Organisers of Britain's largest-ever woodland bat survey say the exercise has picked up calls from some of the UK's rarest and most threatened bat species in the south-west of England, in findings they describe as a "nice story of AI for good".
The survey, which covered woodlands across the country, used artificial intelligence to assist with analysis of the acoustic recordings gathered during nocturnal monitoring. Project leads say the use of machine learning tools allowed them to process large volumes of audio data more quickly than would have been possible by manual review alone.
Conservationists welcomed the detections as an important indicator of the continuing presence of vulnerable bat populations in parts of the country, and said the results will help inform future habitat protection and management. Bats are protected species in the UK, and monitoring their distribution is considered a key part of efforts to maintain biodiversity and assess the impacts of land use change.
Organisers said the survey demonstrates how technological tools can support environmental science and conservation. They added that the findings will be passed on to agencies and local conservation groups to guide follow-up surveys and targeted measures aimed at safeguarding roosts and feeding habitats in the affected areas.
Further work will be needed to build on the initial results, including targeted field studies to confirm species identities and to map important sites more precisely. Project partners said they hope the approach could be repeated and refined in coming years to provide a clearer picture of long-term trends in bat populations across Britain's woodlands.