Accountability regimes typically encourage record-keeping to enable the transparency that supports oversight, investigation, contestation, and redress. However, implementing such record-keeping can introduce considerations, risks, and consequences, which so far remain under-explored. This paper examines how record-keeping practices bring algorithmic systems within accountability regimes, providing a basis to observe and understand their effects. For this, we introduce, describe, and elaborate 'accountability capture' -- the re-configuration of socio-technical processes and the associated downstream effects relating to record-keeping for algorithmic accountability. Surveying 100 practitioners, we evidence and characterise record-keeping issues in practice, identifying their alignment with accountability capture. We further document widespread record-keeping practices, tensions between internal and external accountability requirements, and evidence of employee resistance to practices imposed through accountability capture. We discuss these and other effects for surveillance, privacy, and data protection, highlighting considerations for algorithmic accountability communities. In all, we show that implementing record-keeping to support transparency in algorithmic accountability regimes can itself bring wider implications -- an issue requiring greater attention from practitioners, researchers, and policymakers alike.

In the field of machine learning, getting from point A to point B is never a totally straightforward journey. While designing a deep learning algorithm, there are a plethora of factors that will change over time and affect the way your potential product operates. The path to production will very likely involve all kinds of detours, backtracking, and changes on the fly. Luckily, there are ways to keep track of these changes and ensure that your project is ready for the changing variables that come along; the field is called machine learning for a reason, after all. Here are a few ways to teach your machines to adjust algorithms as they go.