Imagine a world where accessing legal aid is as simple as asking a question. That's the vision researchers are exploring, using the power of large language models (LLMs) to streamline the often complex and time-consuming intake process. For those struggling with legal issues like housing disputes, securing legal help can feel like navigating a maze. Long wait times, complicated eligibility rules, and limited resources create a significant barrier to justice. This research dives into how AI could help by automating the initial screening process. Researchers built a digital intake platform that combines logical rules with LLMs to assess eligibility. Applicants simply describe their problem, and the AI, armed with program-specific guidelines, determines their likelihood of qualifying. Tested with eight different LLMs, the results are promising. The top-performing model achieved an impressive 82% accuracy (F1 Score), holding significant potential for closing the access to justice gap. But how does it work? The LLM receives the applicant's description and the relevant rules as input. It then responds with one of three outcomes: 'qualifies,' 'does not qualify,' or 'needs more information.' This initial assessment helps filter applicants, saving valuable time for both those seeking help and legal aid staff. While the technology is still in its early stages, it offers a glimpse into a future where AI empowers individuals to understand their legal standing quickly and efficiently. However, challenges remain. Ensuring fairness and mitigating potential biases in the training data is paramount. The next steps involve refining the process, integrating the tool with online platforms, and simplifying rule updates for legal aid programs. This research underscores the potential for LLMs to transform the legal aid landscape, opening doors to justice that are often closed to the most vulnerable members of society.