Generative AI raises short-term productivity by completing tasks that learners would otherwise practice on their own. Whether this substitution erodes frontier skill, the skill behind top-tail non-AI-aided performance, is an open question of rising stakes. The sharper question is whether selection mechanisms can screen apart two coexisting types: substitute-users, who use AI in place of deliberate practice, and complement-users, who use it to accelerate skill development. In elite programming, the International Collegiate Programming Contest (ICPC) and the International Olympiad in Informatics (IOI) prohibit AI under proctoring and admit entrants through qualification rounds, whereas online Codeforces (CF) contests are unproctored and open to all. From CF histories we build an AI-prompt signature (more first-attempt acceptances, fewer attempts and retries) consistent with AI-assisted practice. Three patterns triangulate institutional screening. First, CF practice shifted toward this signature across cohorts over two AI rollouts. Second, in open CF contests a stronger signature predicts smaller rating gains for users with no ICPC/IOI affiliation, but not for those who qualified for the AI-prohibited contests. Third, inside the AI-prohibited ICPC environment, a shift toward AI-style practice predicts higher non-AI-aided scores for AI-era entrants. The same practice input carries opposite signs depending on whether the environment screens for it. The contrast points to two levers: how AI is integrated into training, since within the screened pool AI-style practice coincides with stronger non-AI-aided performance; and the design of AI-prohibited evaluation gates as a type-separating institution. Both extend beyond programming to credentialing systems (medical and legal boards, professional certification) that certify skill in a workforce increasingly shaped by AI.

翻译：生成式人工智能通过代劳学习者本应自主练习的任务，短期内提升了工作效率。但这种替代是否会侵蚀前沿技能——即非人工智能辅助下顶尖表现所依托的能力——是一个日益紧迫的未决问题。更尖锐的议题在于，筛选机制是否能够区分并存的两类使用者：替代型用户（用AI替代刻意练习）与互补型用户（借AI加速技能发展）。在精英编程领域，国际大学生程序设计竞赛（ICPC）与国际信息学奥林匹克竞赛（IOI）在监考环境下禁止使用AI，并通过资格赛选拔参赛者；而在线Codeforces（CF）竞赛则无监考且向所有人开放。基于CF的历史数据，我们构建了一个AI提示特征（首次通过率更高、尝试与重试次数更少），这与AI辅助的练习模式一致。三种模式共同验证了机构筛选的作用。第一，在两轮AI部署期间，CF练习活动整体呈现向该特征转移的趋势。第二，在开放的CF竞赛中，更强的AI提示特征预示着无ICPC/IOI背景用户的积分增长更小，但对已通过AI禁止竞赛资格的用户则无此效应。第三，在禁止AI的ICPC环境中，AI时代的参赛者若练习风格更接近AI辅助模式，则其非AI辅助下的表现得分更高。相同的练习输入因环境是否进行筛选而呈现相反的信号方向。这种对比揭示了两个杠杆：首先是如何将AI融入训练体系——因为在被筛选的群体中，AI风格的练习与非AI辅助下的优异表现相关；其次是设计禁止AI的评估门槛作为类型分离机制。这两点不仅适用于编程领域，更可延伸至认证体系（医学与法律执业考试、专业资格认证），这些体系在日益受AI影响的劳动力市场中承担着技能认证功能。