> For the complete documentation index, see [llms.txt](https://corpus-core.gitbook.io/pap-colibri-stateless/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://corpus-core.gitbook.io/pap-colibri-stateless/the-problem/privacy-rethought.md).

# Privacy Rethought

## The Anatomy of the Privacy Problem

Most blockchain privacy discourse focuses on on-chain data: hiding transaction amounts, concealing wallet balances, or anonymizing contract interactions. This focus is understandable but incomplete. On-chain privacy addresses what is permanently recorded on a public ledger. It leaves unaddressed the information exposed during the process of using blockchain applications.

The privacy gap does not lie in what is recorded on the chain, but in what is observed before any record is made. Blockchain applications query external RPC providers and indexers to evaluate state, simulate outcomes, and prepare decisions — all before a transaction is submitted. This access layer operates outside the blockchain's trust model and carries no cryptographic guarantees of privacy.

This distinction is fundamental. On-chain privacy concerns what is written to a permanent, public record. **Read privacy** concerns what is observed in the process of deciding whether and how to act. The attack surface grows with application sophistication and the value of the decisions being made.

The mechanisms that protect on-chain data — zero-knowledge proofs, confidential transactions, mixing protocols — do not extend to the read layer. An application may conceal the content of a submitted transaction while simultaneously exposing, through its read pattern, the intent that preceded it. A system that protects execution while leaving observation unaddressed provides incomplete privacy guarantees.

## A New Perspective

Privacy should not be viewed as the opposite of transparency, nor as a feature that conflicts with trustlessness. A decentralized system can remain fully verifiable while significantly reducing the information exposed during user interactions.

Achieving this requires a broader understanding of privacy. The objective is not simply to hide transactions or anonymize users, but to control which information becomes observable, when it becomes observable, and whether that information can be exploited.

This perspective shifts privacy from an isolated protocol feature to a system-wide architectural property. Applications, communication channels, verification, and data retrieval all contribute to the overall privacy characteristics of the system.

This whitepaper introduces **Pragmatic Adaptive Privacy (PAP)**, a model that treats privacy as an adaptive, trustless, and application-driven property. Rather than prescribing a single privacy mechanism, PAP provides a structured framework for selecting the appropriate level of privacy for each interaction while preserving verifiability and remaining practical for deployment in real-world Web3 systems.