Samsung Invests in Memories.ai for Advanced AI Video Analysis

While artificial intelligence has made impressive strides in video summarization, most models still falter when it comes to analyzing multiple videos across extended timeframes. The current generation of tools can summarize individual clips or short videos with relative ease, but scaling this ability to sift through hours, or even thousands of hours, of footage across diverse sources presents a formidable challenge. This limitation is particularly significant for industries that rely on extensive video archives, such as security firms monitoring feeds from countless cameras or marketing teams managing massive content libraries from varied campaigns.

The Enterprise Demand for Contextual Video Intelligence

For security and marketing organizations alike, the value of video lies not just in isolated visuals but in the patterns, behaviours, and narratives that unfold over time. These sectors require more than a simplistic synopsis; they need in-depth contextualization, the ability to draw connections, and actionable insights from enormous datasets. The traditional manual methods of navigating video archives are no longer tenable at this scale. What’s missing is a platform that brings sophisticated semantic intelligence and real-time querying to large video repositories, a gap that Memories.ai is positioning itself to fill.

Memories.ai: A Context-First Platform for Video Analysis

Memories.ai has developed an AI-powered platform capable of processing up to 10 million hours of video, offering companies a new lens through which to derive value from their visual assets. Rather than just condensing footage into highlights, the platform introduces a deep contextual layer that makes video archives both searchable and intelligible. By incorporating features like automated indexing, intelligent tagging, and segment-based aggregation, it allows enterprises to locate specific moments, track thematic trends, and generate insight-rich reports at scale. This turns passive video footage into an active, data-rich asset.

The Founding Team Behind the Vision

At the core of this innovation is a team with deep roots in advanced AI research and real-world deployment. Co-founder Dr. Shawn Shen brings experience from Meta’s Reality Labs, where he contributed to frontier research while completing his PhD. His co-founder, Enmin (Ben) Zhou also worked at Meta, specializing in machine learning engineering. Their combined expertise underpins the technical credibility and forward-thinking nature of Memories.ai, giving the platform both academic depth and commercial relevance.

“All top AI companies, such as Google, OpenAI, and Meta, are focused on producing end-to-end models. Those capabilities are good, but these models often have limitations around understanding video context beyond one or two hours,” Shen said.

“But when humans use visual memory, we sift through a large context of data. We were inspired by this and wanted to build a solution to understand video across many hours better,” he said.

Strong Investor Confidence Fuels Growth

This vision has not gone unnoticed by the investment community. Memories.ai recently secured $8 million in seed funding, in a round that was initially capped at $4 million but became oversubscribed due to intense interest. Leading the round was Susa Ventures, with key contributions from Samsung Next, Fusion Fund, Crane Ventures, Seedcamp, and Creator Ventures. The oversubscription signals a growing recognition among investors that video intelligence, particularly at scale, is a critical and under-addressed problem.

“Shen is a highly technical founder, and he is obsessed with pushing boundaries of video understanding and intelligence,” said Misha Gordon-Rowe, a partner at Susa Ventures. “Memories.ai can unlock a lot of first-party visual intelligence data with its solution. We felt that there was a gap in the market for long context visual intelligence, which attracted us to invest in the company,” he added.

“One thing we liked about Memories.ai is that it could do a lot of on-device computing. That means you don’t necessarily need to store video data in the cloud. This can unlock better security applications for people who are apprehensive of putting security cameras in their house because of privacy concerns,” said Sam Campbell, a partner at Samsung Next.

A Proprietary Tech Stack with Intelligent Layers

Unlike many startups that rely heavily on open-source AI models, Memories.ai has built its own proprietary tech stack tailored specifically for large-scale video comprehension. The system begins by eliminating irrelevant noise from footage and applies a compression algorithm to retain only high-value data. This refined input is then passed into an indexing engine that enables semantic search through natural-language queries. Beyond this, a segmentation and tagging process makes it possible to isolate key themes and moments. An aggregation layer then synthesizes this information, distilling the index into meaningful reports that offer decision-makers clear, actionable insights.

Focused Applications in Marketing and Security

Currently, the company is targeting two major verticals: marketing and security. In the marketing world, teams can use the platform to analyze social media videos for trend identification, audience sentiment, and content strategy alignment. The system can also aid in content creation by suggesting video formats or narratives that align with current trends. For security firms, the application is even more profound. The AI can detect and analyze behavioral patterns in surveillance footage, offering early insights into potentially harmful or suspicious activity. This moves the role of security AI from passive monitoring to predictive alerting.

Reimagining the Future of Video Intelligence

Memories.ai is not just solving a technical challenge; it is reshaping how businesses interact with video content. By transforming raw footage into an organized, searchable, and actionable knowledge base, the platform enables organizations to make better, faster decisions. Whether it’s identifying consumer behavior trends or preventing a security breach, the implications of contextual video AI are vast. As industries continue to generate visual data at exponential rates, tools like Memories.ai will be essential in making that data not just manageable, but truly meaningful.