Generative AI is increasingly contaminating citizen science databases, with researchers warning that manipulated wildlife imagery could skew critical data on species distribution and climate-driven migration. According to a commentary in the journal Nature, hundreds of AI-generated or enhanced images have been detected on major platforms like iNaturalist and the Macaulay Library, threatening the integrity of records used by conservationists to track biodiversity in real time.
The Risk to Citizen Science Data Integrity
Citizen science platforms rely on the accuracy of user-submitted photographs to monitor how ecosystems respond to environmental changes. Tony Iwane, director of community support at iNaturalist, notes that these platforms function as a real-time sensor for global shifts, such as plants flowering earlier or species moving toward cooler latitudes due to climate change. When AI-enhanced images are uploaded, they introduce “noise” into datasets that scientists depend on for large-scale conservation planning.
While iNaturalist has flagged approximately 1,400 images for AI use out of more than 610 million entries, experts suggest this likely represents only a fraction of the total issue. Many AI-edited photos bypass detection, potentially leading to false reports of rare species in regions where they do not naturally exist.
Did you know?
In one documented case, a photographer used AI to “improve” an image of an epaulet oriole in Brazil. The software added features from a red-winged blackbird—a North American species—resulting in a false record of a bird that had never been sighted in that area.
Why AI Editing Distorts Biodiversity Records
Dr. Alexander Lees, a researcher involved in the Nature commentary, emphasizes that the primary danger is not necessarily malicious hoaxes, but the common practice of “polishing” wildlife photos. Birders often use AI tools to remove distracting branches or improve lighting. However, these algorithms often pull data from different species to fill in gaps, effectively creating a hybrid creature that does not exist in nature.
This creates a significant obstacle for researchers. If an AI-altered image is accepted as a legitimate sighting, it can lead to incorrect conclusions about a species’ range or behavior. As Dr. Lees points out, wildlife photographers are often driven by the aesthetic quality of their work, but this aesthetic pursuit now carries a direct risk of “contaminating” scientific records.
Frequently Asked Questions
Scientific research relies on accurate, verified data to track species. AI tools can add or alter physical characteristics of birds, leading to incorrect species identification and false location records.
No. According to Tony Iwane of iNaturalist, most instances are likely the result of users attempting to improve the visual quality of their photos without realizing the impact on scientific data.
Related reading