Environmental Commission One

FORUM: Environmental Commission One

QUESTION OF: The Use of Artificial Intelligence for the Benefit of our Planet

SUBMITTED BY: Australia

CO-SUBMITTERS: Djibouti, EU, FAO, Gabon, Germany, Greece, Greenpeace, Kazakhstan, Lao PDR, Paraguay, Philippines, Singapore, Syrian Arab Republic, Thailand, Ukraine, Zambia

THE ENVIRONMENTAL COMMISSION,

Recalling UN Resolution A/78/L.49 on “Seizing the opportunities of safe, secure and trustworthy artificial intelligence systems for sustainable development”,

Affirming the need for regulatory frameworks that promote transparency and accountability in AI systems,

Recognizing the acceleration of global warming, spearheading the melting of ice and consequently, the rise of ocean levels,

Understanding the potential of artificial intelligence, including its capabilities to predict and precisely moderate risk factors of sustainable agriculture and the likelihood of natural disasters,

1. Authorises the expansion of machine-learning algorithms - incorporating collaboration with geostationary satellites - to track, monitor, and predict atmospheric risk factors alluding to humidity, temperatures, and air pollutants in order to;

1. predict, track, and assess natural disasters and environmental challenges
2. connect machine-learning predictions with existing early warning systems to trigger alerts for communities at risk;

2. Expands FAO capabilities to collaborate with local and national governments to track, monitor, and promote agricultural sustainability such as but not limited to:

1. water management and crop optimisation
2. regenerative agriculture and agroforestry
3. disaster response systems to lessen the impact of natural disasters
4. environmental monitoring to combat issues such as deforestation and desertification;

3. Minimizes costs associated with AI development whilst maximizing efficiency by following the development patterns of new technologies such as Deepseek that have already minimized operational costs associated with AI development by;

1. prioritizing the research/development of efficient, environmentally-friendly machine-learning complexes by collaborations with international research agencies and the UNEP
2. collaborating with relay-based satellites to expand access to AI technologies in LEDCs
3. investigating energy-efficient AI architectures/energy sources to power machine-learning complexes
4. encouraging investment and collaboration between governments, private companies, and NGOs to fund efficient and cost-effective AI technologies;

4. Ensures the implementation of an ethical and legal framework for AI systems to prevent the misuse of AI systems in a way that could harm communities or the environment such as perpetuating existing inequalities;

5. Requests AI development processes with the inclusion of environmental impact assessments to ensure that technologies will have a net positive effect on the planet through:

1. investigating the effects and extent of AI system’s electricity and water consumption
2. establishing standardised procedures for measuring the environmental footprint of AI which includes but is not limited to - sub clause a
3. managing energy storage by predicting energy demand and organising the integration of stored energy into the grid;

6. Maintains that AI usage will be limited in terms of the influence maintained over governmental systems by:

1. forming lists shall be formed through global consensus to define what global issues are worthy of AI intervention and what issues are not
2. encouraging Member States to submit reports detailing their AI usage on global matters to an unbiased organisation such as the UN who can define whether their usage is responsible or not;

7. Encourages Member States to promote digital assistance to less economically developed countries to implement ethical and effective global use of AI;

8. Implement measures to reduce the ecological impacts of AI use, such as:

1. Incorporating carbon-neutral AI infrastructures in the design and operation of AI systems, including the use of green data centers powered by 100% renewable energy sources such as solar, wind, and hydroelectric power
2. Promoting energy-efficient AI algorithms that reduce computational power requirements, which in turn will lower energy consumption during the training and operation of machine learning models
3. Implementing water conservation measures to minimize the use of water in cooling data centers, including the development of closed-loop cooling systems and the use of AI-based systems to optimize energy and water management.