As government representatives gathered in Geneva for the United Nations' first discussions on global AI governance, environmental advocates warned that a critical issue has received little attention: the potential impact of AI on biodiversity and ecosystems. The talks, aimed at establishing international frameworks for the ethical development and use of artificial intelligence, have largely focused on human-centric concerns such as privacy, bias, and job displacement. However, campaigners argue that the environmental footprint of AI, including its energy consumption and effects on natural habitats, must be integrated into governance discussions.
According to experts, the rapid expansion of AI technologies requires vast computational resources, leading to increased energy demand and carbon emissions. Data centers that power AI models consume significant amounts of electricity and water, contributing to resource depletion. Moreover, the deployment of AI in sectors like agriculture, forestry, and mining can disrupt ecosystems if not properly regulated. For instance, AI-driven precision agriculture may optimize crop yields but could also lead to monocultures and loss of biodiversity. Similarly, autonomous systems in mining might accelerate resource extraction without adequate environmental safeguards.
The United Nations Environment Programme (UNEP) has highlighted that AI could both help and hinder environmental goals. On one hand, AI can enhance monitoring of deforestation, wildlife populations, and pollution; on the other hand, its unchecked expansion poses risks. Campaigners urge that any global AI governance framework must include binding commitments to assess and mitigate ecological impacts. They call for transparency in AI's energy use, lifecycle assessments of hardware, and inclusion of environmental stakeholders in policymaking.
Notably, some companies are exploring ways to leverage advanced computing for environmental benefit. For example, D-Wave Quantum Inc. (NYSE: QBTS) has been investigating how quantum computing can optimize complex systems, potentially reducing energy consumption in AI training and other applications. Quantum computers may offer more efficient processing for certain tasks, lowering the carbon footprint of AI operations. However, such innovations are still in early stages and require supportive policies to scale.
The Geneva discussions mark a starting point for global AI governance, but environmental advocates stress that nature must be a core consideration. Without explicit attention to biodiversity and ecosystems, AI development could exacerbate environmental degradation. As the UN continues its consultations, campaigners will push for a holistic approach that balances technological progress with planetary health.
