Coram's $35 Million Round Backs AI Investigation Across Existing Cameras

Coram AI announced a $35 million Series B co-led by Ansa Capital and Battery Ventures on 11 June 2026, bringing total funding to $66 million. Its central product claim is not that every customer should replace every camera. Coram says its edge appliances can connect existing cameras to a cloud platform for search, access control, visitor management, alerts, and investigation.

That compatibility is the wedge. Physical-security buyers often have mixed hardware, long replacement cycles, and footage spread across sites. A software layer that makes the installed estate searchable can reach useful deployment before a full hardware refresh.

The Series B announcement highlighted Deep Investigation agents working across video, access-control, and visitor data. The important shift is from finding one clip to reconstructing an incident: who entered, which camera saw them, what happened next, and which other records add context.

Coram says Nvidia-powered edge processing helps it work with existing camera networks. That architecture can reduce the need to send every raw stream directly to the cloud, but buyers still need clear answers about retention, biometric handling, access logs, model errors, and what remains available when connectivity fails.

Before the financing, Coram published examples from a zoo and a school district. The Point Defiance story focused on finding people and incidents across a public venue. The Brown County Schools story said staff reduced video-search time from hours to minutes. Both are company-authored case studies, so they demonstrate the workflows Coram wants to own, not independent proof of typical performance.

The access-control launch in October 2025 adds a second system of record beside video. That matters because the product becomes more valuable when a search can connect visual evidence with doors, credentials, visitors, and response actions.

Camera vendors retain device reliability, channel relationships, and installed base. Access-control and security-platform vendors may have deeper identity and response workflows. Coram pressures both by promising a usable AI layer without replacing the underlying hardware.

The response should be proven in the buyer's environment: investigation time, search precision, compatibility, total storage and appliance cost, permission controls, and false-alert handling. Market-share claims cannot be inferred from a few selected case studies.

Coram's public trail is coherent: natural-language video search, cloud access control, customer investigation stories, and then agents that connect the systems. The Series B gives the company more room to repeat that deployment model.

The decisive proof will be broader, independently assessed performance across mixed camera estates and sensitive environments. In physical security, a fast answer is useful only when teams can trust how it was produced and who is allowed to see it.