Researchers from various institutions have developed AdaLens, an interactive system for monitoring and steering long-running agentic data analysis. The system addresses the limitations of conventional interfaces in providing adequate support for observability and steerability in autonomous workflows. AdaLens combines a storyline-based representation with steering interactions to enable analysts to monitor progress and redirect low-value directions or deepen promising ones during execution. --- Why it matters: AdaLens matters because it tackles the challenges of monitoring and controlling complex, long-running AI analysis processes, which are increasingly common in data science applications. By providing an interactive interface for oversight, AdaLens enables analysts to better understand and steer these processes, improving their efficiency and effectiveness. Source: https://arxiv.org/abs/2608.17834