Unverified50% confidenceFactExact time
Common Crawl项目存储了超过2500亿网页的完整HTML快照，压缩后数据量达PB级别，是许多AI模型的主要训练数据来源之一
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
8/11/2026
First Seen
Valid until: 11/9/2026
Sources
AI时代的公地悲剧：谁在透支互联网的公共资源
Related Claims
UnverifiedCLIP的训练数据来自互联网公开爬取的数据集WIT（WebImageText），其规模约为ImageNet（约1400万张标注图像）的近30倍78% similarUnverifiedThe training data for modern image generation models typically comes from billions of internet images and their accompanying text descriptions.76% similarUnverifiedCommon Crawl是非营利组织运营的大规模网络爬取项目，每月抓取数十亿网页，数据集被用于训练包括GPT系列在内的大型语言模型71% similarUnverifiedCLIP was released in 2021 and trained on 400 million image-text paired data collected from the internet.69% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/731023API
curl https://kongchang.com/api/v1/knowledge/claims/731023MCP
get_claim(id=731023)