Purpose: This study examines the impact of the Amira Learning AI-powered reading platform on literacy outcomes for elementary students in 12 school districts in Louisiana during the 2023-24 school year. Methodology: Employing a quasi-experimental design aligned with the Every Student Succeeds Act Level II (Moderate Evidence) standard, a matched-student sample of 79,084 K-5 students (39,542 treatment; 39,542 comparison) was formed. Student demographics included approximately 38% African American, 38% White, 18% Hispanic, 4% multi­racial and 2% Asian; 75% were economically disadvantaged; 10% were English-language learners; and 14% received special education services. Usage data (e.g., total session minutes, number of passages read) for the Amira platform were merged with demographic and assessment outcomes (e.g., DIBELS, LEAP) for analysis. Findings: Students who used Amira demonstrated statistically significant higher literacy outcomes compared to matched peers without access to the platform. Moreover, higher dosage (in terms of minutes and passages read) was associated with larger gains. The magnitude of effect met the threshold for meaningful educational impact under ESSA's Level II guidance. Conclusions: The results suggest that integrating the Amira platform into elementary literacy instruction in high-need contexts can contribute to improved reading growth. Implementation fidelity and student engagement emerged as important moderators of impact. Implications: For practitioners and policymakers in K-5 literacy, this study provides evidence that an AI-based reading tutor can yield measurable benefits when deployed at scale, especially in economically disadvantaged and diverse student populations. Future research should explore longitudinal impacts, differential effects across subgroups (e.g., ELL, special education), and cost-effectiveness of full-scale implementation. [This report was prepared by Instructure for Amira Learning®.]