A recent academic job interview took a sharp turn when a candidate attempted to use ChatGPT during a traditional chalk talk. This interview format, designed to assess spontaneous thinking without aids, clashed directly with the candidate's modern scientific workflow, which heavily relies on large language models for drafting, ideation, and experimental design.

The candidate argued that using AI is now standard practice, comparing the restriction to barring a carpenter from using a hammer. They explained their process involves prompting models like ChatGPT and Claude for introductions, experimental controls, and grant specific aims. This approach, they contended, demonstrates sophisticated understanding and efficient research execution, not a lack of foundational knowledge.

Despite the candidate's defense of AI-augmented cognition as the current reality of scientific collaboration, the search committee cited concerns about independent thinking and foundational knowledge. The rejection highlights a significant disconnect between evolving research methodologies and entrenched academic hiring rituals, leaving the candidate frustrated but optimistic about industry roles.

The incident reveals how academic institutions may struggle to adapt to AI's integration into the scientific process, potentially overlooking skilled candidates who leverage these tools effectively. This chalk talk rejection underscores a broader debate about evaluating expertise in an AI-assisted world.