How can customers balance AI use with sustainability goals?

By TD SYNNEX Newsflash 19th June 2026

The rapid adoption of AI is increasing scrutiny on its environmental impact. According to one report, data centres already account for 6% of electricity in the UK and US, with global consumption rising by 36% globally over the past two years. This raises an important question: is the environmental cost too high a price to pay for expanding access to AI?

► Power-hungry AI data centres under increased scrutiny

► Businesses may need to guardrails to protect sustainability goals

► AI needs to be managed like any other business-critical tool

The figures come from a report by the International Data Center Authority (IDCA), which says annual investment in data centres is approaching $1 trillion, close to 1% of global GDP. Data centres now have a worldwide footprint of 67.7GW and consume 2% of global electricity, up from 1.7% in 2024. The US is the largest data centre market, using 29.2GW annually, representing 43% of global consumption. The UK consumes 2.0GW a year, accounting for 5.8% of national electricity use.

How can customers balance AI use with sustainability goals?

As investment continues, AI is increasingly on a collision course with environmental concerns. Earlier this year, not‑for‑profit organisation Global Action Plan held two days of activity in the UK to highlight the impact of data centres, including a ‘march against the machines’ in London and an online discussion on expansion. Around the same time, the UK government announced a new inquiry into the energy and water demands of data centres and how these may affect net‑zero ambitions.

To balance these competing pressures, organisations may need to introduce guardrails to ensure AI adoption does not undermine sustainability objectives. Sustainability consultancy Tunley Environmental argues that the debate should not be framed as ‘AI is good’ or ‘AI is bad’, but that ‘AI is powerful and needs to be managed like any other major business tool that can significantly affect our performance and risk.’

Carbon consultancy Seedling adds that this may require suppliers to provide clearer information on electricity use, water consumption, and overall carbon impact. All major hyperscalers already offer tools to help measure this impact. Microsoft provides an Emissions Impact Dashboard, AWS offers its Sustainability Console, and Google Cloud has a Carbon Footprint tool.

Seedling also stresses the need for perspective. Organisations do not need to avoid AI, but they should measure and manage its use, avoiding intensive workloads unless necessary. It also advocates encouraging providers to use green electricity, optimise efficiency, locate data centres where low‑carbon power is available, manage water responsibly, extend product lifecycles, adopt circular procurement, and make use of waste heat.