Unverified50% confidenceFactExact time
AI code generation tools are trained on massive amounts of publicly available code, and their output snippets may be highly similar to existing projects governed by licenses such as GPL and Apache, creating a risk of license contamination.
1
Sources
50%
Confidence
Long-term
Relevance
10/3/2026
First Seen
Sources
Related Entities
Related Claims
VerifiedAI生成的代码由于训练数据中包含大量符合规范的开源项目，在遵循标准模式方面往往比手写代码更加一致，这可能提升了审核通过率79% similarUnverifiedAI-generated code is synthesized from countless open-source fragments, causing license compliance, attribution, and liability assignment to become ambiguous.75% similarVerifiedA 2022 Stanford University study found that developers using AI coding assistants had a significantly higher probability of writing code containing security vulnerabilities compared to a control group that did not use AI.75% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/970867API
curl https://kongchang.com/api/v1/knowledge/claims/970867MCP
get_claim(id=970867)