Artificial intelligence could help win the race against extinction of vital plants, botanists say
Artificial intelligence is changing the game in one of the quietest and most urgent battles of our time: saving plants and fungi before they vanish forever.
A report produced by 400 scientists from 40 countries, published by the Royal Botanic Gardens, Kew, delivers data that is both striking and deeply alarming. Around 40% of the 70,000 plant species already assessed are at risk of extinction. Another 330,000 have not even been evaluated yet. On top of that, an estimated 100,000 plant species still do not have a scientific name. That means medicines, foods, and solutions we have not even discovered yet could be disappearing before anyone gets a chance to study them.
It is a massive problem — but technology is opening up paths that would have been unthinkable just a few years ago. 🌿
The scale of a problem we barely notice
When we think about extinction, the image that usually comes to mind involves large animals — lions, elephants, whales. But the plant crisis is just as severe, and its consequences hit much closer to home than most people realize. Plants and fungi form the foundation of virtually everything that sustains life on Earth: they provide food and medicine, store carbon, and regulate the climate across the entire planet.
The Royal Botanic Gardens, Kew report makes it clear that the pace of cataloging and studying species falls far short of what is needed. Around 2,000 new plant species are recorded each year, but as Professor Alexandre Antonelli, executive director of science at RBG Kew, pointed out, that barely scratches the surface of what needs to be done. With hundreds of thousands of species still waiting to be identified and assessed, scientists face a colossal logistical challenge. Each species needs to be collected, described, named, photographed, genetically analyzed, and monitored over time — a process that, done the traditional way, can take decades for a single species.
The situation with fungi is even more alarming. There are an estimated 2 million fungal species on the planet, and 90% of them remain completely unknown to science. Less than 1% of known species have been assessed for extinction risk. Considering that fungi gave us breakthroughs like penicillin and statins, losing this biological wealth before it can even be studied represents a potentially enormous cost to humanity.
Meanwhile, deforestation, climate change, and habitat degradation keep advancing at breakneck speed. The scariest part is not what we already know we are losing. It is what we have not even discovered yet. Species that could carry compounds capable of treating diseases that still have no cure, or food crops resistant to drought and pests that could feed billions of people in the future, are vanishing before they even get a name. That is where artificial intelligence comes in as one of the most promising tools science has at its disposal today. 🔬
Receive the best innovation content in your email.
All the news, tips, trends, and resources you're looking for, delivered to your inbox.
By subscribing to the newsletter, you agree to receive communications from Método Viral. We are committed to always protecting and respecting your privacy.
How artificial intelligence is speeding up plant discovery and protection
Artificial intelligence has a capability that no human scientist can replicate alone: processing massive volumes of data in dramatically less time, identifying patterns that would be invisible to the naked eye. In the context of plant biodiversity, that translates into concrete tools already being used in the field and in laboratories around the world.
One practical example cited in the report involves plants that are notoriously difficult to identify, such as rushes and peat mosses, whose distinguishing features are microscopic. AI models trained to recognize these species can analyze images and accurately pinpoint details that would escape even expert eyes. According to Antonelli, these AI models can sometimes identify species better than specialists — something he described as extremely exciting.
This ability to identify species quickly and accurately is critical for detecting new or vulnerable species with far greater speed. Instead of waiting months or years for confirmation from a taxonomist, AI provides a first filter that accelerates the entire scientific workflow, from field collection to the publication of new records.
Shifts in flowering patterns revealed by AI
The report also highlights a fascinating global study that used an AI model trained to identify flowers in images. Researchers analyzed 8 million digitized specimens and discovered that flowering periods have shifted by an average of 2.5 days per decade over the past century due to the climate crisis. Changes in rainfall patterns and rising temperatures have caused some flowers to bloom earlier and others later.
That shift might sound subtle, but the consequences run deep. It can break apart long-standing relationships between plants and the pollinators and other animals that depend on them at specific times of year. A study using herbarium specimens showed that around 80% of kindal trees in India’s Western Ghats, important for timber production, used to flower at the same time. By the 1990s, that number had dropped to less than half. This desynchronization can compromise plant reproduction and destabilize entire ecological chains. 🌍
Digitization as a bridge between botany’s past and future
Another front where artificial intelligence is making a significant difference is the digitization of historic botany collections. Herbaria around the world hold collections spanning hundreds of years — pressed specimens, handwritten notes, detailed scientific illustrations, and collection records that document past biodiversity in a unique way. The problem is that much of this material has never been digitized, or when it was, it was not in a format that allowed automated analysis.
