The environmental impact of AI sits at the center of many campus conversations. You hear claims about energy collapse on one side and confident assurances about green innovation on the other. Between those poles, you often struggle to find clear, usable evidence.

You likely want to form your own position on AI related carbon emissions, water consumption, and full lifecycle costs. You also know how hard this becomes when information appears as isolated statistics, media headlines, or vendor statements. This gap creates a clear role for universities and libraries.

Librarians and research support professionals work with evidence every day. Their task is not to promote fear or optimism. Their task is to support informed judgment. You gain value when someone helps you trace claims back to peer reviewed studies, technical reports, and transparent methods.

Leo Lo https://www.linkedin.com/posts/leoslo_ais-environmental-impact-understanding-activity-7426267318315352064-VF8l
highlights this responsibility in his recent reflection on AI and sustainability. He points to a practical response.
Nicole Hennig has compiled a living document https://docs.google.com/document/d/1WFWL8oPLXwLetQ4-urYKzskHtVO1pPTbAMSDiS4QADY/edit?tab=t.0 
that brings together research on the environmental footprint of AI systems. The collection covers topics such as:

• energy use in model training and inference
• water consumption in data centers
• carbon accounting challenges
• hardware supply chains and lifecycle impacts
• limits and gaps in current measurement methods

You can use this document as an entry point for deeper reading. It helps you move from opinion to evidence. It also shows where uncertainty remains, which matters for teaching and research.

For faculty, this material supports course discussions on digital responsibility, research ethics, and technology assessment. For students, it provides sources you can cite, question, and compare. You gain a stronger position when you base arguments on documented methods rather than simplified narratives.

Knowledge on this topic grows through shared work. If you teach, research, or study AI, sustainability, computing, or related fields, you likely hold relevant sources, data, or case studies. Sharing these strengthens the collective understanding.

You are invited to contribute in three ways:

• share literature or reports you already use
• discuss how you address AI environmental costs in teaching or research
• raise questions that need further investigation

The Science and Research Institute welcomes this exchange. Contact the institute if you want to continue the conversation, propose a discussion session, or collaborate on curated resources.
Informed debate starts with shared evidence.

Follow us on social media: https://www.linkedin.com/feed/update/urn:li:activity:7426653658592993281

09 фев 2026