How to Write with an LLM

In his latest post, security researcher Thomas Ptacek explores the practical nuances of integrating Large Language Models into the writing process. Rather than viewing LLMs as autonomous content generators, Ptacek argues for a collaborative approach where the human remains the primary architect of intent and structure. He details specific workflows for using AI to overcome writer's block, refine prose, and automate tedious editing tasks without sacrificing the author's unique voice. The article emphasizes that effective writing with an LLM requires iterative prompting and rigorous verification, treating the model as a sophisticated assistant rather than a replacement for critical thinking. By maintaining control over the narrative arc and factual accuracy, writers can leverage AI to enhance productivity while ensuring the final output remains authentic and coherent. This guide serves as a pragmatic framework for professionals looking to incorporate generative tools into their creative and technical writing workflows.