Unverified60% confidenceFactExact time
OpenAI uses the identifier 'GPTBot' for its training data crawler.
2
Sources
60%
Confidence
Long-term
Relevance
8/6/2026
First Seen
Sources
TIME向AI爬虫投喂专属网页内嵌广告：媒体变现新策略
Related Claims
UnverifiedThe training frameworks for models such as OpenAI's GPT series, Meta's LLaMA, and Google's Gemini are all based on PyTorch or JAX (Python interface).75% similarVerifiedgpt-oss is a model series open-sourced by OpenAI.73% similarUnverifiedOpenAI has publicly discussed that their model was specifically trained to efficiently use the apply_patch tool.72% similarUnverifiedOpenAI和Anthropic的大模型都大量使用了GitHub开源代码、Stack Overflow问答、技术文档等公开数据进行训练71% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/697195API
curl https://kongchang.com/api/v1/knowledge/claims/697195MCP
get_claim(id=697195)