In a multiuser environment one would know who sent the benchmark, who to target. I can think of systems where that would be considered disloyal to the company.
Usually in cases like those submitting results outside of the internal network is shunned on anyways and may be even cause for dismissal. Bottom line is you shouldn't be doing activities that the company would consider disloyal in the first place.
You are missing the point. A known user called root already exists on most systems. As it is also pointed out since the IP isn't reported what are the chances of someone finding your machine and then trying to hack a user account instead of the root account which is what they would want.
Knowing a user account name has nothing to do with social engineering. Social engineering would be tricking you into giving your login credentials with password.
The point is you can easily turn off SSH logins via root in the ssh config file and there's really no valid reason to display user logins.