DadgogoAI Dictionary

Dictionary entry

membership inference attack

noun \ˈmem-bər-ˌship ˈin-f(ə-)rən(t)s ə-ˈtak\

also membership inference or MIA

TopicSafety & Ethics First known use2016 Senses1
Advertisement

Definition of membership inference attack

  1. : a privacy attack that tries to determine whether a particular record, person, document, or example was included in a machine-learning model's training data

    • The audit used a membership inference attack to estimate whether patient records could be detected in the trained classifier.
    • High confidence on familiar examples made the overfit model more vulnerable to membership inference.

Origin & history

first known use2016

Reza Shokri and collaborators formalized and experimentally evaluated membership inference against machine-learning-as-a-service models in work released in 2016 and presented at the 2017 IEEE Symposium on Security and Privacy. The attack family later expanded to generative and foundation models.

Test yourself

Which of these is the meaning of membership inference attack?

Related terms

training data the collection of examples (text, images, code, audio, and so on) that a model learns from data poisoning an attack that inserts, changes, or selectively removes examples in a machine-learning… model card a document published alongside a model describing what it is for, how it was trained and… confidential AI the use of confidential-computing technologies to protect AI models, data, and computations… data residency a requirement or arrangement specifying the geographic region in which data is stored… black box a system whose inputs and outputs can be seen but whose internal reasoning cannot easily be…

Cite this entry

"Membership inference attack." AI Dictionary, Dadgogo, https://dadgogo.com/dictionary/membership-inference-attack/. Accessed 8 Oct. 2026.

Dictionary entries near membership inference attack

Advertisement