As generative AI becomes a bigger part of university life, one issue keeps returning in higher education discussions: assessment. The debate is no longer just about whether students are using AI tools, but about how institutions should respond when traditional coursework can be completed with increasing help from automated systems.
A common first reaction has been to focus on detection. That approach reflects concern over academic integrity, but it also highlights a wider challenge for educators. If assessment is designed mainly around catching AI use, universities may spend more time policing submissions than reconsidering whether existing formats still measure learning effectively.
The idea behind designing beyond detection is a shift in emphasis. Instead of relying primarily on tools to spot machine-generated work, the conversation moves toward creating assessments that are more robust, more authentic and better aligned with how students demonstrate understanding. In that framing, generative AI is not only a compliance problem but a prompt to revisit long-standing assumptions about coursework and evaluation.
The wider message is that assessment in the age of generative AI cannot be solved by a single technical fix. For higher education, the challenge is becoming one of design: building tasks, processes and learning experiences that remain meaningful even as AI tools continue to improve.
