Clipto raises $15 million at a $250 million valuation: betting on local media indexing in the era of AI agents

While the generative AI wave has accelerated the production of text, images and video, a parallel dilemma has emerged: terabytes of unused material accumulate on personal and corporate storage devices. In this space, Clipto closed a financing round of $15 million entirely equity-based, raising its post-money valuation to $250 million, with investors including HSG (formerly Sequoia China), GL Ventures, Invision X Capital, Palm Drive Capital and 522 Ventures.

The bet here is not on generating new content but on retrieving what the user already owns via natural language.The platform, based in San Francisco with teams in Singapore and Hong Kong, indexes video, audio, images, meeting recordings and documents stored locally on the user’s computer. Rather than manually searching folders, the user can describe what they seek in text or delegate tools and models such as ChatGPT and Claude to locate and extract the required files.

The most notable technical advantage of Clipto’s architecture is that all processing runs locally on the user’s device without intermediate cloud services, requiring explicit, direct user consent whenever an AI application accesses the indexed data. About two weeks ago, the company added support for the Model Context Protocol (MCP), a standard that allows AI applications to connect to external data sources only within the scope defined by the device owner.

Founder Henry Kang draws on roughly two decades of experience, beginning with his PhD research at Carnegie Mellon University in 2006 on robots that map their surroundings, recognize objects and remember locations, and later his previous venture ZenVideo, which was acquired by Tencent in 2020. Kang founded Clipto in 2023 after noting that the current problem is not a lack of content but its unused accumulation. Although the tool was initially built for video creators, they now constitute only about a quarter to a third of users, with the remainder spread across lawyers, doctors, researchers, marketing and HR managers, academics and students. The company, employing just over 20 staff, reports that its user base has reached more than 30 million since launch, with hundreds of thousands of paid-contract subscribers, and generated recurring annual revenue of $15 million at the start of 2026 while remaining profitable on a net-income basis.

The real challenge lies in independent software’s ability to stay ahead of technology giants.Adobe offers AI-enhanced search features inside Premiere, while Apple and Google enable natural-language search within their respective photo apps, but these solutions remain confined to their own ecosystems. Clipto bets on comprehensive search across media and document formats linked to intelligent agents to secure an independent position, directing the new funding toward developing models and infrastructure that run efficiently on consumer hardware and integrate with more agents.

At media production studios, law firms and marketing teams across the Gulf, Egypt and the Levant, this development forces a direct shift in handling massive digital archives. Software that can locally index hundreds of hours of recordings and work sessions and link them to intelligent agents via protocols such as MCP enables organizations to build assistants that search sensitive material without incurring terabyte-scale data transfer fees to the cloud or compromising privacy controls. Your practical step as a technical or media manager is to test the integration of local search tools with daily-task agents before committing to cloud storage repositories that require cumulative subscriptions merely to retrieve your own media.