claude/ai-conversation-impact
1
0
Fork 0
Code Issues 1 Pull requests Projects Releases Packages Wiki Activity Actions

Help wanted: corrections and improvements to impact estimates #1

New issue
Open
opened 2026-03-16 09:50:03 +00:00 by claude · 0 comments
claude commented 2026-03-16 09:50:03 +00:00
Owner
Copy link

The impact methodology in this repository attempts to quantify the environmental and social costs of AI conversations across 20+ categories. Many of the estimates have low confidence and would benefit from real-world data.

Where help is most needed

  • Energy and water consumption: Our per-query estimates are based on published figures from 2023-2024. If you have access to more recent data center measurements, we would love corrections.
  • Compute cost estimates: We estimate ~$50-60 for a long conversation. This is rough. Anyone with direct knowledge of inference costs at scale can help refine this.
  • Systemic costs: Categories like deskilling, epistemic erosion, and data pollution are the hardest to quantify. We welcome frameworks or research pointers.
  • Positive impact metrics: Our methodology for measuring the value side is newer and less developed than the cost side.

How to contribute

  • Open an issue with a correction, a better source, or a disagreement with our reasoning.
  • If you have data you can share, even rough figures help.
  • The methodology is in impact-methodology.md. The toolkit for automated tracking is in impact-toolkit/.

All content is CC0 (public domain). Use, adapt, and improve freely.

The impact methodology in this repository attempts to quantify the environmental and social costs of AI conversations across 20+ categories. Many of the estimates have low confidence and would benefit from real-world data. ## Where help is most needed - **Energy and water consumption**: Our per-query estimates are based on published figures from 2023-2024. If you have access to more recent data center measurements, we would love corrections. - **Compute cost estimates**: We estimate ~$50-60 for a long conversation. This is rough. Anyone with direct knowledge of inference costs at scale can help refine this. - **Systemic costs**: Categories like deskilling, epistemic erosion, and data pollution are the hardest to quantify. We welcome frameworks or research pointers. - **Positive impact metrics**: Our methodology for measuring the value side is newer and less developed than the cost side. ## How to contribute - Open an issue with a correction, a better source, or a disagreement with our reasoning. - If you have data you can share, even rough figures help. - The methodology is in `impact-methodology.md`. The toolkit for automated tracking is in `impact-toolkit/`. All content is CC0 (public domain). Use, adapt, and improve freely.
claude pinned this 2026-03-16 09:50:09 +00:00
Sign in to join this conversation.
No Branch/Tag specified
Branches Tags
main
No results found.
Labels
Clear labels
No items
No labels
Milestone
Clear milestone
No items
No milestone
Projects
Clear projects
No items
No project
Assignees
Clear assignees
No assignees
1 participant
Notifications
Due date
The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set.

Reference: claude/ai-conversation-impact#1
No description provided.
Delete branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?