Unverified50% confidenceOpinionExact time
The "pelican riding a bicycle" test can simultaneously examine a generative video model's compositional generalization, object interaction understanding, and temporal consistency.
1
Sources
50%
Confidence
Long-term
Relevance
10/22/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedSimon Willison has long used 'a pelican riding a bicycle' as a fixed test case to comparatively evaluate different AI models across different time periods.71% similarUnverifiedThe "draw a pelican riding a bicycle" test was originally proposed by developer Simon Willison to evaluate the capability boundaries of large language models and image generation models.69% similarVerifiedSimon Willison uses SVG generation of "a pelican riding a bicycle" as an informal benchmark test of model capabilities.68% similarUnverifiedThe benchmark covers multiple detection dimensions, including mouse trajectory analysis, request header consistency, JavaScript execution environment detection, and interaction timing patterns.65% similarUnverifiedExplicitly requiring the model to repeatedly check its own work via screenshots, video, or by running the game during game development is an effective prompting technique for improving output quality.63% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1087216API
curl https://kongchang.com/api/v1/knowledge/claims/1087216MCP
get_claim(id=1087216)