Meet VOGIC AI, which turns raw CCTV footage into searchable, actionable intelligence

Big Brother is watching you.

This line from English novelist George Orwell’s book, 1984, succinctly expresses what the hundreds of thousands of CCTVs and surveillance drones mean to the human race—keeping an eye on our every movement.   

Government agencies and corporations get terabytes of video footage each year from thousands of CCTVs and drones. Yet, most of this data sits unused—buried in archives, only reviewed when an incident demands it. 

But what if this vast, underutilised resource could be transformed into a real-time, searchable knowledge base?

Enters VOGIC AI—a video intelligence platform founded by Arijit Biswas(CEO), Rahul Singh Thakur (CTO), and Nikhil Mehta (Co-founder) in 2024. The Gurugram-headquartered startup transforms raw video into a structured, searchable knowledge base, allowing users to retrieve contextual insights instantly by leveraging vision language models (VLM).

"VOGIC AI makes video search as easy as having a conversation. Need to find specific moments in hours of footage? Just ask: ‘When did the cashier leave the station?’ Want real-time alerts? Set one for ‘Someone climbing over the wall.’ Looking for quick data? Simply query, ‘How many vehicles entered yesterday?’," Arijit Biswas, CEO of VOGIC AI, tells YourStory.

Unlike traditional surveillance tools that rely on rigid parameters—detecting only pre-defined objects like shirt colours or license plates—the startup enables users to search, analyse, and set alerts using natural language queries.

“Instead of relying on predefined detections, our platform extracts deeper contextual insights from video, allowing users to retrieve specific moments and patterns dynamically, without prior labelling or training. This enables a flexible, search-driven approach to video intelligence,” he highlights.

According to the co-founder, VOGIC AI provides context-aware intelligence—empowering security teams, city planners, and business leaders to make faster, data-driven decisions, where conventional systems generate overwhelming false alarms and require hours of manual footage review.

Biswas explains, "The future of video intelligence isn’t just about watching—it’s about understanding. And with VOGIC AI, organisations can finally harness the full potential of their video archives, turning them into powerful assets for security, operations, and beyond."

Co-founders Biswas, Thakur and Mehta have over 30+ years of combined experience across business-to-business (B2B) SaaS, Internet of Things (IoT), and artificial intelligence (AI). The trio had previously built HopCar, a peer-to-peer pre-owned car platform (acquired by Spinny in 2016), and EnrichAI, a profitable IoT-as-a-Service business serving Ooredoo Telecom, Software AG, and RIL Unlimit.

It was at EnrichAI that they realised how CCTVs—one of the richest and truest data sources—remain largely unutilised because there is no easy way to search for contextual information, paving the way for VOGIC AI.

Business model and revenue

The video intelligence platform operates on a scalable AI SaaS model, offering two primary revenue streams designed for multi-sensor intelligence, including feeds from fixed, mobile, and aerial vision systems:

It offers an annual software licensing subscription plan, where organisations pay a per-device fee to deploy AI-powered video intelligence on their existing infrastructure of CCTVs or drones. 

The second revenue source is a usage-based model, where businesses can purchase API bundles based on the volume of hours of footage indexed, analysed, and queried.

VOGIC AI serves a diverse clientele, including video alarm monitoring centres, construction firms, drone survey providers, smart cities, and educational institutions. It also serves Prakhar Aviation in India and OmniVision Security in South Africa.

Way ahead

According to a Markets and Markets report, the global video surveillance market is projected to reach nearly $89 billion by 2030, with India’s share estimated at about $8 billion. 

“With growing demand for AI-driven video intelligence, multiple players can co-exist and contribute toward making physical spaces safer, smarter, and more secure,” says Biswas.

VOGIC AI competes with indigenous video management system (VMS) providers like Videonetics, which focuses on efficient storage, retrieval, and streaming of CCTV footage, as well as video analytics companies like IntelliVision, Awiros, and Axis Communications, which specialise in point-based analytics, detecting objects, humans, vehicles, and predefined events in surveillance feeds.

The company is also helping the Uttar Pradesh government manage the crowd in Prayagraj at the ongoing Mahakumbh Mela. Biswas says, “Our Vision Language Model, will ingest feeds from existing CCTV networks in the city and provide a real-time contextual understanding of crowd movement, triggering public announcements and digital signage alerts to aid in decongestion and ensure safer, more accessible, and secure crowd flow across the holy city.”

The Gurgram startup was a semi-finalist at the Toyota Mobility Foundation’s Sustainable Cities Challenge for Varanasi and received a $50,000 grant. The challenge attracted 80+ high-quality global innovators, including Fractal Analytics, Tiami Networks, Graymatics, etc., focused on enhancing crowd management. 

VOGIC AI is in discussion to raise an institutional funding round soon to expand its operations in the Middle East and North America. It also aims to reach $5 million in annual revenue by 2027.

Speaking about future plans, Biswas says the company wants to partner with law enforcement, traffic police, and the Indian Army to develop AI agents tailored to India’s needs.

He adds, “We also aim to make the platform more actionable and cohesive. We will be integrating GIS, satellite imagery, public announcement systems, digital signage access control systems, etc., to build a one-stop platform for cities to manage crowds, improve traffic flow, monitor infrastructure, and enhance public safety.”