How to scale AI without scaling your carbon footprint

Global, Jul 16, 2026

Written by Nick Zinzan, Head of ESG, Datatec and Logicalis

Artificial intelligence has rapidly moved from experimentation to execution. Across every industry, organisations are investing in AI to improve productivity, accelerate innovation, enhance customer experiences and unlock new revenue streams. The race is on. 

But as AI adoption accelerates, a crucial question is emerging in boardrooms around the world: how do you scale AI without scaling energy consumption, operational costs and carbon emissions?

For many organisations, AI strategy and sustainability strategy are still being treated as separate conversations. That is becoming increasingly difficult to justify. As AI workloads grow, leaders must consider not only what AI can achieve, but how those outcomes are delivered. 

The most successful organisations will be those that balance innovation with responsibility, ensuring AI delivers long-term value without creating unintended environmental consequences. 

AI’s hidden sustainability challenge

The benefits of AI are well documented. From automating repetitive tasks to enhancing decision-making and generating new business insights, AI is helping organisations work smarter and move faster. 

Behind the scenes, however, AI requires significant computing power. Training and running large language models, processing vast datasets and supporting increasing demand for AI-powered applications places additional pressure on cloud environments, networks and data centres. 

This is creating a new leadership challenge. Every AI initiative must now be evaluated across four dimensions:

  • Business value
  • Security and governance
  • Operational resilience
  • Environmental impact

Many organisations are already recognising this shift. The 2026 Global CIO Report highlighted that less than half of CIOs feel confident their organisation actively manages AI’s environmental impact, suggesting there is still significant work to be done to align AI innovation with sustainability objectives. 

The reality is simple: AI may be digital, but its environmental footprint is very real. 

Why sustainable AI matters to business leaders

Sustainable AI, despite its name, is not just an environmental issue but a business issue. Organisations that scale their AI investments with poor infrastructure or badly governed workloads can drive up operational costs, increase energy consumption and create new compliance challenges. At the same time, stakeholders are demanding greater transparency around environmental performance and responsible technology adoption. 

When approached correctly AI can in fact help operations to reduce costs, improve operational efficiency and create a more resilient digital foundation. In other words, sustainability should not be viewed as a constraint on AI adoption. But it does need to be accounted for to enable long-term success. 

Measure what matters – using AI to measure and manage emissions

Measurement is where sustainable AI starts to become actionable. 

IBM Envizi, for example, uses AI-assisted sustainability data management to help organisations capture, calculate and report Scope 1, 2 and 3 emissions, while supporting forecasting and scenario planning. Logicalis uses Envizi internally to support its carbon reporting and net zero journey, helping improve the consistency and confidence of its sustainability data. Other organisations are applying the same principle to operational efficiency: Cisco has used AI and advanced analytics to optimise energy, airflow and cooling in its labs, while Google has used DeepMind’s machine learning to reduce energy used for data centre cooling by up to 40%. 

Four actions leaders should take now

1. Build governance into every AI initiative 

Organisations need clear ownership, accountability and oversight for AI deployments. This includes establishing policies that address ethical considerations, data quality, transparency and environmental impact. 

2. Optimise infrastructure for efficiency

Many organisations can significantly reduce energy consumption through smarter infrastructure decisions including:

  • Modernising legacy environments
    • Optimising cloud architectures
    • Retiring underutilised applications 
    • Improving data management practices

Sustainability and efficiency often go hand in hand. Reducing waste in your technology estate typically benefits both your environmental footprint and your budget. 

3. Measure what matters

Incorporate sustainability indicators into technology reporting, including energy consumption, infrastructure utilisation and carbon impact. 

4. Align AI strategy with sustainable business outcomes

Rather than deploying AI simply because it is available, organisations should focus on initiatives that create measurable value while supporting wider business priorities, for example, reducing operational waste, optimising resource utilisation or improving productivity. 

Sustainable AI is part of a wider sustainable IT strategy

AI doesn’t exist in isolation. Its environmental impact is influenced by the wider technology ecosystem that supports it, from data centres and cloud platforms, to networking infrastructure and device lifecycles. By embedding AI into broader sustainable IT strategies, leaders gain a holistic view of technology decisions, considering factors such as energy efficiency, circularity, asset lifecycle management and carbon reporting, all alongside AI performance. 

Organisations that thrive in the AI era will not simply be those that deploy the most AI solutions. It will be those that deploy them most effectively. By building technology environments that are efficient, resilient and sustainable by design organisations can gain long-term competitive advantage. Because ultimately the winners of the AI race will not just be the fastest adopters. It will be those who build AI for the long term, with sustainability in mind. 

Related Insights