AI密集型软件开发成本测量：早期结果与警示
原标题：Building AI-Intensive Software with AI: Early Results and a Cautionary Tale on Measuring Development Cost
AI 摘要
一项案例研究显示，六人学生团队在一个学期内使用AI辅助构建了基于RAG的对话式入职助手，并采用三层成本模型测量开发成本。初始报告的成本比为19.4，但后续发现两个测量错误（订阅制下每token成本推断错误和反事实劳动力定价错误），修正后为9.9。研究强调AI辅助开发成本测量容易出错且难以察觉，呼吁更严谨的测量方法。
正文节选
Building AI-Intensive Software with AI: Early Results and a Cautionary Tale on Measuring Development CostConference: a; September 11, 2026; São Paulo, SP, Brazil Abstract. Empirical reports on the true cost of AI-intensive software development remain scarce, and the few that exist are easy to get wrong in ways that never surface in the final number. We report early results from an ongoing case study: a six-person student team built a full conversational onboarding assistant — RAG-based code chat