Artificial intelligence is becoming a critical tool in the fight to prevent plant extinction, with new technology enabling scientists to identify vulnerable species faster than ever before and unlock genetic data that could lead to new medicines. According to a major report from Royal Botanic Gardens, Kew, about 40% of the 70,000 plant species assessed are at risk of extinction, while another 330,000 species have yet to be analyzed. With roughly 2,000 new plant species recorded each year, scientists warn that potential life-saving medicines and sustainable crops are vanishing before researchers even discover them.

How Can AI Help Save Plants From Extinction?

AI models are transforming plant conservation in several concrete ways. The technology can learn to identify challenging species like sedges and peat mosses, whose distinguishing features are microscopic and difficult for humans to spot. Machine learning systems trained on digitized plant images can now sometimes identify species more accurately than specialist botanists, accelerating the discovery and protection of vulnerable plants before they disappear.

Beyond identification, AI is unlocking genetic secrets from historical specimens. Scientists have developed methods to extract high-quality genomes from fungal specimens up to 180 years old, turning museum collections into what researchers call a "genomic goldmine" for developing new medicines and predicting disease outbreaks. Penicillin and statins, two of the most important medicines in modern medicine, were both derived from fungi, illustrating the potential hidden in these collections.

• Species Identification: AI models can identify microscopic plant features faster and sometimes more accurately than human specialists, enabling rapid discovery of new or vulnerable species.
• Genetic Analysis: Machine learning can extract high-quality genetic data from museum specimens dating back 180 years, creating a resource for developing new medicines and understanding disease patterns.
• Global Collaboration: Digitization and online access to millions of specimens previously locked in archives enables international research partnerships, especially in biodiversity hotspots like Madagascar.
• Climate Tracking: AI trained on digitized specimens can detect how flowering times have shifted across the globe, revealing disruptions to ecosystems caused by climate change.

What Do Digitized Plant Collections Reveal About Climate Change?

A global study using AI trained to spot flowers analyzed 8 million digitized specimens and revealed a striking pattern: flowering times have shifted by an average of 2.5 days per decade over the last century due to climate change. Rising temperatures and changing rainfall patterns have caused some flowers to arrive earlier while others bloom later, disrupting ancient relationships between plants and the animals that pollinate them at specific times of year.

The consequences are already visible in real ecosystems. A study using herbarium specimens showed that about 80% of kindal trees in India's Western Ghats, which are important for timber production, used to flower at the same time historically. By the 1990s, that synchronized flowering had collapsed to less than half, demonstrating how climate disruption is unraveling ecological relationships that evolved over millennia.

"While documenting and protecting all life on Earth remain formidable challenges, digitisation and accompanying technologies make me increasingly hopeful that we'll succeed," said Prof Alexandre Antonelli, executive director of science at Royal Botanic Gardens, Kew.

Prof Alexandre Antonelli, Executive Director of Science at Royal Botanic Gardens, Kew

How Big Is the Digitization Effort Underway?

Royal Botanic Gardens, Kew has digitized all 7.4 million of its specimens, including plants collected by Charles Darwin, making them freely available online. The four-year digitization program involved taking 20,000 high-resolution images daily at its peak. Globally, there are now 145 million digital specimens available online, but this represents less than 16% of all specimens held in herbarium collections worldwide, leaving what scientists describe as "huge blind spots in understanding".

The effort extends beyond Kew. In Madagascar, one of the world's most extraordinary biodiversity hotspots, digitization of 37,000 physical specimens has unlocked centuries of knowledge about the region's biodiversity. This work is particularly important because many developing nations hold vast plant collections that were previously inaccessible to international researchers.

"Madagascar is one of the world's most extraordinary biodiversity hotspots. By digitising [37,000 physical specimens], we've unlocked a treasure of knowledge spanning centuries, offering invaluable insights into today's biodiversity," said Landy Rajaovelona, a senior botanist at Kew Madagascar.

Landy Rajaovelona, Senior Botanist at Kew Madagascar

What About Fungi, and Why Does It Matter?

The extinction crisis is even more severe for fungi than for plants. An estimated 2 million fungal species exist on Earth, but 90% remain unknown to science, and less than 1% of known species have been assessed for extinction risk. This matters because fungi are not just curiosities; they are essential to human survival and health. Beyond penicillin and statins, fungi produce countless other medicines and play critical roles in soil health, nutrient cycling, and ecosystem function.

Climate change is also affecting fungi in ways scientists are only beginning to understand. As temperatures rise and seasons lengthen, some human pathogens appear to be spreading from warmer regions into temperate zones, raising concerns about disease outbreaks in areas previously protected by cold winters.

What Are the Challenges Ahead?

The report, produced by 400 scientists across 40 countries, acknowledges significant concerns about the environmental cost of AI itself. Data centers now consume 6% of electricity in the UK and US, raising questions about whether the energy required to train and run AI models undermines the environmental benefits of using AI for conservation. The authors caution that digitization and AI could amplify existing biases and inequalities unless the underlying data is expanded and improved, particularly in the global south where many biodiversity hotspots are located.

To address these challenges, the report calls for partnerships between technology companies and environmental organizations, as well as increased government and philanthropic funding for plant and fungi collections. The scale of the task is enormous: with about 100,000 plant species still to be named by scientists and millions of fungi species unknown to science, the race against extinction will require sustained investment and international collaboration for decades to come.