Verifiable Redactable Audit Log Using Segmented Hash Chains

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current log verification methods are inadequate in detecting tampering, as attackers can conceal changes by altering both log messages and hashes, and existing solutions fail to maintain verifiability after tampering is detected.

Innovation Solution

The system employs a logging system that uses one-way hashing algorithms, digital signatures from a trusted time server, and ephemeral keys to create a verifiable and redactable log structure, where each log entry includes multiple hashes dependent on previous entries, allowing for segmentation and periodic commitment to prevent tampering and maintain verifiability.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If traditional hash chains are used to verify log entries, then log integrity can be detected, but attackers can still conceal tampering by altering both log messages and their corresponding hashes

Engineering Contradiction:
Improvelog integrity detectionVSAvoidtampering concealment
Core Design Contradiction:
ReliabilityVSObject-affected harmful factors

Solution Approach 1:

The patent segments the log verification process into multiple independent hash chains, where each log entry is associated with multiple hashes from different chains. This segmentation ensures that tampering with one chain does not compromise the entire log, as other chains can still detect the alteration. The segmentation principle directly addresses the vulnerability of traditional single-chain approaches where attackers could coordinate message and hash alterations.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces commitment entries as intermediary elements that act as anchors between log entries and verification points. These commitment entries contain hashes of log entries and are signed by a trusted authority, creating an intermediary layer that prevents direct manipulation of the hash chain. This intermediary mechanism ensures that even if attackers alter log messages and hashes, the signed commitment entries serve as verifiable reference points that expose the tampering.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Reliability

If the entire log is presented for verification, then complete auditability is achieved, but efficiency is reduced due to the large amount of data that must be processed

Engineering Contradiction:
ImproveauditabilityVSAvoidverification efficiency
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The patent extracts only the essential verification elements (commitment entries and their signatures) from the complete log and presents them for verification. Instead of requiring the entire log to be processed, the system extracts the critical anchor points that prove log integrity. This extraction principle allows verifiers to confirm log authenticity by examining only a small subset of committed entries rather than the entire log, dramatically improving verification efficiency while maintaining complete auditability.

Inventive Principle:
Principle #2Taking out (Extraction)

3Reliability

If all log data is retained for verification, then complete verification capability is maintained, but confidentiality is compromised by exposing sensitive information

Engineering Contradiction:
Improveverification capabilityVSAvoidconfidentiality
Core Design Contradiction:
ReliabilityVSLoss of information

Solution Approach 1:

The patent extracts and presents only the verification-critical elements (hashes and signatures) while leaving sensitive log data concealed. The commitment entries contain cryptographic commitments to log entries without exposing the actual log content. This selective extraction maintains verification capability by providing provable integrity anchors while preserving confidentiality by keeping the underlying sensitive information hidden until verification is complete.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent creates cryptographic copies (hashes and commitments) of the log entries that serve as verification substitutes. Instead of presenting the original sensitive log data, the system presents cryptographic copies that contain all necessary verification information but reveal no sensitive content. These cryptographic copies can be verified independently while the original confidential data remains protected, effectively decoupling verification capability from information disclosure.

Inventive Principle:
Principle #26Copying

Data Source

PatentUS11849023B2Verifiable redactable audit log
Publication Date: 2023.12.19 PALANTIR TECHNOLOGIES INC
  • US11849023B2 patent drawing
  • US11849023B2 patent drawing
  • US11849023B2 patent drawing

AI summary

A verifiable, redactable log, which, in some embodiments, may contain multiple hash values per entry in order to sever confidentiality of a log from verifiability. Logs may be verified using recalculation of hashes and verification of trusted digital signatures. In some embodiments, the log may be divided into segments, each signed by a time server or self-signed using a system of ephemeral keys. In some embodiments, log messages regarding specific objects or events may be nested within the log to prevent reporting omission. The logging system may receive events or messages to enter into the log.