# Preprint tests reversible personal-data substitution for LLM prompts

_Published Monday, October 19, 2026 at 9:33 AM EDT · Science, AI · Latest · Tier 2 — Notable_

Researchers report in a preprint that Veil-then-Reveal replaces personal information on a user's device before sending prompts to remote language models, then restores original values in responses. Tests on three datasets containing 11,322 annotated personal-information instances achieved 97.9% detection recall and 95.2% protection of detected values.

The framework substitutes realistic values of the same type, using encryption that preserves formats for structured information and a keyed dictionary and secure vault for names, locations and organizations. Session-specific mappings prevent links across sessions. The researchers report 100% restoration of echoed substitutes, but the detection and protection results leave gaps in coverage.

## Sources

- [cs.CR updates on arXiv.org](https://arxiv.org/abs/2610.22800)

---
Canonical: https://techandbusiness.org/newswire/OaunO2fBAttf7X12R6tX3b
Published: 2026-10-19T13:33:22.998Z
Story chronology: 2026-10-19T04:00:00.000Z
Retrieved: 2026-10-19T16:13:31.668Z
Publisher: Tech & Business (techandbusiness.org)