The frustration of attempting to join AI clips to form a cohesive video is something everyone has experienced before. The main character’s coat color alters in the third scene. The kitchen acquires a window. And your leading man ends up becoming an unfamiliar face with somewhat familiar hair. This is what the Eloelo Group claims to have solved and named it Dolphin AI. Dolphin AI was introduced at FICCI FRAMES 2026 held on September 29.
Also read: GPT-6.1 Sol: Everything that’s better than the week old GPT-6 Sol
It couldn’t be simpler according to CEO Saurabh Pandey. Today, AI video is all about lack of context and inability to go past 8 to 10 seconds. You only need to have seen a generated video deteriorate at the 10 second mark to agree with this.
Dolphin’s solution is multi-shot planning and continuity technology. Rather than producing each individual shot in isolation, it produces the video as a whole, taking care of the characters, voices, wardrobes, props and even locations through the individual shots. There is an agentic layer that interprets the creative intention behind the video and divides a longer video into individual shots and tasks.
This is what Dolphin claims. However, there aren’t any demos, benchmarks or max video length available from the launch release.
Also read: OpenAI Dots explained: An AI coworker that never logs off
Dolphin has four modes. The Chat mode allows you to explain the video in simple words. The Templates mode introduces more structure into the process. The Playground mode gives you shot level controls. And finally, Film Studio is the most adventurous: enter your script, and it will analyze it for the theme, tone, chronology, and character development. It will identify characters, settings, and props, and create reusable visual, voice, and wardrobe assets. Lock. The word is crucial.
Looks like a consumer app firm venturing into AI tooling. Well, it has been nibbling on its own cooking all along. Story TV, Eloelo’s vertical storytelling app, has already created over 5,000 original dramas, and the teams at Eloelo screen over 100,000 creatives every month. Dolphin was originally designed as a solution for the problem and is currently being made available to creators, filmmakers, brands, and studios. Those who have shipped dramas in such huge quantities know the pain spots in the production process well.
The most opinionated statement from the announcement comes from CTO Sagar Gaonkar who says that the model is becoming a component and the creative system is the product. Dolphin is not banking on one video model out there winning. Instead, the company is planning on using the appropriate model at each step while maintaining consistency when all the steps are combined.
Smart play in an environment where the leaderboard keeps getting rearranged on a monthly basis. Here’s the problem – quality depends entirely on the underlying models and none of them are named in the announcement. Dolphin has also made its technology compatible with MCP. As a result, Claude, ChatGPT, Gemini and Cursor can take advantage of Dolphin’s video generation capabilities inside their systems.
Pandey aspires to transform India from being one of the largest content consumers into one of the largest content exporters, a transition which is associated with the Orange Economy. The announcement estimates the creative economy in India to be around 2.5% of GDP, with around 80 million creators working in the creator economy. Dolphin aims at advertising, product video production, microdramas, podcasts, cinematic movies, user-generated content, AI avatars, and animation. Dolphin is currently open for early access. Creators can sign up for the waitlist through trydolphin.ai or through an invite code.
Also read: Apple’s Vision Pro, Samsung’s Android XR, Meta VR Glasses: The battle for the future of XR