Change Provenance

Cryptographically binds the identity of the modifier to each specific edit. This moves beyond simple 'Track Changes' usernames to include digital signatures and public key infrastructure (PKI) verification.

• Records the exact user ID, role, and authentication method
• Prevents repudiation of modifications in high-stakes negotiations
• Enables granular blame annotation at the character level, not just the paragraph level
• Example: Proving that a specific liability cap was altered by opposing counsel, not your team

Establishes a strict happened-before relationship between every edit in a negotiation session. This is not just a timestamp; it is a directed acyclic graph (DAG) of edit operations.

• Uses vector clocks or Lamport timestamps to resolve concurrent edits
• Distinguishes between causal dependency and simple chronological coincidence
• Critical for reconstructing the exact sequence of a multi-party negotiation
• Enables rollback to any specific point in the negotiation history without data loss

Stores the semantic rationale behind a change, not just the textual delta. This bridges the gap between a raw diff and the negotiating party's strategic objective.

• Links edits to specific comments, emails, or negotiation playbook rules
• Classifies intent using taxonomies: Risk Mitigation, Clarification, Commercial Concession
• Enables downstream AI to learn negotiation patterns from historical intent data
• Provides the 'why' required for advanced obligation change detection audits

Treats the document's edit history as a Merkle tree or blockchain-like structure. Each modification is hashed and linked to the previous state, creating a tamper-evident seal.

• Detects unauthorized insertions or deletions in the audit log itself
• Integrates with clause-level hashing to verify individual clause integrity
• Provides mathematical certainty that the final 'redline' is a true representation of all agreed changes
• Essential for regulatory compliance under SEC Rule 17a-4 or similar recordkeeping mandates

Records the specific software agent, algorithm, or API endpoint that executed the modification. This is vital when AI agents autonomously negotiate or redline contracts.

• Distinguishes between human edits, AI-suggested modifications, and automated formatting
• Captures the model version, prompt template, and temperature settings for AI-generated changes
• Enables debugging of conflict resolution algorithms by tracing errors to specific engine versions
• Provides the metadata required for algorithmic explainability audits in automated contracting

Stores not just the text diff, but the extracted semantic state of the document at each version. This allows querying the provenance of legal meaning, not just text strings.

• Links provenance data to the obligation graph diff to track duty modifications
• Records changes to defined term reconciliation mappings across versions
• Enables queries like 'Show me who altered the indemnity cap and when'
• Provides the grounding data for Material Adverse Change (MAC) clause diff audits

A technique that generates a unique, fixed-size cryptographic fingerprint for an individual clause to efficiently detect any modification to its content across document versions. Even a single character change produces an entirely different hash, making tampering immediately evident.

• Uses algorithms like SHA-256
• Enables rapid integrity verification
• Forms the basis for blockchain-based document provenance systems

A concurrency control algorithm that transforms editing operations to ensure eventual consistency across all replicas in a real-time collaborative document editing system. OT preserves the intention of each user's edit while resolving conflicts, making it foundational to platforms like Google Docs.

• Handles concurrent edits without locking
• Maintains a causal ordering of operations
• Each operation carries implicit provenance metadata

A distributed data structure designed so that concurrent, uncoordinated edits from multiple users can be merged mathematically without conflicts. Unlike OT, CRDTs do not require a central server to coordinate transformations, making them ideal for peer-to-peer legal collaboration tools.

• Guarantees strong eventual consistency
• Each edit carries a unique identifier and vector clock
• Powers modern offline-first document editors

A standardized method for recording and visualizing inline modifications, comments, and formatting adjustments within a document. This protocol embeds provenance directly into the document structure, capturing the author identity, timestamp, and nature of change for every insertion and deletion.

• Industry standard in Microsoft Word and Google Docs
• Supports asynchronous collaborative review
• Generates a human-readable audit trail natively

The process of creating a compact, machine-readable file containing only the differences between two documents. A patch serves as a portable provenance artifact—it captures exactly what changed, and when applied to the original, recreates the modified version with full fidelity.

• Standard formats include Unified Diff and JSON Patch (RFC 6902)
• Enables efficient transmission of changes
• Acts as a lightweight, replayable audit record