In a GOV.UK case study published on 13 August 2026, UK Defence Innovation presented Vizgard as an example of how small-firm defence funding can convert a niche software product into a usable operational tool. The London company’s FortifAI platform was developed with UKDI funding and Defence Science and Technology Laboratory support to assist drone operations, including policing use beyond visual line of sight and the detection of hostile drones. (gov.uk) That matters because the underlying issue is not simply flying more drones. It is managing a busier airspace in which the same low-cost platforms can support safety and security tasks but can also be used by adversaries. UKDI’s own framing places Vizgard within that wider security problem rather than treating it as a standalone software story. (gov.uk)
FortifAI is designed to sit on existing camera infrastructure across land, air and sea rather than require a wholly new fleet. The GOV.UK case study says the package combines onboard AI for police drones, edge computing placed next to the camera so sensitive police data can remain local without internet access, and ground-based steerable cameras that scan the sky and investigate alerts from radar and radio-frequency sensors. (gov.uk) Set out plainly, the system is intended to take on the repetitive visual work that otherwise absorbs operator time: spotting, tracking and classifying objects across several feeds at once. Dstl’s current competitions show why that matters. The department is actively seeking more autonomous sensor management because ISR remains too human-intensive in fast-moving and deceptive environments. (gov.uk)
The procurement story is the most useful part of the case study. According to UK Defence Innovation, Vizgard had to rebuild its software so that a system once limited to one camera stream for one drone could process dozens of streams and now more than 100. That redesign was financed through UKDI support at the point where the technology had shown promise but was not yet robust enough for operational use. (gov.uk) This matches the published logic of the Open Call for Innovation, which says funded work should have realistic impact within three years, whether by reaching an end user, informing procurement or shaping a wider capability programme. On that reading, Vizgard is not just a company success story. It is a live example of government trying to shorten the distance between prototype and procurement. (gov.uk)
UKDI’s case study traces that path back to Vizgard’s first commercial contract in 2021, when the company joined sea trials on the Royal Navy’s uncrewed test submarine. The department says the trial fused radio tracks with visual AI to identify vessels masking their identity and to automate the tracking of fast-moving craft, giving the firm an early operational proof point rather than a laboratory-only demonstration. (gov.uk) For policy readers, the significance is practical. Early government contracts do more than cover research costs; they generate test data, expose software to real conditions and give larger suppliers and public buyers evidence that a small firm can deliver. UKDI’s current guidance now presents that route as a central offer to SMEs seeking entry into defence and security markets. (gov.uk)
The human-factors case is also clear. UKDI says the system provides automated tracking of friendly drones while continuously scanning for other aerial objects, a function intended to reduce the chance that an operator misses a relevant movement on one of several live feeds. The design assumption is not that personnel are dispensable, but that attention is scarce when multiple sensors are running at once. (gov.uk) That sits neatly alongside Dstl’s wider work on autonomous sensing. A 2026 UKDI-Dstl competition document on autonomous sensor management links this family of technology to the planned Digital Targeting Web and to a shift away from ISR processes that rely too heavily on manual interpretation. Vizgard’s platform therefore fits a clear departmental direction of travel: more edge AI, more sensor fusion and faster exploitation of visual data in contested environments. (gov.uk)
The company’s growth figures suggest the model has commercial traction as well as policy relevance. UKDI states that Vizgard has grown from about five employees to 20 since first working with the programme, while the company announced a £1.5 million seed round in January 2025 to expand engineering capacity and international distribution. The GOV.UK case study also says Vizgard is exporting through a Defense Innovation Unit contract for field testing with the US Marines. (gov.uk) International recognition has followed. NATO DIANA’s published 2026 cohort lists Vizgard under Autonomy and Unmanned Systems, describing its offer as distributed visual autonomy for scalable management of an AI-enabled autonomous fleet. NATO DIANA has separately said the 2026 cohort contained 150 companies selected from more than 3,600 proposals, placing Vizgard within a competitive allied acceleration route rather than a purely national one. (diana.nato.int)
Closer to home, the GOV.UK case study says UKDI support has also led to work with Dstl on edge-based AI for quadcopters identifying unexploded mines during automated ISR missions. Separate Ministry of Defence reporting from April 2026 shows the Army and Dstl are already trialling AI-enabled drones to detect landmines and explosive ordnance, indicating that this is part of a broader push to field autonomous sensing rather than an isolated pilot. (gov.uk) The broader lesson is straightforward. On the evidence published by UKDI, Dstl and NATO, Vizgard is best understood as a procurement case study: public money absorbed early technical risk, an SME used that runway to reach production scale, and the resulting software is now being tested across policing, defence and allied channels. For ministers and officials arguing that dual-use AI funding can produce deployable capability, this is the kind of pathway they are trying to institutionalise. (gov.uk)