The research critically examines the impact of generative artificial intelligence (GenAI) on educational assessment by highlighting both its transformative potential and associated risks. Findings reveal that GenAI enhances scalability, provides instant feedback, and supports personalized learning, particularly in structured disciplines like STEM. However, it underperforms in humanities assessments that require varied interpretation and creativity. The analysis also shows that while AI can reduce human grading bias, it introduces algorithmic biases that disadvantage AI can reduce human grading bias but diverse learners. Stakeholder analysis identifies benefits for students and educators but emphasizes the need for AI literacy and ethical oversight. Key concerns include academic dishonesty, data privacy violations, and the erosion of human judgment in pedagogy. The study concludes that GenAI should not replace human evaluators but complement them through hybrid models. Future research must focus on explainable AI systems, ethical frameworks, and AI-resistant assessment formats to ensure educational equity, integrity, and long-term learning effectiveness.
Srinivasa Kalyan Vangibhuratha (2025) studied this question.