Synthesized launches unstructured data masking

London, 12 October 2026. Synthesized today launched unstructured data masking. It finds the personal and confidential data inside documents, emails, chats, call transcripts and free text, and replaces it consistently, under the same policies teams already use for their test databases.

Why it matters

Test, analytics and AI environments rarely hold only tables. They also hold the PDFs, emails, support tickets and notes attached to those records. Column masking does not reach them, so real names, account numbers and health details can end up in lower environments, prompts and vector indexes.

What is new

• Masking for PDF, Word and PowerPoint documents, emails, chat logs and tickets, call transcripts, clinical notes, free-text fields and JSON, XML and log files.
• Consistent replacements across a document set, so the same person gets the same replacement everywhere it appears.
• Synthetic documents for cases where no real file should be used at all.
• Protection for RAG corpora, prompts and agent traces, so confidential data stays out of AI pipelines during development.

Ways to start

• 48-hour proof: see the results on 50 of your own documents in 48 hours.
• Free synthetic sample documents, including bank statements, pay stubs and lab reports, for testing document workflows.
• Guides to the rules that apply to test data: GDPR, HIPAA, PCI DSS and DORA.

Read more about unstructured data masking, or talk to our team.