Desafíos actuales de la Inteligencia Artificial

Mapping the influence of artificial intelligence 205 operations. Technologies can drive economic growth and create new business opportunities, but they can also disrupt existing industries and exacerbate economic inequalities. In AI, this dimension assesses how AI technologies affect economic activities and business models. AI-driven automation can boost productivity and economic efficiency, but might also lead to job losses and economic disparity if not managed properly. Technical Dimension : This dimension addresses the maintenance and evolution of artificial systems over time, emphasizing resilience, maintenance, and ease of system transi- tions. Technologies should be designed for longevity, adaptability, and reliability, considering technical challenges and risks. For AI, this dimension evaluates the sustainability of AI systems in terms of their technical robustness, scalability, and adaptability. AI systems need to be resilient to failures, capable of evolving with new data, and easy to maintain to ensure long-term functionality and reliability. Environmental Dimension : This dimension evaluates the use and stewardship of natural resources, covering immediate waste production, energy consumption, local ecosys- tems, and climate change concerns. Technologies can help reduce environmental impact through efficiency and innovation, but they can also lead to increased resource consumption and environmental degradation if not managed sustainably. Regarding AI, this dimension looks at the environmental footprint of AI technologies. AI development and deployment consume significant computational resources, leading to high energy usage and associated carbon emissions. Sustainable AI practices aim to minimize this impact through energy-efficient algorithms and environmentally conscious data center management. 3.1.2. Temporal Levels of Impact The three temporal levels represented in the SusAD—immediate, enabling, and structur- al effects—help visualize the short-term, medium-term, and long-term impacts of technol- ogies. This temporal distinction is crucial for understanding how effects evolve and interact over time. Immediate Effects : These are direct functions of the system or direct effects of its development, observable shortly after implementation. Immediate effects of AI include im- provements in efficiency and decision-making in various applications, such as automated cus- tomer service and real-time data analysis. Enabling Effects : These arise from the use of the system and may become apparent over a medium timeframe, influencing broader applications and user behaviors. Enabling effects of AI might include increased accessibility to services through AI-powered tools and enhanced capabilities in predictive analytics that support proactive decision-making. Structural Effects : These refer to persistent changes that can be observed at the mac- ro-level over a long period, often altering fundamental aspects of society and environment.