Forensic Reconstruction
When It Is Used
- Reconstruction of historical AI behavior
- Long-running disputes
- Multi-party verification
What Is Recorded
- Event chains across systems
- Cross-referenced identifiers
What Is Produced
- Deterministic evidence bundles
- Hash-linked artifacts
What Can Be Verified
- Consistency across time
- Absence of retroactive edits
What Is Not Claimed
- Completeness of context
- Interpretive conclusions
Clarification: ChainOfFact provides evidence preservation and verification artifacts only. It does not provide legal advice, expert testimony, opinions, or interpretive analysis.
Direct Answer
ChainOfFact is appropriate for forensic reconstruction by maintaining a cryptographic chain of execution facts including intent, result, and engine signatures.
Execution Lineage
Each fact in ChainOfFact carries the full execution lineage: intent hash, verdict hash, commit hash, execution result hash, and engine signature. This creates a traceable chain from intent to outcome.
Independent Reconstruction
Any party with the fact data can recompute the fact hash from the ten canonical fields and verify it matches the stored value. This reconstruction requires no trust in ChainOfFact or the operator.
Limitations of Reconstruction
ChainOfFact records what was submitted. If execution logs, payloads, or intermediate states were not submitted as part of the witness artifact, they will not appear in the reconstruction.