AI Sprints
David M. Berry Figure 1: Gemini generated image The emergence of Large Language Models (LLMs) presents a remarkable opportunity for humanities and social science research. As I have argued elsewhere, we face an “algorithmic condition” where computational systems increasingly mediate not just our analytical tools but how we understand culture and society more generally (Berry 2025). This also opens the possibility for new methodological approaches capable of exploiting AI’s augmentative capacities whilst maintaining the critical reflexivity that is key to humanistic inquiry. In this article I want to suggest that “AI sprints” might be a possible method for using those capacities, building upon my recent experiments in critical code studies where I have developed an educational application through iterative dialogue with generative AI (Berry 2025b). That experiment, which adapts the idea of “vibe coding” following Karpathy’s (2025) neologism, revealed the suggestive potential of human-AI...
