GhostWriter is an AI-enhanced writing environment designed to personalize the writing process through style and context personalization. It leverages large language models (LLMs) to learn the user's intended writing style implicitly as they write, while also allowing explicit teaching moments through manual style edits and annotations. The system aims to address the challenges of personalization and control in LLM-powered writing systems, which can frustrate users due to their limited personalization and control. In a user study with 18 participants, GhostWriter was evaluated on two different writing tasks: professional editing and creative writing. The results showed that GhostWriter helped users craft personalized text generations and empowered them by providing multiple ways to control the system's writing style. Participants found the system easy to use and intuitive, and they appreciated the various methods to interact with style and give feedback. The study also revealed that participants felt in control of the experience and believed the system was learning from them. The key contributions of GhostWriter include its ability to offer content personalization through style and context, the user study results demonstrating its utility in generating personalized text, and a set of design recommendations for future research on LLM-assisted writing. The findings provide insights into how people view the role of AI in the context of AI-assisted writing and offer guidance for practitioners creating similar solutions.GhostWriter is an AI-enhanced writing environment designed to personalize the writing process through style and context personalization. It leverages large language models (LLMs) to learn the user's intended writing style implicitly as they write, while also allowing explicit teaching moments through manual style edits and annotations. The system aims to address the challenges of personalization and control in LLM-powered writing systems, which can frustrate users due to their limited personalization and control. In a user study with 18 participants, GhostWriter was evaluated on two different writing tasks: professional editing and creative writing. The results showed that GhostWriter helped users craft personalized text generations and empowered them by providing multiple ways to control the system's writing style. Participants found the system easy to use and intuitive, and they appreciated the various methods to interact with style and give feedback. The study also revealed that participants felt in control of the experience and believed the system was learning from them. The key contributions of GhostWriter include its ability to offer content personalization through style and context, the user study results demonstrating its utility in generating personalized text, and a set of design recommendations for future research on LLM-assisted writing. The findings provide insights into how people view the role of AI in the context of AI-assisted writing and offer guidance for practitioners creating similar solutions.