Even before AI, I can't imagine a single bug bounty researcher doing that. Pre-AI, everyone ran a tool like subfinder to enumerate subdomains, httpx to resolve them, nuclei to scan them, etc. There's no human review involved there at the subdomain level.
And I don't know anyone that would really look at the intermediary of a CNAME even during a manual test. Maybe if it was obviously a third party service.
And I don't know anyone that would really look at the intermediary of a CNAME even during a manual test. Maybe if it was obviously a third party service.