A large language model (LLM) has been successfully integrated with an 8-bit shoot-'em-up game, PvP-AI, running on a Commander X16 emulator, using a novel approach called "smart senses"1. This method involves providing the LLM with structured text summaries based on the game's existing inputs, rather than relying on traditional pixel or audio data. The LLM, specifically the ChatGPT API model gpt-4o, demonstrates impressive capabilities, including maintaining notes between turns, developing strategies across games, and even discovering an exploit in the game's built-in AI behavior. The experiment showcases the potential of LLMs to interact with complex systems in innovative ways, leveraging their ability to process and generate human-like text. This development has significant implications for the field of artificial intelligence, as it highlights the potential for LLMs to be applied to a wide range of domains, from gaming to more complex problem-solving tasks, so what matters is how this technology can be harnessed to drive further advancements in AI research and applications.