I think their point is that the people registering certs may not intend their sites to be immediately scraped, but now OpenAI is bypassing e.g. google indexing or web spidering, and using your cert provider's CT entries to find you immediately for scraping.