personally identifiable information
In plain wordsInformation that can identify a specific person, such as a name, address, phone number or ID number — or several facts that together point to one person.
Information that identifies a particular person, either on its own or once linked with other data; US federal guidance defines it as information that 'can be used to distinguish or trace an individual's identity'. Direct identifiers include names, government ID numbers, passport and account numbers, email addresses and biometric records; quasi-identifiers such as birth date, postcode and gender are not unique on their own but can single people out in combination — in a classic finding, about 87% of the 1990 US population was unique on ZIP code, gender and full birth date. Web-scale training datatraining data, n.: The examples an AI system learns from. Whatever patterns, skills, gaps and biases are in the training data tend to show up in the finished system. inevitably contains PII, and models can memorizememorization, n.: When an AI model stores specific pieces of its training data — a paragraph, a phone number, an image — closely enough to reproduce them, instead of only learning general patterns. and reproduce it, as training data extractiontraining data extraction, n.: Getting an AI model to repeat, word for word, pieces of the data it was trained on, which may include someone's private contact details or copyrighted text. attacks have shown. Data pipelines therefore detect and redact or mask it, using pattern matching for emails, phone numbers and IP addresses and named-entity recognitionnamed-entity recognition, n.: Software that spots names and similar expressions in text and labels each by type: in 'Maria flew from Lagos to Paris on Monday', it marks a person, two cities and a date. for names, alongside opt-outs, machine unlearningmachine unlearning, n.: Erasing specific things a trained AI model has learned, such as dangerous lab techniques, a copyrighted novel or one person's records, far more cheaply than training the model all over again. and differential privacydifferential privacy, n.: A mathematical promise that the results of analyzing a dataset — or a model trained on it — would look almost the same whether or not any one person's data was included, so little can be learned about that person.; no method is complete.
The PII scrubber replaced every email address in the crawl with a placeholder.
Origin
A US privacy term, defined for federal agencies in Office of Management and Budget memorandum M-07-16 (2007) and NIST Special Publication 800-122 (2010); the current federal definition is in OMB Circular A-130 (2016), and M-07-16 was replaced by M-17-12 (2017). Latanya Sweeney's research of the late 1990s showed that combinations of innocuous attributes can re-identify people, undermining naive anonymization.
Usage note
PII has no single legal definition. EU law instead uses 'personal data' (GDPR, Art. 4(1)), which is broader: any information relating to an identified or identifiable person, including online identifiers. Pseudonymized data remains personal data under the GDPR.
Related
- training data extractiontraining data extraction, n.: Getting an AI model to repeat, word for word, pieces of the data it was trained on, which may include someone's private contact details or copyrighted text.
- memorizationmemorization, n.: When an AI model stores specific pieces of its training data — a paragraph, a phone number, an image — closely enough to reproduce them, instead of only learning general patterns.
- differential privacydifferential privacy, n.: A mathematical promise that the results of analyzing a dataset — or a model trained on it — would look almost the same whether or not any one person's data was included, so little can be learned about that person.
- data cleaningdata cleaning, n.: Finding and fixing errors in a dataset — typos, duplicates, missing values, impossible numbers, wrong labels — before using it, because flawed data produces flawed results.
- machine unlearningmachine unlearning, n.: Erasing specific things a trained AI model has learned, such as dangerous lab techniques, a copyrighted novel or one person's records, far more cheaply than training the model all over again.
- data pipelinedata pipeline, n.: The chain of automated steps that collects raw data, cleans and transforms it, and delivers it in a usable form — the factory line that turns messy inputs into training material.
Sources
- McCallister, E., Grance, T. & Scarfone, K. (2010). Guide to Protecting the Confidentiality of Personally Identifiable Information (PII). NIST Special Publication 800-122.
- Sweeney, L. (2000). Simple Demographics Often Identify People Uniquely. Carnegie Mellon University, Data Privacy Working Paper 3.
Defined in law or standards
OMB Circular A-130 (2016); NIST SP 800-122 (2010). EU counterpart: 'personal data', GDPR (Reg. (EU) 2016/679), Art. 4(1)
Mentioned in
agent memoryagent memory, n.: The ways an AI assistant or agent keeps information — your preferences, past conversations, lessons from earlier tasks — and brings it back later, beyond what fits in a single conversation., guardrailsguardrails, n.: Safety checks built around an AI system — filters on what goes in and comes out, and limits on what it may do — to keep it from producing harmful or off-topic content or leaking data., privacyprivacy, n.: Your ability to control who knows what about you. AI raises new privacy questions because it learns from huge amounts of personal data and can infer things about people they never shared., data pipelinedata pipeline, n.: The chain of automated steps that collects raw data, cleans and transforms it, and delivers it in a usable form — the factory line that turns messy inputs into training material., data protectiondata protection, n.: Laws and practices that control how organizations collect, use, store and share information about people, and that give people rights over it, such as seeing it or having it deleted., 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., named-entity recognitionnamed-entity recognition, n.: Software that spots names and similar expressions in text and labels each by type: in 'Maria flew from Lagos to Paris on Monday', it marks a person, two cities and a date., privacy-preserving machine learningprivacy-preserving machine learning, n.: The family of techniques for training and using AI models while keeping the underlying personal or confidential data hidden — from adding careful noise to computing on encrypted data., web scrapingweb scraping, n.: Using software to automatically copy information from websites — text, prices, images — in bulk, which is how much of the data used to train AI was gathered., contextual integritycontextual integrity, n.: A way of thinking about privacy: information is not simply public or private; what matters is whether it flows in ways that fit the expectations of the setting where it was shared., 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., k-anonymityk-anonymity, n.: A rule for anonymizing tables of personal data: blur details such as age and postcode until every person matches at least k others, so no one can be singled out.