Unverified50% confidenceFactExact time
早期大语言模型在某些领域的幻觉率高达15-25%
1
Sources
50%
Confidence
Long-term
Relevance
7/9/2026
First Seen
Sources
The Trap of AI Visibility Dashboards: Why Most Metrics Are Meaningless
hackernewshackernews7/7/2026
Related Claims
UnverifiedResearchers have found that the hallucination rate of large language models when answering clinical questions can reach 5% to 15%.79% similarUnverifiedEarly large language models suffered from serious overconfidence, tending to give definitive statements even for uncertain answers, which is the hallucination phenomenon.74% similarVerified研究表明大语言模型规模越大，幻觉的伪装性往往越强，能生成更连贯、更具说服力的错误内容72% similarUnverified研究表明模型在低资源语言上的幻觉率显著高于高资源语言69% similarVerifiedLLM幻觉（Hallucination）是大语言模型的已知问题，会导致AI生成不准确的内容66% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/328272API
curl https://kongchang.com/api/v1/knowledge/claims/328272MCP
get_claim(id=328272)