Balancing Innovation and Sustainability: Weighing the Benefits and Environmental Costs of AI in Healthcare Administration

The healthcare landscape is undergoing a rapid digital transformation, with Artificial Intelligence (AI) increasingly embedded across the system, from diagnostic innovations to tools that improve operational efficiency. AI’s ability to handle massive datasets, classify patterns, and make forecasts has made it particularly appealing in healthcare (Mohsin Khan et al., 2025). Machine Learning (ML) and Deep Learning (DL) algorithms help anticipate patient health issues, evaluate medical imaging, and support surgical procedures, ultimately improving patient outcomes. Alongside these clinical contributions, AI has also transformed healthcare administration by enabling seamless communication across healthcare networks and optimizing hospital workflows, many of which now rely on AI-supported systems (Krotkiewicz et al., 2025).

It is non-negotiable that ethical efforts to improve patient outcomes should be encouraged. At the same time, growing reliance on energy-intensive technologies must be considered within the broader context of climate change. Healthcare already accounts for approximately 4–8% of total global CO₂ emissions (Ibrahim Alzoubi et al., 2025). While sustainability-focused approaches are emerging in AI development, many widely used models still require substantial computational resources. This reality underscores the importance of being intentional and selective about when and how AI is used in healthcare, particularly in administrative contexts where its necessity is often less clear.

The rapid adoption of AI in healthcare administration has been driven by both system-level pressures and technological advances. Since the COVID-19 pandemic, increasing patient demand and resource constraints have accelerated the need for digital transformation (Krotkiewicz et al., 2025). At the same time, refined computational models, powerful hardware, and access to vast datasets have enabled advances in Natural Language Processing (NLP), AI voice technologies, automated assistants, and robotics (Chen & Decary, 2020). These developments have made it possible for AI to predict patient volumes and staffing needs, automate documentation through voice integration with electronic health records, analyze physicians’ notes using NLP, streamline operational workflows, and detect fraud, waste, and abuse.

Empirical evidence highlights the scale of these administrative gains. AI-driven automation has been shown to reduce patient wait times by 30%, robotic process automation has lowered administrative workloads by 40%, and predictive analytics has decreased hospital readmission rates by 22% (Krotkiewicz et al., 2025). These improvements demonstrate why AI is often seen as a solution to chronic inefficiencies in healthcare systems.

However, these benefits come with environmental costs that are often overlooked. The economics of AI is grounded in physical commodities, particularly energy consumption (Bogmans et al., n.d.). Healthcare systems already contribute significantly to greenhouse gas emissions, with hospitals and their supply chains identified as major sources (Ibrahim Alzoubi et al., 2025). Projections suggest that from 2025 to 2030, additional emissions driven by AI growth could cumulatively equal Italy’s entire energy-related greenhouse gas emissions over a five-year period (Bogmans et al., n.d.).

The scale of AI’s infrastructure helps illustrate this impact. Data centers have expanded rapidly, with server-filled warehouse space in Northern Virginia alone now roughly equivalent to the floor area of eight Empire State Buildings (Bogmans et al., n.d.). Training a single large AI model can emit as much carbon dioxide as five cars over their lifetimes (Ueda et al., 2024). Beyond electricity use, AI systems require frequent hardware replacement, generating electronic waste that often contains toxic materials such as lead, cadmium, and mercury. Additional environmental burdens stem from mining rare earth elements and from the transportation and logistics involved in producing and maintaining AI infrastructure (Ueda et al., 2024).

Given this growing footprint, a more balanced approach to AI adoption in healthcare administration is needed. The American Medical Association’s framing of AI as “augmented intelligence” offers a useful starting point, emphasizing careful evaluation of whether AI meaningfully supports or surpasses human performance for a given task and considering the consequences of potential errors (Chen & Decary, 2020). From a sustainability perspective, this evaluation should also include an assessment of energy efficiency and environmental impact.

A task-by-task approach can help healthcare leaders determine when AI use aligns with both operational priorities and sustainability goals. AI is most justifiable when tasks are clearly defined, long-term benefits outweigh simpler non-AI alternatives, and organizations have the expertise and governance structures needed for responsible implementation. Prioritizing energy-efficient or task-specific models, limiting AI use to clearly bounded functions, and maintaining human oversight are essential to preventing overreliance and unnecessary environmental harm.

AI undeniably offers powerful tools to improve efficiency and coordination in healthcare administration, but innovation should not come at the expense of sustainability. Selective, context-driven AI adoption allows healthcare leaders to distinguish when AI is truly necessary and when simpler solutions are sufficient. By making these distinctions explicit, healthcare systems can continue to innovate while aligning administrative practices with broader ethical and environmental responsibilities.

References

Bogmans, C., Gomez-Gonzalez, P., Melina, G., Pescatori, A., & Thube, S. (n.d.). Power Hungry: How AI Will Drive Energy Demand . International Monetary Fund.

Chen, M., & Decary, M. (2020). Artificial intelligence in healthcare: An essential guide for health leaders. Healthcare Management Forum, 33 (1), 10–18. https://doi.org/10.1177/0840470419873123

Ibrahim Alzoubi, Y., Topcu, A. E., & Elbasi, E. (2025). A systematic review and evaluation of sustainable AI algorithms and techniques in healthcare. IEEE Access, 13 , 139547–139582. https://doi.org/10.1109/ACCESS.2025.3596189

Krotkiewicz, M., Szynkaruk, A., & Stachyra, A. (2025). Digital transformation in healthcare management: From artificial intelligence to blockchain. Wiadomości Lekarskie, 3 , 578–583. https://doi.org/10.36740/WLek/202445

Ueda, D., Walston, S. L., Fujita, S., et al. (2024). Climate change and artificial intelligence in healthcare: Review and recommendations towards a sustainable future. Diagnostic and Interventional Imaging, 105 (11), 453–459. https://doi.org/10.1016/j.diii.2024.06.002