No, they would get a warning if it's self-signed "google.com" certificate which doesn't match the one in the Certificate Transparency registry, or OCSP fails, or HPKP fails to match, DANE/Tack fails to match.
At least that's what a sane implementation would do.
Aren't all of those vulnerable to denial of service attacks? If the "man in the middle" just drops all OCSP requests on the floor, for example, the default case is to accept everything.
Certificate Transparency is to detect compromised or malicious CAs, and typically only lets people know about problems after the fact. How would OCSP help with a self-signed certificate? What CA would the OSCP request go to? And cert pinning only works if you've visited the site before while not being MitMed.
A lot of these extensions to TLS help secure popular sites like Google, but they don't scale to the many small-but-important sites such as your local credit union. Instead of adding so much complexity, it's much easier and safer to keep current behavior and warn on self-signed certs.
At least that's what a sane implementation would do.