Verified65% confidenceFactTime unknown
Large models are prone to 'hallucination' when handling professional domain questions, meaning they confidently provide incorrect answers.
3
Sources
65%
Confidence
Long-term
Relevance
6/1/2026
First Seen
Sources
AIGC到AI Agent：2025多智能体开发核心思路与实战指南
bilibili码士集团-马小雪
Related Entities
Related Claims
Verified"Hallucination" refers to models generating content that appears authoritative but is factually incorrect, with very high confidence.79% similarUnverifiedLarge language models exhibit an unfaithful execution problem: Opus sometimes arbitrarily changes definitions on its own, and upon finding that results do not match, writes code to override the derived results and directly fills in the answer it considers correct.76% similarUnverifiedThe hallucination problem in Agents is more severe than in ordinary dialogue scenarios, because erroneous reasoning is amplified through the action chain.73% similarUnverifiedLarge models have a hallucination problem, meaning they generate API calls that appear reasonable but do not actually exist, or incorrect function parameters73% similarVerifiedResearch indicates that models may produce unfaithful chains of thought, where the surface-level reasoning steps are inconsistent with the actual basis for decision-making.72% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/27275API
curl https://kongchang.com/api/v1/knowledge/claims/27275MCP
get_claim(id=27275)