As artificial intelligence (AI) revolutionizes industries and daily life, its environmental footprint has become a pressing concern.

While AI holds promise for addressing environmental challenges, it simultaneously poses a significant threat due to its massive energy consumption and the acceleration in generating e-waste. This paradox presents a complex challenge for companies integrating AI into their IT strategies.

Similarly, MIT's AI Risk Repository identifies “Environmental Harm” as one of the major risks posed by AI, warning that unchecked growth in AI technologies could exacerbate existing environmental issues, such as climate change and resource depletion.

The Environmental Risks of AI

AI's energy demands are particularly alarming. Training and maintaining large AI models require vast amounts of computational power, which in turn draws heavily on electricity. The usage of ChatGPT, answering hundreds of millions of daily queries, can cost around 1 GWh each day, which is the equivalent of the daily energy consumption for about 33000 U.S. households.

Furthermore, the upkeep of AI systems, which requires frequent updates and new hardware, leads to the generation of significant amounts of e-waste. Without proper disposal or recycling, this e-waste can release toxic chemicals into the environment, harming ecosystems and human health.

Last but not least, the increasing demand for AI-related computing power puts additional strain on global energy infrastructure.

The role of IT Strategy

For IT strategy managers, addressing these challenges is crucial but complex. The lack of transparency from major AI companies regarding their emissions complicates immediate action: this opacity makes it difficult for organizations to quantify and address their environmental impact.

However, IT strategy is inherently forward-looking, requiring a long-term vision to guide present decisions. By formulating clear objectives and roadmaps, organizations can begin to manage AI's environmental footprint while aligning with sustainability goals.

Assess the environmental impact

Begin with a comprehensive evaluation of how AI adoption affects your organization's environmental footprint. Understanding current and future energy consumption and e-waste generation can help identify areas for improvement and sustainability optimization.

Integrate environmental criteria into decision-making

Embed environmental considerations into all strategic decisions related to AI. Whether it's evaluating new AI investments or planning deployments, every initiative should be scrutinized for its potential environmental impact. Engage stakeholders across departments to raise awareness and promote sustainability in AI development, deployment, and maintenance.

Set environmental KPIs

Define Key Performance Indicators (KPIs) that measure environmental performance across the organization at the level of the IT strategy, not limited to single initiatives. By regularly measuring the evolution of these KPIs, organizations can track the overall environmental impact of AI integration, aligning with broader sustainability goals.

Adopt a lifecycle perspective to limit e-waste

In addition to responsible e-waste management, IT strategy managers should adopt a lifecycle perspective for AI technologies. This means considering the environmental impact at every stage—from development and deployment to maintenance and disposal. By assessing the full lifecycle of AI systems, organizations can make more sustainable choices that align with both technological needs and environmental goals.

Conclusion

AI has the potential to play a positive role in addressing the environmental crisis, but only if its own environmental risks are managed effectively. As energy consumption and e-waste from AI technologies continue to rise, organizations cannot afford to ignore these issues.

While transparency from AI providers is still lacking, IT strategy managers can start laying the groundwork for long-term sustainability. By integrating environmental considerations into their strategic frameworks, organizations can turn the challenge of AI's environmental impact into an opportunity for innovation.

IT strategy is a marathon, not a sprint, and the road ahead requires developing a clear vision for sustainability that accounts for the unseen risks of today's technologies.

This proactive stance not only mitigates risks but positions companies as leaders in sustainable technology adoption. As the AI landscape evolves, so too must our strategies.

By Fabio Turel

Keywords: Sustainability, Project Management, IT Strategy