Galleri5, "AI is bringing down the production costs to just one-fifth of traditional filmmaking in mythology, fantasy genres, etc. He added that production time is "down to a quarter. If we are to take one production team's account to industry reporters as a practical example of what this means at the project level, then an animation from scratch would have cost about a few million dollars with conventional techniques and took between 6 months and 1 year to produce. AI can accomplish the same with a few hundred dollars and in a matter of weeks.

Consultancy firm EY has placed a few more grand numbers on this trend. Over the medium term, AI could increase Indian media and entertainment companies' revenues by 10% and lower their costs by 15%. More ambitiously still, EY research indicates that the technology could ultimately have a US$15-20 billion impact on the sector, roughly the size of the domestic media and entertainment industry today. By the start of 2026, the Indian generative AI video market alone is expected to be valued at $1.4 billion, boasting a 72% adoption rate in the leading production houses for B-roll and pre-visualization.

So the difference between the two and Hollywood is not only cultural, it's structural. Yet this stands in stark contrast to Hollywood, where use of AI is still more limited because of job loss fears and union restrictions. Recent years' writers' and actors' strikes resulted in a set of contractual protections, which put a cap on the extent of experimentation American studios could pursue. There is no such framework in India. But there are no guilds powerful enough to slow the machine's downhill momentum and the economic imperative is everything but caution.

With the launch of JioHotstar, JioStar is now set to launch a bouquet of series, which would be written, animated, voice acted and edited by AI. The company is looking to recruit 80 AI specialists and engineers for the push and the number is telling AI adoption on this level is not a down-sizing effort, it's a workforce replacement and retooling effort at the same time. Old work opportunities are disappearing, new work opportunities are vastly different.

aiON is a Rs 100 crore AI-powered film studio, which Abundantia Entertainment and InVideo officially announced, planned to make five full-length commercial feature films. On the other hand, Collective Artists Network (CAN), a leading talent house for Bollywood A-listers, is creating multiple episodic series, digital-first short-form universes and feature length stories in all of them through its in-house generative AI company Prismix Studios.

It is mythological genre that has become the default laboratory for such experimentation – and there is a reason. Studios in India are venturing into bold AI experiments, even in mythological Hindu stories, which is a significant market for this devout nation. Collective is in the process of creating 8 such AI-generated titles, which include Hanuman, Krishna, Durga, Kali and more. It's a genre that naturally calls for precisely what AI excels at creating: fantastical settings, far from realistic character designs, epic scope, and an audience that expects the visual worlds to be different than the one they're in. In a way, it's a training ground that couldn't be any better.

Hence the importance of the critical reception to the first experiment. Since the show debuted in October, it has garnered at least 26.5 million views, but the AI-created Mahabharat has had a rocky run with viewers. This is that nagging tension that no price number of the production can ever eliminate that people can see when something was made by a machine and they often don't like the results, and still they keep on watching. The efficiency increases are genuine. The art quality question is a true question.

Not all the creative community in India is pleased. Gangs of Wasseypur and Dev director Anurag Kashyap is an acclaimed director.He isn't alone among numerous influential industry leaders who have voiced significant concerns regarding the use of AI. At the International Film Festival of India (IFFI) in late 2025, filmmaker Shekhar Kapur was more nuanced, calling AI the most democratic filmmaking tool, which is dismantling traditional barriers and opening up new voices. The optimists and the skeptics aren't talking past each other either; they're observing different aspects of the same animal.

The one thing they agree on is that the issue of what becomes of people who get their skills "digested" by artificial intelligence is not being resolved. In 2025, India's advertising, media, and entertainment sector slashed over 1,000 jobs in response to a surge in the adoption of AI. The losses were heaviest on the mid-level and support positions people who were doing basic reporting, production support, low-level creative adaptations and account work which was now being done by software that could create first-cut drafts, edits, and analytics quickly and at scale.

This is the part of the story that's hardly ever covered in the hypnagogic frenzy of AI-made launch day view counts of Mahabharat. AI proponents say that's the same thing that happened with the personal computer, and that the workforce evolved, not vanished, which is true overall and not much consolation individually. The animator, who was displaced to a prompt engineer in Bengaluru, isn't that consoled by the historical arc.

But there is a worldwide component to this, and it's not just about the Indian box-office numbers. Dominic Lees, a film and AI researcher at Britain's University of Reading, has been arguing that if the Indian film makers can make their aspirations a reality, the hub of filmmaking with AI will move to India. The implications from that change to content markets in Southeast Asia, the Middle East, and the African continent where Indian entertainment already has significant audiences is huge. India is not only in the process of developing AI filmmaking capacity for its own use. It's set to become the "behind-the-camera content processing hub", as EY calls it, for the world.

The tools powering this desire go by names that are all the more familiar in production circles: Seedance, Minimax, Veo platform by Google's Studio product, Midjourney, Adobe Firefly. Companies like Google, Microsoft and Nvidia have been investing early, forging partnerships with local filmmakers, who could stress test technologies in volume that can't be replicated in any American test environment. The feedback cycle from Indian studios to AI tool development is now a significant force shaping the AI tools themselves.

All this does not answer the main artistic question. The Raanjhanaa controversy, which involves Eros re-releasing the 2013 film with an AI-ed happy ending, and lead actor Dhanush claiming that it had taken the "very soul" of the film, is indicative of the point where efficiency logic meets something that's tougher to quantify. People can be made to watch AI-generated content. But whether or not they will appreciate it like a director who spent three years producing one story is another matter and it's this that will be the deciding factor on what of what is being built now will survive contact with real commercial markets in the decade to come.

But for the time being, nothing is abating. The Indian studios have collectively bet that the economics is compelling enough to overcome the aesthetic discomfort, that viewers will be more willing to accept some AI-generated glitches as the technology advances, and that the cost and time savings of creating content with AI are far too significant to pass up. Hollywood's bet is that the maintenance of old creative labor forms will pay off in market share.

There will be one of these bets that will be correct. The answer, given where investment is going is likely already being written frame by frame, in pixels, in Bengaluru.