1. Headline & intro
OpenAI’s new deal with Reliance isn’t just another flashy AI feature. It’s a bid to control the most valuable real estate in entertainment: the moment when you pick what to watch. By wiring conversational AI deep into India’s largest streaming ecosystem, OpenAI is testing what an AI-first TV experience actually looks like at national scale. If it works in India, expect the rest of the world’s living rooms – including Europe’s – to follow. In this piece we’ll unpack what was announced, why India is the proving ground, what this means for the streaming wars, and how it could reshape media discovery in Europe.
2. The news in brief
According to TechCrunch, OpenAI has partnered with Reliance to add AI-powered conversational search to JioHotstar, the streaming service within Reliance’s JioStar joint venture. Using OpenAI’s API, viewers will be able to search for films, series and live sports through text or voice prompts in multiple languages and receive recommendations based on their viewing history and preferences.
The partnership was announced at the India AI Impact Summit in New Delhi, where OpenAI’s leadership appeared alongside executives from Anthropic and Google. TechCrunch reports that OpenAI plans to open offices in Mumbai and Bengaluru this year, expanding on its existing New Delhi presence, as part of a broader “OpenAI for India” initiative that already includes collaborations with Tata Group, Pine Labs, Eternal and MakeMyTrip.
Crucially, the integration is two-way: JioHotstar recommendations will also surface inside ChatGPT, with contextual suggestions and deep links into the JioHotstar catalogue.
3. Why this matters
This partnership is less about a fancy new search bar and more about who owns the interface between viewers and content.
For OpenAI, this is a distribution coup. India reportedly already has over 100 million weekly ChatGPT users. Embedding OpenAI not just in a chat app but in the default entertainment experience of hundreds of millions of Jio customers turns ChatGPT from a destination into infrastructure. Every time someone presses the remote and asks in Hindi or Tamil for “a family film like the one we watched last weekend,” OpenAI’s models sit in the middle of that decision.
That position comes with three advantages: recurring enterprise revenue via API usage, incredibly rich behavioural data (even if not used for training, it shapes product development), and proof that OpenAI can power large, latency‑sensitive consumer services.
For Reliance and JioHotstar, this is about differentiation and retention. Indian streaming is brutally competitive, especially around live sports. Cricket rights are staggeringly expensive; if AI-powered discovery keeps users inside JioHotstar for a few extra minutes per session – and reduces churn after marquee tournaments – the economics change. A conversational interface also lowers the barrier for users in smaller cities and in households where typing on a TV remote is painful or where English skills are weaker.
The losers? Any streaming service that continues to treat discovery as a basic search bar and a carousel of thumbnails. If users get used to telling the TV why they want to watch something – mood, company, language, time available – they’ll quickly feel that old-school interfaces are dumb and frustrating.
The risk is that recommendation power becomes even less transparent. When a general-purpose AI sits between you and content, it becomes harder to know whether you’re seeing the “best” option, the one that maximises watch time, or eventually the one that maximises ad revenue or partner priorities.
4. The bigger picture
This deal lands in the middle of a larger shift: entertainment platforms are racing to turn AI assistants into the primary discovery layer.
TechCrunch notes that Netflix has been testing a natural-language search experience using OpenAI’s technology since 2025, letting users describe what they’re in the mood for and receiving tailored suggestions. Google, meanwhile, has added Gemini-powered discovery features to Google TV, letting people ask open-ended questions like “good films for kids who love dinosaurs” across multiple apps.
JioHotstar goes one step further by integrating discovery into both the streaming app and ChatGPT itself. Ask ChatGPT on your phone what to watch tonight, and it can push you directly into a specific JioHotstar title or live match. That turns ChatGPT into a kind of cross‑platform remote control – exactly the role Google, Amazon and Apple have been pursuing with voice assistants for a decade, but now with far more context and reasoning.
Historically, whoever controlled the electronic programme guide (EPG) or home screen controlled viewer attention. In the cable era it was the set‑top box; in the early streaming era it became individual apps. Now, the discovery layer is abstracting away from any one app. A handful of AI assistants could soon decide how traffic is distributed across video platforms, much like search engines decide which websites get visited.
This is also a strategic signal from OpenAI. After years of focusing on a direct‑to‑consumer product (ChatGPT) and horizontal APIs, it is increasingly striking vertical partnerships: with office productivity, customer support – and now media and live sports. Each vertical foothold makes it harder for competitors like Anthropic, Google, or open-source players to dislodge it.
5. The European angle
For European viewers, this might look like a distant India story. It isn’t.
Europe faces a similar challenge to India: fragmented languages, strong local broadcasters, and audiences overwhelmed by content across linear TV, national streamers and global giants. A multilingual conversational interface that actually works for a German‑Turkish household in Berlin or a Catalan‑speaking family in Barcelona is just as valuable as one that supports Hindi, Tamil and Bengali.
Yet the regulatory environment is very different. Under the Digital Services Act (DSA), large platforms operating in the EU must explain how their recommendation systems work and offer meaningful controls and alternatives. GDPR puts strict limits on profiling and the reuse of behavioural data. The upcoming EU AI Act will require extra safeguards and documentation for general-purpose AI used in consumer-facing services.
If a JioHotstar‑style integration came to a European broadcaster or telecom operator – think Sky, Deutsche Telekom’s MagentaTV, Orange or Canal+ – they would need to provide far more transparency about why a particular show or match was recommended, and clear options to opt out of personalised profiling.
There’s also the industrial angle. European AI firms like Mistral, Aleph Alpha or Stability AI are trying to position themselves as sovereign alternatives to US and Chinese models. Media and telecom groups in the EU may come under political and regulatory pressure to work with European AI vendors where feasible, especially for core discovery functions that shape information diets.
India is becoming the scale testbed for AI‑first entertainment interfaces. The question for Europe is whether it wants to import that model largely from US companies, or co‑develop its own equivalents.
6. Looking ahead
Several things are worth watching over the next 12–24 months.
First, user behaviour. Do people actually talk to their TV, or do they default back to scrolling? Voice assistants on smart TVs existed for years without becoming dominant. The key difference now is that LLMs can handle follow‑up questions, constraints (“nothing violent, under 40 minutes, kids can watch too”), and reference previous sessions. If engagement data shows that AI queries meaningfully reduce search time and increase completed views, every major streamer will feel compelled to copy.
Second, the depth of integration. Today it’s discovery. Tomorrow it could be real‑time sports analytics, personalised highlight reels, or commerce layered on top of live events. Imagine asking during a cricket match, “show me every boundary this player has hit today” and getting an instant reel – or being nudged toward merchandise or betting partners based on your viewing.
Third, platform power dynamics. As ChatGPT starts sending users directly into JioHotstar, we move closer to an AI layer that sits above individual apps, not unlike how Google Search sits above websites. At some point, streamers will ask whether they’re comfortable being so dependent on a single AI gateway, especially one controlled by a foreign company.
Finally, regulation and trust. India’s own data protection regime is tightening, and any high‑profile error – say, the AI confidently recommending content that isn’t available, or surfacing controversial material during major sports events – will attract scrutiny. Europe’s regulators will be watching closely, using India as a case study when they assess AI‑powered interfaces under the DSA and AI Act.
7. The bottom line
OpenAI’s deal with Reliance turns JioHotstar into a live laboratory for AI‑first entertainment, with India as the scale experiment for how we’ll all search for TV in a few years. It strengthens OpenAI’s grip on the discovery layer while raising hard questions about transparency, data use and platform dependence. The real question for readers is simple: when your TV asks “What are you in the mood for?”, are you comfortable letting one AI system decide what comes next?
