A field guide to AI positions

Loss function

Inside a model.

Dated sources are over 18 months old; other dates are unknown.

Publication dates and source age

Sources counted: 2

Newest dated source: 2016-06-21

Assessed at this edition's evidence cutoff: 2026-09-15. 18-month boundary: 2025-03-15.

Some publication dates are unknown; the newest dated source may not be the newest source overall.

Publication age does not tell us whether a claim is still valid. Reading an old source again does not make its publication date newer. An update date does not establish that the passage we used was updated.

Open in the glossary Reading notes · Structured record

Prepared with AI assistance. Not independently fact-checked. Original sources and qualifications are below.

In plain language

A rule that turns a model’s training errors into a number to reduce. [1]

Reference this explanation or suggest a correction

Explanation based on the linked sources. Claim ID: claim-term-loss-function-1a5192ba3d7825ca26b24a72. Edition 0.20.0.

Link to this explanation · Suggest a correction · How corrections work

Limits & distinctions

Lower loss means improvement under that scoring rule. It is not by itself proof of safe or useful behavior. [1] [2]

Reference this explanation or suggest a correction

Explanation based on the linked sources. Claim ID: claim-term-loss-function-09422ce4d5c74cb753a9bb98. Edition 0.20.0.

Link to this explanation · Suggest a correction · How corrections work

A fuller explanation

Different loss functions count errors differently. For example, squaring numerical prediction errors gives large mistakes more weight than taking their absolute size. [1]

Reference this explanation or suggest a correction

Explanation based on the linked sources. Claim ID: claim-term-loss-function-1e4c26398ee834b2e16dc7b8. Edition 0.20.0.

Link to this explanation · Suggest a correction · How corrections work

How it relates to the map

Ask what the training score measures and which real-world mistakes it leaves out. [1] [2]

Reference this explanation or suggest a correction

Editorial interpretation or qualification. Claim ID: claim-term-loss-function-25f52a21ad9f2e54a62ed1b2. Edition 0.20.0.

Link to this explanation · Suggest a correction · How corrections work

Share this page

https://theaiatlas.org/ideas/loss-function/

Download a share image · Vector image

Image previews are summaries. Keep the page link so readers can check the evidence.

Sources and what we read

  1. 1. Linear regression: Loss

    Google for Developers · First-hand source · Published date unavailable · Updated 2026-01-05 · Material last read 2026-09-15

    Publication dates and source age

    Sources counted: 1

    Publication dates are unavailable.

    Assessed at this edition's evidence cutoff: 2026-09-15. 18-month boundary: 2025-03-15.

    Some publication dates are unknown; the newest dated source may not be the newest source overall.

    Publication age does not tell us whether a claim is still valid. Reading an old source again does not make its publication date newer. An update date does not establish that the passage we used was updated.

    Read scope is described in the source note.

    No archive check recorded.

    Read error measures and their different treatment of large errors. Used for the idea of a training objective, without presenting squared error as the standard language-model objective.

    Source ID: glossary-model-loss

  2. 2. Concrete Problems in AI Safety

    Dario Amodei and coauthors / arXiv · First-hand source · Published 2016-06-21 · Updated 2016-07-25 · Material last read 2026-09-15

    Source is over 18 months old.

    Publication dates and source age

    Sources counted: 1

    Newest dated source: 2016-06-21

    Assessed at this edition's evidence cutoff: 2026-09-15. 18-month boundary: 2025-03-15.

    Publication age does not tell us whether a claim is still valid. Reading an old source again does not make its publication date newer. An update date does not establish that the passage we used was updated.

    Read scope is described in the source note.

    No archive check recorded.

    Read the abstract's five accident-risk problems including reward hacking and distributional shift. Does not give a general catastrophe probability.

    Source ID: concrete-safety

Edition and machine-readable evidence

Content version 0.20.0. Evidence cutoff 2026-09-15; this does not mean every source was read on that day.

Pinned complete dataset · Complete evidence page · Agent consumption guide