Flipkart said on December 20, 2025 it has acquired a majority stake in Bengaluru-based GenAI startup Minivet AI to deepen its AI-led e-commerce capabilities. The deal, for an undisclosed sum, will be used to power video-first catalogs, semantic search and conversational shopping across Flipkart’s platforms.
This article aggregates reporting from 6 news sources. The TL;DR is AI-generated from original reporting. Race to AGI's analysis provides editorial context on implications for AGI development.
Flipkart’s move on Minivet AI is a classic example of how incumbent platforms are racing to hardwire generative models into every part of the commerce stack. Rather than just buying access to a model API, Flipkart is locking in proprietary video generation, semantic search and conversational interfaces as in-house capabilities. That matters because whoever controls the AI layer that shapes how products are discovered and displayed will control consumer attention and, ultimately, margin structure in online retail. ([financialexpress.com](https://www.financialexpress.com/business/news/flipkart-acquires-majority-stake-in-genai-startup-minivetnbsp/4082509/))
Strategically, this deal underlines a broader shift: GenAI is no longer a lab experiment or marketing add-on in retail—it’s becoming core infrastructure. India is a particularly important testbed, with price-sensitive consumers, massive mobile usage, and relatively low video content density in e-commerce. If Flipkart can pull off low-cost, AI-generated video catalogs and truly conversational shopping at scale, it will raise the bar for Amazon, Meesho and others in the region, and set expectations globally for what an AI-native marketplace looks like. ([retail.economictimes.indiatimes.com](https://retail.economictimes.indiatimes.com/news/e-commerce/e-tailing/flipkart-acquires-majority-stake-in-genai-startup-minivet-ai/126075258?utm_source=openai))
For the race to AGI, this is another datapoint showing that value is shifting from generic foundation models to tightly integrated, domain-specific systems wrapped around them. The more commerce, finance, and other verticals embed these specialized agents, the more economic and behavioral data they generate—fuel that will continue to train and benchmark the next wave of frontier models.