To AI or not to AI? Per-task energy and water demands of AI inference are orders of magnitude lower than human labor across all scenarios

Abstract excerpt

<title>Abstract</title> <p>The aggregate resource footprint of artificial intelligence infrastructure is growing rapidly, but per-unit comparisons against the human labor that AI augments or replaces have not been established. We construct a sector-decomposed lifecycle accounting framework for a representative US adult, covering transportation, aviation, residential energy, food systems, manufactured goods, and w...

Topics

Open a Topic to create a Post that cites this publication.

• Artificial Intelligence
• Decision Support Systems, Clinical
• Diagnosis, Differential
• Electronic Health Records
• Genetic Variation
• Genetics, Bioinformatics, and Biomedical Research
• Phenotype
• Rare Diseases

Identifiers and source

Literature Corpus work

2f3066b9-499b-5572-90ad-ec29e210d8dc

DOI

10.21203/rs.3.rs-9962169/v1