Artificial Intelligence indeed is having its moment in the public zeitgeist. However, as awe-inducing and incredible as AI seems, it is nothing short of a wrecking ball to the planet. However, a team of researchers from the Stockholm International Peace Research Institute have published a paper that can help reverse the adverse effects of AI on the climate, by leveraging the very thing that has accelerated its destruction.
AI is using gallons upon gallons of water. Consider this — Microsoft alone used more than 2,500 Olympic-sized swimming pools of water in its data centers last year. OpenAI is used even more than that. And these figures only take water used to cool down the power-hungry servers into account. Just how much electricity are these AI-running data centres.
Authored by Dr Kyungmee Kim and Dr Vincent Boulanin, both researchers at SIPRI, the paper, titled Artificial Intelligence for Climate Security: Possibilities and Challenges, explores various ways in which AI can be a godsend in disaster management, climate security, and, more importantly, in minimising actions that contribute to climate change.
AI-powered algorithms can analyze vast amounts of data to provide more accurate and timely warnings for natural disasters such as hurricanes, floods, and wildfires, helping communities better prepare and evacuate.
Optimized Resource Management
Machine learning algorithms can optimize resource allocation and management, particularly in agriculture, by predicting crop yields, optimizing irrigation, and identifying areas susceptible to drought or soil degradation.
Precision Agriculture
AI-enabled precision agriculture techniques can minimize resource usage while maximizing crop yields, thereby reducing the environmental footprint of farming practices.
Autonomous Robots in Disaster Response
AI-driven autonomous robots can efficiently deliver humanitarian aid and support during and after climate-related disasters, ensuring faster and more effective response efforts.
Improved Climate Hazard Modeling
Machine learning algorithms can enhance climate hazard modelling, enabling better predictions of extreme weather events and their potential impacts on communities and infrastructure.
Data Collection in Conflict-Affected Areas
AI technologies, including satellite imagery and social media analysis, can help gather crucial climate-related data in conflict-affected and fragile regions where traditional data collection methods may be challenging or inaccessible.
Community Engagement
Researchers utilizing AI to study the intersection of climate change and security should maintain close ties with affected communities, ensuring that their work addresses local needs and concerns while respecting human rights and privacy. For this, AI and ML tools can be a great enabler.
However, to facilitate all of this and to enable AI and ML to function in a more open and transparent manner, there has to a be series of major policy changes. In order to harness the virtually limitless potential of deploying AI for good, a global, multidisciplinary and multistakeholder conversation is needed.
The conversation needs to be focussed on:
Ethical Considerations
Policymakers and researchers must address ethical concerns surrounding AI applications in climate security, including issues of data privacy, bias in algorithmic decision-making, and the potential for misuse by authoritarian regimes.
Funding Digital Literacy and Access
Supporting digital literacy and access to digital infrastructure is essential for developing AI ML tools effectively in climate change mitigation and adaptation efforts, especially in vulnerable communities.
Prioritising Funding into Research
Policymakers should prioritize funding for critical research on AI and climate security, focusing on addressing methodological challenges, ensuring data reliability, and minimizing the risk of bias in AI models.
