<![CDATA[Background: Artificial Intelligence (AI) has transformed English language learning through personalized instruction and real-time feedback. However, current literature shows a critical gap: studies focus either on technological capabilities or pedagogical problems in isolation, lacking frameworks that balance innovation with humanistic teaching principles. Aims: In order to investigate AI integration in English language learning, this systematic review will: (1) identify opportunities that AI presents; (2) analyze challenges and constraints; (3) develop a conceptual framework that strikes a balance between technological innovation and humanistic pedagogy; and (4) offer evidence-based recommendations for educators and policymakers.Methods: Thirty peer-reviewed publications from Web of Science, Scopus, ERIC, and Google Scholar were systematically reviewed between January 2023 and August 2025 in accordance with PRISMA principles. A modified CASP checklist was utilized for quality assessment. Humanistic Learning Theory and TPACK frameworks served as the basis for the analysis.Results: 73% of studies (n=22) found that accessibility and personalization improved writing accuracy and student motivation. But 50% (n=15) expressed worries about emotional engagement and ethical integrity. Problems included decreased personal connection, especially in collectivist societies (27%, n=8), algorithmic prejudice (17%, n=5), and plagiarism facilitation (30%, n=9). The four guiding concepts of the Innovation-Empathy Balance Model are educational intentionality, professional capacity building, ethical vigilance, and strategic complementarity.Conclusion: AI-enhanced English language instruction has a number of prospects, but it necessitates striking a balance between humanistic pedagogy and technology innovation. The Innovation-Empathy Balance Model emphasizes intentional deployment, teacher preparation, and preserving human connection in education, positioning AI as a strategic supplement to human-centered teaching.]]>
