4 articles.
The most-cited study of AI grading put 80 multi-turn questions to GPT-4 and 58 expert annotators. GPT-4 matched human preference 85% of the time, above the 81% two humans managed. But when researchers took an answer that was already correct, added a redundant restatement of it at the top, and re-judged, Claude and GPT-3.5 were fooled 91% of the time, every judge favoured whichever answer appeared first, and on maths a reference answer cut the judge's failure rate from 70% to 15% while plain step-by-step reasoning left 6 failures in 20.
A 2026 audit of 125 million references found fabricated citations rising more than twelvefold, and a benchmark of 13 models found citation hallucination rates spanning 14% to 95%. The references your AI hands you are the least trustworthy thing on the page.
Turnitin scores are losing legal and academic ground in 2026. Process analytics and oral defenses are replacing them, and that changes what you actually need to keep.
Real April 2026 benchmark numbers on the detectors your professor might be using, plus what actually protects you when a false flag lands on writing you wrote yourself.