Unverified60% confidenceFactTime unknown
Research shows that the probability of common vulnerabilities such as SQL injection, XSS, and insecure dependency references in code generated by large language models is significantly higher than that of experienced human developers.
2
Sources
60%
Confidence
Long-term
Relevance
5/31/2026
First Seen
Sources
Related Entities
Related Claims
Unverified研究表明LLM生成的代码在SQL注入、XSS防护、密钥硬编码等方面的错误率显著高于有经验的人类开发者85% similarUnverified直接让 LLM 生成 SQL 风险很高，因为模型不了解表结构、字段含义和业务定义，容易生成语法正确但结果错误的查询69% similarUnverifiedEarly large models in cross-source data research tasks were prone to missing key sources, numerical confusion, or broken inference chains, forcing agents to repeatedly supplement queries.67% similarUnverified提示注入与SQL注入有概念相似性，都源于未能有效分离代码与数据的信任边界67% similarUnverifiedFor PostgreSQL, which does not support MATCH_RECOGNIZE, workarounds based on recursive CTEs or window functions exist in the community, but code complexity increases significantly and performance varies66% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/2488API
curl https://kongchang.com/api/v1/knowledge/claims/2488MCP
get_claim(id=2488)