As 2025 concludes, the information technology sector faces a pivotal moment in its environmental responsibilities. This year marked a significant shift, transitioning sustainability from an aspirational goal to a regulatory requirement. The urgency of this change is underscored by the increasing energy demands of modern computing, particularly driven by the rise of artificial intelligence (AI). The industry now recognizes that IT sustainability is integral to business operations and compliance, rather than merely a parallel initiative.
The past year has seen a transformation in how organizations approach sustainability, elevating the topic to a core focus for executive leadership. No longer is the dialogue centered on reputation; discussions now revolve around compliance with frameworks such as the Corporate Sustainability Due Diligence Directive (CSDDD) and the Corporate Sustainability Reporting Directive (CSRD). These regulations compel companies to confront the complexities of their environmental impact, holding them accountable for detailed, auditable data related to every asset and vendor.
For Chief Information Officers (CIOs), this shift manifests in critical ways. Firstly, energy efficiency is increasingly viewed as an essential component of operational costs, particularly in the face of fluctuating global energy prices. Secondly, the rapid growth of generative AI has sparked internal debates on responsible computing practices. Organizations now find themselves justifying AI investments not just through traditional return on investment (ROI) metrics, but through a new lens of “return on compute,” which incorporates carbon emissions into financial calculations.
Despite these advancements, challenges remain. The IT sector often struggles with the authority and resources needed to enforce significant changes across various departments. Moreover, the budget constraints and risk tolerance necessary for transformative initiatives continue to hinder progress.
Promising Developments and Systemic Challenges
In 2025, tangible progress emerged in areas that could pave the way for future net-zero efforts, though widespread adoption remains limited. Three notable advancements illustrate this trend:
1. **Decoupling Cloud Growth from Carbon Emissions:** Major cloud providers have made substantial strides in renewable energy procurement. The focus has now shifted to optimizing physical operations. For instance, there has been increased interest in advanced liquid cooling technologies, primarily in hyperscale environments, which are essential for supporting the energy demands of AI. Enterprises are optimizing workloads for low-carbon regions and adopting serverless computing models, effectively disconnecting cloud expansion from rising emissions.
2. **Advancing the Circular IT Model:** The Managed Device-as-a-Service (MDaaS) model emerged as a key environmental strategy in 2025. By outsourcing the entire device lifecycle, companies commit to refurbishment and efficient reverse logistics, significantly reducing electronic waste. While some enterprises have successfully implemented these contracts, widespread adoption of MDaaS remains limited, and verification processes for these circular supply chains still need robust third-party oversight.
3. **Emergence of Green Software Engineering:** The development of green software engineering (GSE) represents a promising evolution in the industry. Until recently, sustainability efforts focused predominantly on hardware. This year, organizations began evaluating the energy consumption of their code and optimizing algorithms to minimize reliance on resource-intensive computing. A notable milestone was the release of the W3C Web Sustainability Guidelines, which outlines best practices for reducing the environmental impact of web-based services.
Addressing Gaps in Net-Zero Goals
Despite these encouraging developments, 2025 also highlighted critical gaps that could impede progress towards net-zero targets. Two key challenges persist:
1. **Challenges with Scope 3 Emissions:** One of the most significant obstacles remains the measurement and reduction of Scope 3 emissions, particularly those arising from purchased goods and the disposal of assets. Many companies still depend on aggregated, industry-average data, which lacks the precision required for compliance. The absence of detailed product carbon footprints from suppliers complicates efforts to accurately report emissions. This situation creates a “Scope 3 plateau,” where companies set ambitious targets but fail to address the underlying emissions effectively.
2. **Energy Demands of Generative AI:** While AI offers potential benefits for sustainability, its immediate energy consumption poses a substantial challenge. The rapid adoption of Large Language Models (LLMs) necessitates high-performance computing resources, creating an “energy debt” that undermines sustainability efforts. Many organizations prioritize initial ROI from AI deployments but overlook essential policies for efficient energy use and resource management. Without a framework for responsible computing, the environmental impact of AI could overshadow its intended benefits.
Looking ahead to 2026, the focus must shift from identifying challenges to actively closing these gaps with disciplined strategies. Addressing these priorities is essential for enhancing business resilience in the face of mounting regulatory pressures. The transition of IT sustainability from aspirational goals to mandatory compliance is a significant step forward. The upcoming year should emphasize gathering accurate data, enforcing accountability, and managing the growing energy demands of innovative technologies. The time for aspirational goals has passed; the imperative now is to ensure verifiable progress in the journey toward sustainability.