Advancements in artificial intelligence (AI) technologies have led to a rapid increase in the use of AIenabled tools to curate content across various application contexts (e.g., news, social media, search engines, product reviews). This content can include multiple data sources and formats. Some commercially available AI-enabled content curation integrations seem useful, while others are underdeveloped, or not considered in terms of utility, contextual relevance, or user experience. We present an annotated portfolio of nine discrete interface design patterns, exploring how an AI-in-theloop approach can be used to present contextually relevant, AI curated content — across varying degrees of AI involvement. To illustrate these patterns, we use a case study of online news content, to reflexively examine how different content types and use cases are suited to the different interface design patterns. We view this work as a provocation for advancing the discourse on AI-enabled content curation applications.