model card
In plain wordsA short document published with an AI model explaining what it is for, how it was trained and tested, how well it works for different groups of people, and its known limits.
A standardized disclosure published alongside a trained modelmodel, n.: A simplified working version of something. In AI it usually means the trained program, a big set of learned numbers, that takes an input and produces an answer, prediction or piece of content. in which its developers set out what it is meant for and what it should not be used for; who built it and how (version, architecture, license); which data trained and evaluated it; how well it performs within relevant subgroups, such as demographic groups, environments or instruments; and the ethical issues, caveats and recommendations that apply. The disaggregated evaluation is the distinctive element: an aggregate accuracy figure can hide large failures for subgroups, as audits of face-analysis systems showed. Model cards became the de facto way to publish models: on Hugging FaceHugging Face, prop. n.: A company whose website is the main public library for sharing AI models and datasets — often called the GitHub of machine learning — and which makes widely used free AI software. a repository's README, with machine-readable metadata, serves as its model card (though many repositories have only a stub or none), and frontier developers publish longer system cardssystem card, n.: A public report released with an AI model that explains what it can do, how it was tested for safety, what risks were found and what protections were put in place. covering safety evaluations. Quality varies widely and many cards omit training-data details or evaluations, a gap regulators address through mandatory documentation for general-purpose AI modelsgeneral-purpose AI model, n.: The EU's legal name for the large, adaptable AI models behind chatbots and similar tools: models that can turn their hand to many different jobs and be built into many other products..
The model card says the detector was evaluated only on English text, so don't use it on Hindi.
Origin
Proposed by Margaret Mitchell, Simone Wu, Andrew Zaldivar, Parker Barnes, Lucy Vasserman, Ben Hutchinson, Elena Spitzer, Inioluwa Deborah Raji and Timnit Gebru in 'Model Cards for Model Reporting' (arXiv, October 2018; FAT* 2019), with example cards for a smiling-face detector and a toxic-comment classifier.
Usage note
A model card documents a model; a datasheet for datasetsdatasheet for datasets, n.: A standard document that travels with a dataset, describing why it was made, what is in it, how it was collected and labeled, and what it should and should not be used for. documents data; a system cardsystem card, n.: A public report released with an AI model that explains what it can do, how it was tested for safety, what risks were found and what protections were put in place. documents a deployed system, including safeguards and red-teaming results. The terms are sometimes used loosely for one another.
Related
- datasheet for datasetsdatasheet for datasets, n.: A standard document that travels with a dataset, describing why it was made, what is in it, how it was collected and labeled, and what it should and should not be used for.
- system cardsystem card, n.: A public report released with an AI model that explains what it can do, how it was tested for safety, what risks were found and what protections were put in place.
- AI transparencyAI transparency, n.: Openness about AI: telling people when they are dealing with one, how a system was built and tested, and why it reached a particular decision.
- Hugging FaceHugging Face, prop. n.: A company whose website is the main public library for sharing AI models and datasets — often called the GitHub of machine learning — and which makes widely used free AI software.
- data provenancedata provenance, n.: The documented history of a piece of data: where it came from, who created and changed it, under what permissions, and how it was processed before reaching an AI system.
- evaluationevaluation, n.: Testing an AI system to find out what it can and cannot do, how well, and how safely — from exam-style quizzes to hands-on trials and expert reviews.
Sources
- Mitchell, M. et al. (2019). Model Cards for Model Reporting. Proceedings of FAT* 2019. arXiv:1810.03993. arxiv.org/abs/1810.03993
Mentioned in
Hugging FaceHugging Face, prop. n.: A company whose website is the main public library for sharing AI models and datasets — often called the GitHub of machine learning — and which makes widely used free AI software., AI modelAI model, n.: The trained core of an AI: a large file of learned numbers plus the recipe for using them, which turns an input such as a question or a photo into an output., AI transparencyAI transparency, n.: Openness about AI: telling people when they are dealing with one, how a system was built and tested, and why it reached a particular decision., algorithmic accountabilityalgorithmic accountability, n.: The idea that the people and organizations behind automated decision systems must be able to explain and justify those systems' effects, answer for the harm they cause, and fix it., data provenancedata provenance, n.: The documented history of a piece of data: where it came from, who created and changed it, under what permissions, and how it was processed before reaching an AI system., model registrymodel registry, n.: A central catalog where a team stores every version of its trained models, with records of how each was made and which version is approved for use., model risk managementmodel risk management, n.: The rules banks and other firms follow to make sure the mathematical models and AI systems they rely on are built correctly, checked independently and used only for what they are fit for., reproducibilityreproducibility, n.: Whether other people, or the same people later, can get the same results by rerunning an AI experiment. Many published AI results have proved hard to reproduce, which makes real progress harder to judge., responsible AIresponsible AI, n.: An umbrella name for the practices organizations use to keep their AI systems fair, safe, private and accountable, from the first design meeting to everyday use., system cardsystem card, n.: A public report released with an AI model that explains what it can do, how it was tested for safety, what risks were found and what protections were put in place., Timnit GebruTimnit Gebru, prop. n.: An Ethiopian-born American computer scientist who exposed racial and gender bias in AI systems, co-wrote the 'stochastic parrots' paper, and founded an independent AI research institute., woke AIwoke AI, n.: A politically charged label, used mainly by US conservatives, for AI systems seen as building progressive views into their answers. Since 2025 it also names a US rule on which AI the government may buy., datasheet for datasetsdatasheet for datasets, n.: A standard document that travels with a dataset, describing why it was made, what is in it, how it was collected and labeled, and what it should and should not be used for., FAccTFAccT, prop. n.: A yearly academic conference on the fairness, accountability and transparency of computer systems, where much influential research on AI bias and ethics has been published., Margaret MitchellMargaret Mitchell, prop. n.: An American computer scientist who proposed 'model cards' for documenting AI systems, co-led Google's Ethical AI team, and is chief ethics scientist at Hugging Face., model hubmodel hub, n.: An online library where trained AI models are published so that others can find, download and reuse them instead of training their own from scratch.