Resources

This influential paper introduces the concept of Green AI, which encourages AI research that yields better results while consuming less computing power and thus lower environmental impact. The authors contrast Green AI with what they call Red AI: research that seeks to improve accuracy through massive computational power, regardless of the environmental cost. The paper proposes new evaluation criteria for AI research that include computational efficiency alongside accuracy, encouraging more sustainable approaches to AI development.

This foundational paper examines the ethical implications of artificial intelligence development and deployment. The authors present a comprehensive framework for ensuring AI systems are developed and used in ways that benefit humanity. The research addresses key ethical challenges including algorithmic bias, transparency, accountability, and the long-term societal impact of AI systems. The paper proposes concrete guidelines for ethical AI development and governance structures to ensure responsible innovation.

This pioneering study examines the carbon footprint of training natural language processing models. The authors quantify the financial and environmental costs of training various NLP models. The study reveals that training a single BERT model can emit as much CO2 as a trans-Atlantic flight, and that the computational costs of NLP models double every 3-4 months. The authors provide concrete recommendations to reduce environmental impact, particularly by prioritizing energy efficiency in model design and using renewable energy sources for training.

This landmark document provides comprehensive ethical principles and guidelines for computing professionals. It addresses fundamental ethical considerations in computing practice and research. The code establishes clear principles for professional responsibility, including avoiding harm, respecting privacy, ensuring fairness, and promoting sustainable computing practices. This work has become a cornerstone reference for ethical decision-making in computer science and information technology.