<p>Confronted with the profound challenges posed by the rapidly aging global population, smart elderly care has emerged as a vital solution for enhancing the quality of life for older adults and alleviating the socioeconomic burden on healthcare systems. Grounded in socio-technical systems theory, the Unified Theory of Acceptance and Use of Technology (UTAUT), and welfare pluralism theory, this study identifies and analyzes key barriers to the implementation of Artificial Intelligence of Things (AIoT) in elderly care from the perspectives of multiple stakeholders, proposing an integrated, theoretically-informed policy framework to facilitate effective adoption. Methodologically, a qualitative research design was employed, incorporating semi-structured interviews and field observations conducted in Beijing’s Xicheng District—a representative urban area with a high aging rate and active smart elderly care initiatives. The findings reveal five major challenges: systemic fragmentation and a lack of unified standards, a persistent mismatch between service supply and elderly needs, low technology adoption among older adults due to digital literacy and trust issues, a severe shortage of interdisciplinary professional caregivers, and over-reliance on government subsidies with limited market participation. In response, the study proposes an integrated policy framework that emphasizes standardizing AIoT systems, tailoring services to elderly needs, increasing financial investment, empowering communities, enhancing digital inclusion for older adults, and professionalizing the caregiving workforce. The study concludes that coordinated multi-stakeholder efforts are essential to leverage AIoT for building sustainable, responsive, and human-centered elderly care ecosystems.</p>