“In 2026, AI Is Driving Sustainability Through Action, Not Reports”

Sustainability and AI are no longer being discussed as two separate ideas inside companies. What started as future talk is now influencing daily decisions, from how much energy a factory uses to how supply chains are run in real time. With tighter rules, rising energy costs and closer scrutiny from investors and customers, AI is moving out of pilot mode and into core business operations. In an interview with ResponsibleUs, Pablo Orvananos, Global Sustainability Lead at Hitachi Digital Services, talks about how this shift is unfolding on the ground, why 2026 is shaping up as a turning point for real, measurable impact, and how companies can tell genuine progress apart from well-packaged claims.

In 2026, How AI Is Transforming Sustainability in a Silent Way
In business, sustainability and AI are not talking in parallel anymore. They are quickly becoming the same conversation. At the same time, companies are getting stricter regulations, rising energy costs, and increasing pressure to show actual progress, AI is getting from being an experimental tool to an operational backbone. In 2026, the question is already decided that AI can support sustainability, but it is the other way around; how it is already changing the course of decisions, accountability, and impact throughout the organisations. We had the chance to speak with Pablo Orvanosos, Global Sustainability Lead at Hitachi Digital Services, for this article to see what this shift is like in practice and how to spot genuine progress from the hype that is just well-intentioned.

By 2026, the use of AI in sustainability will go beyond dashboards and reports to the actual implementation of AI. AI is less about static reporting in 2026 and more about actions taken based on AI's decision-making. No more quarterly emissions reports, AI is continuously analysing and adjusting operations embedded in the company. Factories conserve electricity by shifting their loads, the reduction of fuel consumption for the transportation of goods is achieved by the optimisation of logistics routes and the supply of renewable energy is equilibrated in real time. Sustainability is now managed like cost, safety, or quality; it is not anymore something that is looked at after it has happened.

Why do we consider 2026 the year when companies will transition from small tasks to having a significant, measurable impact? Several factors point to this moment on the timeline. The foundations for data are more solid, both cloud and edge infrastructures are up to par, and the regulations are demanding higher-quality disclosures. A lot of companies did pilot testing from 2022 to 2024 and most of the times they were isolated. Such experiments helped in identifying both the effective and ineffective areas. In 2026, the pressure coming from all corners - investors, regulators, and customers - is such that organizations are forced to not only scale proven use cases but also connect them to financial and operational metrics that leadership teams actually trust.

AI is progressively taking over routine operational decisions. It determines during which times machines will operate to minimize emissions from the power grid, how to arrange production according to the supply from renewable sources, or when an asset needs maintenance to avert waste. AI is utilized in procurement to assess suppliers not only based on price but also on the carbon footprint. These moves are made thousands of times daily, thus silently shaping the final results without the need for human supervision at each stage.

Do you think AI is another view for companies to see changes in performance, and what does this mean for the outcomes?
Indeed, and this is the most significant change that has taken place. The transparency of real-time information gives the companies the power to react right away rather than after weeks. The moment the emission levels rise, the system pinpoints the culprit and suggests a solution. If the water consumption goes above the permitted level, the processes are changed right away. This extremely fast feedback loop has a huge positive impact on the outcome because minor inefficiencies get fixed before they turn into huge environmental and financial losses.

AI agents, on the other hand, anticipate what will happen and can suggest countermeasures. AI is already capable of predicting the energy used by a certain company and recommending changes to its operations to save energy. Furthermore, one can look to how the technology currently operates at major companies like Amazon, Google, and Microsoft for examples of the power of prediction in driving corporate behaviour change. Machine learning applications find areas and times where cooling or heating can be switched off without any inconvenience to the staff at the office or the customers in a shop. AI agents go further by deciding what to do next. They operate continuously, learn from outcomes and act within defined boundaries. An agent is able to strike a compromise between conflicting objectives, such as price, carbon footprint and dependability, and pick the most suitable one instantly. This allows sustainability management to be proactive instead of reactive, transferring human functions to supervision and strategizing.

AI consumption is increasing, but how much of that electricity is actually used for data centers?
AI takes considerable power but mostly during the training of big models and operating data centers. The regions where major infrastructure is located experience the greatest power consumption impact, so it is not equally distributed. The concern is real, but it is also becoming better understood and measured.

As AI use grows, companies are also finding ways to use less energy per task. Better hardware, smarter model design and more efficient cooling are helping bring down energy use, even as demand rises. Still, if AI is not managed carefully, it can become a sustainability problem in itself. Many companies are now responding by choosing lighter, more efficient models, timing heavy computing work when renewable power is available, and tightening how they track energy use. Data centre operators, meanwhile, are investing in renewable energy, water efficiency and ways to reuse waste heat.

More boards are also asking a harder question: does the carbon footprint of an AI project make sense compared to the emissions it is meant to reduce?
At the board level, conversations about sustainability have undergone noticeable changes. With political backing for climate action becoming less reliable in some parts of the world, including the US, companies are no longer looking at sustainability mainly through the lens of compliance or image. The focus has shifted to resilience, cost pressures and staying competitive over the long term. AI fits into this shift because it links environmental performance directly with efficiency and business outcomes. The debate now is less about whether sustainability matters and more about how fast and how seriously companies can improve in a less predictable policy environment.

The line between real action and greenwashing is also clearer. Serious companies build AI into everyday operations, measure results properly and are willing to make tough trade-offs. Greenwashing usually shows up as broad claims, cherry-picked data or small pilots that never move beyond slide decks. Clear data, independent checks and decisions that tie sustainability to financial performance are good signs that the work is genuine.

Over the next couple of years, business leaders would do well to focus on basics before chasing the next big tool. Good data, clear ownership and the right skills matter more than novelty. Targets need to be realistic, and sustainability teams must work closely with digital and operations teams. Starting small is fine, as long as successful ideas are built to scale. The companies that do well will use AI to run their businesses better, not as a bolt-on sustainability story.