Grading exams using large language models: A comparison between human and AI grading of exams in higher education using ChatGPT