Al-altered images on birdwatching forums putting research at risk

2026-07-20 · Source: AI (artificial intelligence) | The Guardian · Field: Science & Research — Environmental Science & Earth Systems, Life Sciences & Biology, Research Methodology & Innovation · Depth: Fundamental Awareness, short

Summary

AI-altered images are increasingly appearing on birdwatching forums, jeopardizing the credibility of citizen science platforms like iNaturalist and Macaulay Library, which scientists use to monitor species' habitat ranges. The proliferation of generative AI tools such as ChatGPT and Google Gemini enables users to create high-quality fake images or enhance existing photographs, inadvertently introducing significant changes, such as combining features from different bird species. Researchers, including Dr. Alexander Lees, warn that hundreds of fake images have already been discovered on popular databases, with the true extent of contamination unknown. This "AI slop" makes it difficult to accurately track species distribution and undermines the real-time data collection vital for conservation efforts, as exemplified by a false red-winged blackbird sighting in Brazil caused by AI enhancement.

Key takeaway

For citizen science platform managers and ecologists relying on public data, the rise of AI-altered images demands immediate attention. You must implement robust detection mechanisms and clear guidelines for image submissions to preserve data integrity. Unverified AI enhancements can corrupt long-term ecological records. It is crucial to educate contributors on these risks and emphasize authentic, unedited observations for reliable scientific monitoring.

Key insights

AI-altered images on citizen science platforms compromise ecological data integrity, risking scientific research.

Principles

• AI image editing can subtly falsify biological records.
• Citizen science data demands uncompromised authenticity.
• Unchecked AI use degrades scientific data reliability.

In practice

• Exercise vigilance for AI-altered images.
• Limit AI use in wildlife photo editing.
• Verify unusual species sightings rigorously.

Topics

• AI-altered Images
• Citizen Science
• Data Integrity
• Ecological Monitoring
• Generative AI
• Wildlife Photography

Best for: AI Scientist, CTO, VP of Engineering/Data, Research Scientist, Domain Expert, AI Ethicist

Related on AIssential

• Distributed Denial of Science: How Indirect Data Poisoning of AI Systems Can Industrialize Scientific Fraud — Indirect data poisoning weaponizes open data ecosystems and AI agents to industrialize scientific fraud at scale.
• AI in nature conservation: powerful tool or dangerous shortcut? — AI offers powerful conservation tools but demands careful application to avoid amplifying biases and alienating communities.
• Will AI ruin the social sciences — or revolutionize them? — Large language models are simultaneously compromising social science research data integrity and offering tools for enhanced analytical robustness.
• Addressing Image Authenticity When Cameras Use Generative AI — Camera-integrated AI can introduce "hallucinations," necessitating methods to recover original, unhallucinated image content.
• AI-generated images fool Spokane city officials twice in six days — Unverified AI-generated content poses significant risks of misinformation and intellectual property infringement for public and private organizations.
• scosman/pelicans_riding_bicycles — Intentional data poisoning can disrupt AI model training by injecting misleading associations.

See Counsel's argued verdicts on the open AI decisions leaders are weighing →

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by AI (artificial intelligence) | The Guardian.