aceAI-altered images on birdwatching forums putting research at risk
For many birdwatchers, recording a species outside its normal range is the holy grail. In the UK, the discoveries often make national headlinesThe western reef heron, for example, usually found in Africa and southern Europe, spotted in a seaside town in north Wales in June, which was widely celebrated on birding forums. But a new scourge is threatening to disrupt the fun: AI slop. Scientists are appealing to birders to limit their use of AI when editing images over fears that it could undermine the credibility o…
On 20 July 2026, The Guardian reported that scientists are warning birdwatchers against using generative AI to create or enhance wildlife photos. Researchers writing in Nature said hundreds of AI-altered images have been found on citizen science databases such as iNaturalist and Macaulay Library, which scientists use to track where species occur.
The contamination matters because these public platforms function as large-scale sensors for research, including monitoring how species move in response to climate breakdown. If AI edits introduce false range records, the credibility and research value of the datasets could be degraded, and the true number of undetected fakes remains unknown.
- Researchers warned in Nature commentary that hundreds of fake images have been found on species recording databases.
- iNaturalist has flagged 1,400 of more than 610m images for AI use, but true scale is unknown.
- Example cited: a false red-winged blackbird sighting in central Brazil created when AI editing added features to an epaulet oriole photo.
AI-generated and AI-enhanced bird images posted to wildlife forums are contaminating citizen science records and undermining platforms like iNaturalist that researchers use to monitor species ranges.
The rundown
The piece describes how generative tools like ChatGPT and Google Gemini enable users to create fake rare-bird images in seconds or to ask AI to remove a branch, which can inadvertently add features from other species. A cited case involved a photographer asking AI to make a picture 'look better', turning an epaulet oriole into a false red-winged blackbird record in central Brazil.
iNaturalist director of community support Tony Iwane, co-author on the Nature commentary, said most cases were unlikely to be malicious and appealed for vigilance. The article notes outright hoaxes like a 'toucan sighting in Siberia' remain rare and easy to spot, while subtle AI edits pose the harder detection challenge.
Sources
- JournalismThe Guardian2026-07-20
How should this claim be treated?
The debate