Scientists invented a fake eye disease to see if AI chatbots could spot it, but the experiment took an unexpected turn

In a groundbreaking experiment, researchers created a fictitious eye disease, bixonimania, to evaluate the accuracy of AI chatbots. Many large language models accepted this non-existent condition as real, emphasizing the dangers of misinformation. Additionally, scientists found themselves referencing bogus research papers without realizing their inauthenticity, showcasing how AI learns patterns from the internet rather than confirming facts.

from Science News: Fresh Discoveries, Research & Breakthroughs https://ift.tt/aOlXNdq