Notice how "evidence" and "courts" aren't words used anywhere in this.
That's because universities are not the criminal justice system. If a victim wants to file criminal charges, that's a completely separate process, independent of whatever the university does. The university data may be subpoenaed, and an online database is almost certainly not protected by patient-counselor privilege, but the court would need to validate that data for trial.
This is about university disciplinary hearings, which have the historical perception of bias against the woman. ie "she was asking for it," "she only changed her mind afterward," or "keep this quiet because it would hurt the university." There's perception, at least on the part of women, that this bias is facilitated by most of the faculty being men, thus identifying with the alleged perpetrator, and by the secretive nature of the hearings, thus making it hard to find patterns of repeated behavior. Those university-internal investigations are serious, even if not criminal, and a disciplinary panel that accepts a bunch of blog posts without following up to verify is dangerously incompetent. University disciplinary findings are protected student information, beyond the outcome, so other schools or potential employers won't know whether the expulsion is for sexual assault or for pissing in the dean's lunchbox.
Sure, people could conspire to use this tool to make false accusations, but conspirators so set on disrupting a boy's life don't need a tool. This is not "Click a button to expel Johnny Boogerhead." This is a way to collect names of people who might corroborate an open accuser's claims.