RBG Kew completed the digitization of all 7.4 million of its specimens, including materials collected by Charles Darwin, and made the entire collection available for free online. The program, which spanned four years, was producing 20,000 high-resolution images per day at its peak. Globally, 145 million digital specimens are now available online, but that represents less than 16% of the total stored in herbaria around the world, leaving what scientists described as enormous blind spots in our understanding of biodiversity.
The case of Madagascar and the importance of global access
Digitization is also opening doors in regions that are true biodiversity treasure troves but have historically had limited access to their own collections. Madagascar is a prime example. Landy Rajaovelona, a senior botanist at Kew Madagascar, emphasized that the island is one of the most extraordinary biodiversity hotspots in the world. With the digitization of 37,000 physical specimens, it became possible to unlock a treasure trove of knowledge spanning centuries, offering valuable insights into current biodiversity.
This kind of initiative is particularly relevant for the Global South, where much of the still-unknown plant and fungal biodiversity is concentrated. By digitizing and making these collections available online, scientists from anywhere on the planet can collaborate on research without having to travel physically, speeding up knowledge production and democratizing access to botanical information. 📂
Fungi: a genomic gold mine hidden in century-old specimens
One of the most impressive findings cited in the report involves the ability to extract high-quality genetic information from fungal specimens up to 180 years old. Scientists managed to produce complete genomes from materials that had been stored in collections for nearly two centuries, transforming historic fungal collections into what researchers called a genomic gold mine.
This achievement has direct implications for the search for new medicines and for predicting disease outbreaks. Fungi are extremely opportunistic organisms — as Dr. Esther Gaya, a senior researcher at RBG Kew, explained, they love heat and moisture. Some human fungal pathogens appear to be spreading from warmer regions as the warm season grows longer in temperate areas. Understanding the genetics of these organisms over time can help science anticipate risks and develop faster responses.
The challenges technology still needs to overcome
Despite all the progress, it is important to understand that artificial intelligence cannot solve the plant extinction problem on its own. The technological tools are powerful, but they depend on quality data to work well — and collecting that data still requires human fieldwork, robust funding, and public policies that protect the habitats where species live.
The report’s own authors acknowledge that intensive AI use raises environmental concerns. The data centers powering these models consume enormous amounts of energy and water. In May, it was reported that data centers already consume 6% of electricity in the United Kingdom and the United States, a figure that is expected to grow as demand for AI processing increases.
Biases and inequalities that need to be addressed
There is also the issue of data equity. The report warns that digitization and AI can amplify existing biases and inequalities unless the underlying data is expanded and improved. AI models tend to perform better for regions and species that already have plenty of available records. Tropical forests in Africa, Asia, and South America, which hold much of the still-unknown plant biodiversity, have historically received less scientific attention and have less complete databases. That means algorithms may be least accurate precisely where they are needed most.
To tackle this gap, the report proposes partnerships between tech companies and environmental organizations, along with investments from governments and funders in plant and fungal collections. The idea is for technological infrastructure and scientific knowledge to advance together, ensuring that the benefits of AI reach every region of the planet — not just those that already have abundant resources and data.
A race against time that now has powerful allies
The Royal Botanic Gardens, Kew report reinforces that the race to catalog, protect, and understand the planet’s plants demands a collective, multidisciplinary effort — scientists, technologists, governments, local communities, and the general public. As Professor Antonelli summed it up, although documenting and protecting all life on Earth remain formidable challenges, digitization and the technologies that accompany it make the possibility of success increasingly hopeful.
Artificial intelligence is proving to be an extraordinary ally on this journey. It can identify species with accuracy surpassing that of human experts, reveal climate changes invisible to the naked eye in historical records, unlock genetic information from century-old specimens, and accelerate global scientific collaboration in ways that were unimaginable just a few years ago. But the tool is only as good as the human willingness to use it with purpose and responsibility.
The clearest message this report delivers is that the time to act is now — because every species that disappears takes with it possibilities we can never get back. 🌱
