There are a few ways to look at this.
1) Regions in the arrange all play back at the session sample rate. Example: 44k session, 96k audio file in arrange=slower playback
2) Logic automatically converts output sample rate so you can record independent of CA devices. Example: You have a session recorded at 96k, your interface is not connected, Logic will load Built in Audio, Logic runs tyhe sessions at 96k and converts the SRate of the session to match the supported sample rate. So you can run sessions at unsupported sample rates, this rarely makes sense if you cannot capture your audio at session sample rate (if needed).
3) There are a few other options for handling this, such as EXS24, which automatically handles SRC.
4) When Importing Audio Files there is a song preference which you can en/disable to automatically convert SR upon import.
I think your friend may have referred to point #2 and it was interpreted as point #1...perhaps. Hope this clears things up. J