Eco Artificial Intelligence
0 likes33 views
The document discusses the intersection of artificial intelligence (AI) and environmental sustainability, highlighting strategies to reduce the carbon footprint and energy consumption associated with AI operations. It details the benefits and challenges of sustainable AI practices, including increased energy efficiency and better governance, while pointing out the lack of policies and awareness promoting these strategies. Additionally, it provides statistics on AI-related emissions and the efforts of global companies to leverage AI in achieving sustainability goals.
Eco Artificial Intelligence
- 2. ENVIRONMENTAL SUSTAINABILITY SLIDESHOWS FROM s2adesign.com S2ADESIGN.COM
- 4. A Branch of Computer Science that Focuses on Developing Intelligent Machines Capable of Performing Tasks Requiring Human Intelligence.
- 5. Strategies to Reduce the Carbon Footprint & Energy Consumption Associated with AI Operations.
- 6. Improve energy efficiency Monitorcompliance reduce water consumption Benchmark efforts Improve social impact Improve DATA governance risk MANAGEMENT Reduce emissions MCKINSEY.COM SUSTAINABLE ai goals
- 8. EFFICIENCY Driving More Efficient Business Processes and Daily Operations. transparency Increasing Supply Chain Transparency Towards More Responsible Sourcing and Labor Practices. AUTOMATION Automating the Collection and Reporting of Data Across Complex Operations and Endpoints. support Supporting Corporate Diversity Equity, and Inclusion (DEI) Goals such as AI-Enabled Recruiting Practices. IBM.COM SUSTAINABLE ai STRATEGIES
- 9. SUSTAINABLE ai ECONOMY Saves on Energy and Resources. Less Waste Generated Means Less Waste Management Costs. SOCIAL Enterprise Social Image Improvement: Many Customers Prefer to Buy from Businesses with Sustainable Practices. ENVIRONMENT Minimizes Negative Environmental Impacts, Carbon Footprint, and GHG Emissions. benefits THEDIGITALSPEAKER.COM
- 10. Over the Past Decade, the Amount of Data Used to Train AI Systems has Significantly Increased, Consuming a Large Amount of Energy. Energy use The Growth of Data has Resulted in a Rapid Increase in the Size of Computational Models, Consuming a Large Amount of Energy. Model size The Growing Use of AI is Leading to an Increase in the Carbon Footprint such as Emissions & Energy Consumption. Ai expansion Lack of Policies & Regulations Supporting Sustainable AI Strategies. policies Lack of Awareness & Education Promoting Sustainable AI Practices. awareness SUSTAINABLE ai CHALLENGES TECHOPEDIA.COM
- 11. Prioritizing Data Quality over Quantity can Significantly Reduce Energy Use: Using High-Quality Data & Avoiding Data Duplication. Energy use Using Smaller Models Eliminates the Need for Powerful Computers. Techniques Include Model Compression & Knowledge Distillation. Model size Developing Energy-Efficient Hardware Specifically for AI Tasks. Using Renewable Energy Sources Like Solar Power for Data Centers. Ai expansion Implementing Policies to Support Sustainable AI Operations as well as Research & Development. Rewarding Energy-Efficient AI Systems. policies Launching Programs to Teach AI Professionals about Energy-Saving Practices. Raising Awareness of Individuals and Organizations. awareness SUSTAINABLE ai solutions TECHOPEDIA.COM
- 13. In 2021, Under UNESCO's Auspices the International Research Center on Artificial Intelligence (IRCAI) was Established to Advance Research on the Use of Artificial Intelligence to Achieve the SDGs by Showcasing Relevant Projects Worldwide. IRCAI.ORG UNESCO Center in AI AI for Sustainability
- 14. This is a placeholder text. This text can be replaced with your own text. This text can be replaced with your own text. This is a placeholder text. This text can be replaced with your own text. This text can be replaced with your own text. 9.7 6.6 4.1 3.6
- 16. tons – CO2 Equivalent AI CAR HUMAN LIFE AIINDEX.STANFORD.EDU ANNUAL AI EMISSIONS comparison IN 2023
- 17. 502 352 76 25 GPT-3 GROPHER OPT BLOOM EMISSIONS FROM AI based LLMs IN 2023 tons – CO2 Equivalent ARXIV.ORG
- 18. 1,287 1,066 433 324 GPT-3 GROPHER BLOOM OPT ENERGY CONSUMPTION FROM AI based LLMS IN 2023 megawatt hours ARXIV.ORG
- 19. Ai environmental impact worldwide in 2023 502Tons Global Emissions from AI Based GPT-3. The Largest by LLM. 1.3GWH Energy Consumption of AI Based GPT-3. The Largest by LLM. 955Tons Global Emissions from AI Based Large Language Models (LLMs). Key highlights ARXIV.ORG
- 21. 53 47 46 36 31 Developing Markets Asia-Pacific Greater China Europe North America companies takingsteps to reduce AI emissions percent (2022) MCKINSEY.COM
- 22. Share initiative 62% Improving the environmental impact (e.g. energy efficiency, emissions) 51% Evaluating sustainabilityefforts(e.g. benchmarking) 45% Improving governance(e.g. regulatory compliance,risk management) 34% Improving the socialimpact (e.g. sourcingethical products) Sustainability efforts using ai worldwide Share of companies(2022) MCKINSEY.COM
- 23. 61 54 44 39 30 Greater China Asia-Pacific Developing Markets Europe North America MCKINSEY.COM Sustainability efforts using ai worldwide Share of companies(2022)
- 24. 62% Share of Companies Using AI to Reduce their Environmental Impact Worldwide. 61% Share of Companies in the Greater China Region Using AI in their Sustainability Efforts. 53% Share of Companies in the Developing Market Region Taking Steps to Reduce AI Emissions. Key highlights MCKINSEY.COM Sustainability efforts using ai worldwide in 2022
- 25. S2ADesign.com