A trial in Ecuador reveals the huge potential of the new tech to record biodiversity, but opinion remains divided
Stuart Blackman
Artificial intelligence (AI) may not be ready to take over the world just yet, but according to new research, it is poised to revolutionise how we monitor wildlife. A proof-of-concept trial shows that AI is capable of identifying the species present in tropical forests based on sound recordings alone.
Biologists working in Ecuador trained an AI system to recognise the calls of 70 bird species and then tested its ability to pick them out from recordings made in forests at various stages of regeneration from farmland. As reported in the journal Nature Communications, the system could identify not only the calls, but also the stage of regeneration based on the communities of species present.
The technique presents an opportunity to process huge amounts of data, which would not be possible using an expert human workforce. And it can be done without disturbing the wildlife. “It gives us an efficient way of measuring the success of conservation projects,” says Jörg Müller, from the University of Würzburg, Germany, who led the research.
Will Cresswell, professor of biology at the University of St Andrews, who was not involved in the study, says the work “shows that we are on the cusp of making a machine that will tell us accurately and efficiently what
