The traps themselves work by luring female mosquitoes, the only ones that bite, with cues that mimic humans: heat similar to skin temperature, a scent imitating human skin odour, and light. The flagship Grand model, intended for larger areas and public spaces, also releases CO2 to imitate human breathing. Catching females before they lay eggs, typically 100 to 200 at a time on a roughly three-week cycle, is meant to suppress population growth.

The AI side is a computer-vision system that analyses biological characteristics to distinguish species. Each identification runs an image through a neural network, checks the result against external datasets, and has entomologists validate the findings before a confidence score is assigned. Environmental data such as weather, wind and humidity is also fed in, and the company says it plans to use these inputs to forecast changes in mosquito populations and potential health risks.

The company stresses that Febris does not diagnose disease: it can identify which mosquito species are present, which can indicate an increased potential risk of diseases that species can transmit, but it cannot say with certainty that a particular mosquito carries dengue. The context for the push is a growing problem: the World Health Organization reports that dengue cases more than doubled in a year, from 6.5 million globally in 2023 to a record 14.6 million in 2024, as rising temperatures and global travel reshape where disease-carrying mosquitoes can thrive. Mosqitter has grown accordingly, with more than 5,000 devices across five continents, up from just over 250 in 2022, and local manufacturing and sales operations in Brazil, a high-burden dengue country.

Unlike the hardware, Febris targets institutional customers such as governments, public-health agencies, airports, municipalities and universities. The company is already working with the UK Health Security Agency to deploy the system at Heathrow and Gatwick airports, and is seeking B2B distribution partners globally.