§ feed · storyline
Researchers create hallucinatory image that confuses AI models
Researchers develop a hallucinatory image that causes GPT-4o to rate it 6.5 out of 10 and triggers a malfunction in Qwen.
Researchers created a "hallucinatory image" for AI: GPT rated it 6.5 out of 10, while Qwen simply malfunctioned. eu.36kr.com
§ sources1 publication · timeline below