I Asked AI to Grade a Viral "Christian Persecution" Paper.
Read on to find out what Gemini & ChatGPT thought about it.
On December 1, a grading dispute at the University of Oklahoma (OU) exploded into a national culture war. The headlines were perfect for social media outrage: A Christian student, Samantha Fulnecky, claimed a transgender instructor gave her a zero for expressing biblical views on gender.
According to the narrative pushed by Turning Point USA (TPUSA), this was a clear case of religious persecution. The group labeled the instructor, Mel Curth, as “mentally ill” and framed the dispute as a battle between sanity and “lies”. Even Oklahoma Governor Kevin Stitt got involved, calling for an investigation.
But behind the viral outrage lies a much simpler, drier question about academic standards: Did Fulnecky fail because of bias, or was the grade a fair judgment of her work?.
To answer this, we have to look past the politics and conduct an audit of the assignment, the student’s submission, and, I submit, the independent grading by two AI models. The results seem to reveal a fundamental misunderstanding of what a university education is actually for.
