SECAI Core · phase 12 of 15
Data anonymization
The process of removing personally identifiable information from data sets, so that the individuals the data describes remain anonymous.
The Explain card
- Plain English
- Data anonymization strips personally identifiable information out of a data set so the people it describes cannot be identified, while keeping the data useful.
- Example
- Before using customer support transcripts to fine-tune a chatbot, a team removes names, email addresses, phone numbers and account IDs, and generalises exact locations to regions.
- Why it matters
- Models memorise. If personal data goes into training, it may come back out in a response to a stranger. Anonymizing upstream shrinks that risk and eases privacy obligations, though re-identification from combined fields remains a concern.
- Hook
- Take the name off the jersey, keep the stats.
Where it sits in the deck
Phase 12: Defensive Technologies and Secure Development Practices
Map defences directly to the attacks just catalogued — the technical controls, secure coding practices, and protective tools that harden AI and traditional systems alike.
