At a recent AMD conference, OpenAI CEO Sam Altman outlined his vision for the future energy demands of artificial intelligence.

Altman emphasised that running advanced AI models like ChatGPT could eventually require a significant portion of the world’s electricity. His remarks bring into focus the immense computational power needed to sustain AI operations.

This projection raises important questions about the sustainability and scalability of AI technologies. As AI tools become increasingly integrated into various sectors, their energy consumption and related infrastructure requirements present a growing concern, especially when service outages occur.

To add context, AI models such as ChatGPT rely on extensive computations, involving the processing of vast amounts of data to generate accurate and relevant responses.

These processes are computationally intensive, driving demand for powerful hardware and, in turn, significant energy resources. As AI evolves, balancing its benefits with its environmental and infrastructural demands will be critical.

Altman’s comments also hint at a broader challenge: aligning AI development with the global push for carbon neutrality. Data centres powering AI models already consume substantial energy, and as models grow more sophisticated, their energy footprints are expected to rise sharply.

This has led to a surge of interest in “green AI” – an emerging movement focused on designing algorithms and infrastructure that prioritise energy efficiency without sacrificing performance. Companies are now exploring custom silicon, optimised data pipelines, and innovative cooling techniques to reduce emissions associated with AI workloads.

Additionally, the geopolitical implications of AI’s energy needs are becoming harder to ignore. As nations compete to build out AI capabilities, access to reliable and affordable energy sources may influence their strategic positions.

Regions with abundant renewable energy could become new hubs for AI innovation, shifting the traditional centres of technological gravity.

In this light, Altman’s vision not only underscores the scale of AI’s future energy appetite but also the imperative to rethink where and how we build the backbone of tomorrow’s intelligent systems.