AI Governance Engineer
GitHub

Evals

By Updated

Automated tests of a model's or agent's behaviour (capability, safety and adversarial), run as controls, not as one-off research.

Developed in
ch. 04, Layer 03: Evals & Red Teaming as Evidence
Chapters
ch. 04, The Stack
Source
Defined by this Body of Knowledge: the section it is developed in is the source

Where it is used

20 chapters of the Body of Knowledge use the term. Each link opens the first section that does.

Patterns that use this term

10 pattern pages use the term, most mentions first.

  • Reinforcement learning

    Learning to maximise a reward signal through trial and feedback.

  • Machine unlearning

    Techniques that remove a training record's influence from a model without full retraining.

  • Open-weight model

    A model whose trained weights are published for download under a licence that may be permissive, copyleft, use-restricted or custom.

  • Multimodal model

    A model that takes or produces more than one modality (text, image, audio, video).

  • Prompt injection

    An input that alters a model's behaviour or output in ways its designers did not intend.

  • Model anonymity

    The EDPB's test for when a trained model falls outside the GDPR: both direct extraction of training subjects' data and obtaining it through queries must be…

  • Generative AI

    AI that outputs new content (text, images, audio, video, code) rather than an estimate about something that exists.

  • Regurgitation

    A model reproducing memorised training data verbatim, including personal data, whether prompted deliberately (training-data extraction) or not.

Sources

No external source: the term is coined or used in a specific sense by this Body of Knowledge, and the section linked under "Developed in" is its source.

Definitions of legal terms paraphrase the cited text, which governs. Dated statements are as of .

Cite this term

García Aibar, J. (2026). Evals. In AI Governance Engineering: The Thesis & Body of Knowledge (v0.5.0), Glossary. https://doi.org/10.5281/zenodo.22956197. https://aigovernanceengineer.com/glossary/evals. CC BY 4.0

BibTeX

@misc{aige2026evals,
  author  = {Jorge García Aibar},
  title   = {{Evals}},
  note    = {Glossary, AI Governance Engineering: The Thesis \& Body of Knowledge, version 0.5.0},
  year    = {2026},
  doi     = {10.5281/zenodo.22956197},
  url     = {https://aigovernanceengineer.com/glossary/evals}
}