Imagine a chatbot that not only engages in flowing conversations but also understands the nuances of human interaction, remembers your preferences, and even offers insightful recommendations. This isn't science fiction; it's the reality of AutoCompanion, a new socialbot built by researchers that combines the strengths of large language models (LLMs) like GPT-4 with the logical reasoning power of Answer Set Programming (ASP). LLMs excel at generating human-like text, but they often struggle with consistency, factual accuracy, and staying on topic. AutoCompanion addresses these shortcomings by using LLMs for natural language processing—translating user input into a structured format that a computer can understand—and then employs ASP to reason over this knowledge. Think of it like this: the LLM is the smooth-talking social butterfly, while the ASP is the meticulous planner behind the scenes, ensuring the conversation stays engaging, coherent, and factually grounded. AutoCompanion's knowledge base spans movies, books, and notable figures, allowing it to discuss plot points, character analysis, and even recommend relevant titles based on user preferences. The ASP engine acts as a topic controller, strategically shifting the conversation based on shared properties between topics. For example, mentioning Leonardo DiCaprio in "Titanic" might lead to a discussion about "Catch Me If You Can." Furthermore, AutoCompanion accurately answers questions by querying its knowledge base and honestly admits when it doesn't know something, avoiding the "hallucinations" that often plague LLMs. This innovative approach allows AutoCompanion to achieve greater reliability, scalability, and controllability than LLM-only socialbots. What makes AutoCompanion truly stand out is its ability to mimic the human thought process during conversation. It considers user preferences, maintains consistency in opinions, and dynamically adjusts its responses to keep the conversation flowing naturally.While still under development, AutoCompanion demonstrates the potential of combining LLMs with symbolic reasoning systems like ASP. Future work aims to expand the knowledge base, enhance multi-modality by incorporating images and voice interactions, and most importantly, deploy AutoCompanion on online platforms to gather user feedback. This technology promises not only more engaging and reliable socialbots but also a deeper understanding of how we can use AI to augment human communication and connection.