ai-agents misinformation healthcare-ai llms

Scientists invented a fake disease. AI told people it was real

nature.com

April 10, 2026

9 min read

🔥🔥🔥🔥🔥

49/100

Summary

Scientists created a fictitious disease and used AI to convince people of its existence. The experiment demonstrated AI's capability to influence public perception about health issues.

Key Takeaways

Researchers created a fictitious disease called bixonimania and published fake studies to test AI's ability to recognize misinformation.
Major AI systems began to treat the invented condition as real, demonstrating a vulnerability to misinformation.
The fabricated studies were cited in peer-reviewed literature, indicating that some researchers rely on AI-generated references without verifying the sources.
The lead author of the fake studies was a fictional character created using AI, highlighting the potential for AI to generate misleading academic content.

Read original article

View story →

Coverage & discussions

Hacker News · nature.com

Scientists invented a fake disease. AI told people it was real ↗

Community discussions

Hacker News discussion ↗

Historical count: 5 · 5 saved in sample · Historical sample

Analysis ready

Community Sentiment

Negative

Concerns

The lack of counter narratives for LLMs to learn from raises concerns about the integrity of information, making them susceptible to manipulation.
There is a significant risk that individuals or groups could intentionally 'poison' LLMs, undermining their reliability and potentially causing financial harm to AI companies.
The ease with which fake information can influence AI product recommendations highlights vulnerabilities in the systems that rely on LLMs for decision-making.
Trustworthiness of sources is critical for both humans and AI, and the current publishing system may not adequately ensure this, leading to misinformation.