Last week, Pl@ntNet was invited by the Royal Botanic Gardens, Kew (UK) to take part in its 2026 Symposium. On this occasion, our colleagues at Kew unveiled their flagship report, State of the World’s Plants and Fungi 2026, to which the Pl@ntNet team actively contributed.
Pl@ntNet’s work was highlighted several times throughout the event. During her presentation, the Director of the Harvard University Herbaria notably referred to our research on the automated identification of herbarium specimens. In addition, our long-standing collaboration with Dr Sue Han Lee, researcher at Swinburne University of Technology Sarawak Campus and specialist in computer vision, machine learning and deep learning, was also featured during her presentation. For the past 17 years, we have worked closely together, notably on the development of tools for identifying plant pathogens, as part of her broader research on applying artificial intelligence to image analysis and the identification of living organisms.
The platform was also represented during the closing plenary session entitled “Supercharging biodiversity conservation: The powers and pitfalls of AI.” Pierre Bonnet, coordinator of the Pl@ntNet platform and botanist at CIRAD, represented the platform alongside Drew Purves, Nature Lead at Google DeepMind, and Juhie Radia, member of the Kew Gardens Youth Council and data analyst.
The discussion focused on the potential of artificial intelligence to transform the digitisation of natural history collections, the use of biodiversity data and scientific practices. It highlighted the many opportunities offered by these technologies while also addressing their limitations, ethical implications and the challenges they raise. Particular attention was also given to issues of data governance, access to technology for countries in the Global South and digital sovereignty.
Pl@ntNet, now deployed at a very large scale, was presented as a concrete example of how artificial intelligence can support biodiversity conservation. The platform was showcased as a citizen science tool that uses AI to facilitate plant identification and biodiversity inventories, relying on several million contributors across more than 200 countries.
Beyond this success, the discussions emphasised one essential point: data quality. Research carried out within Pl@ntNet focuses in particular on automated data curation, the definition of model evaluation standards, large-scale automated mapping and the development of tools capable of turning biodiversity data into information directly usable for conservation. Achieving this level of quality relies to a large extent on citizen participation, and the role of citizen science was therefore strongly emphasised.
Particular attention was also paid to the interoperability between citizen science observations, digitised collections and global taxonomic reference systems—an issue that is especially important within our collaboration with the Royal Botanic Gardens, Kew to ensure that biodiversity data remain robust, well-structured and accessible to both the scientific community and society at large.
These topics, together with the broader work carried out through the international collaborations coordinated by the Royal Botanic Gardens, Kew, are presented in detail in the 2026 report available here. We warmly encourage you to explore this 55-page synthesis.
Beyond the sections dedicated to Pl@ntNet, the report highlights a major step forward for digital biodiversity. In just one decade, the digitisation of more than 145 million plant and fungal specimens has profoundly transformed not only access to biodiversity data but also the opportunities for research, conservation and informed decision-making. However, this momentum will only fulfil its potential if efforts continue to digitise collections, enrich data and strengthen international collaboration, particularly in the world’s most biodiversity-rich regions. This is a vision that we fully share and to which Pl@ntNet is proud to contribute alongside many partners around the world.
