Home / Blog / Bitemporal AI memory

Bitemporal AI memory preserves what the agent knew then.

Ordinary memory stores the latest version of a fact. Bitemporal memory preserves when the fact was true and when the system learned it, so historical decisions can be reproduced without future knowledge.

2026-07-19 · Ethan Beirne · 7 min read

Definition

What is bitemporal AI memory?

Bitemporal AI memory records two independent timelines for every fact. Event time describes when the fact was valid in the real world. System time describes when the AI system received, revised, or retired that fact. Together they let an agent answer current questions while also reconstructing the exact information available during a prior decision.

The two clocks

Event time and system time answer different questions.

Event time

When did this fact apply in the world? A contract amendment may become effective on January 1 even if it arrives later.

System time

When did the agent learn this fact? If the amendment was ingested on January 12, decisions made on January 5 could not have used it.

Validity windows

Each version receives start and end boundaries on both timelines instead of overwriting the previous record.

Why it matters

Single-timeline memory silently rewrites history.

Suppose a customer risk rating is corrected from low to high. A normal database updates the row. A later audit then sees the high rating and may incorrectly conclude that an earlier decision ignored it. Bitemporal memory preserves the original value, the correction, its effective date, and the moment the system received the correction.

This prevents lookahead bias in evaluations, backtests, compliance reviews, investigations, and any workflow where facts or policies change after an agent acts.

Query model

A point-in-time query applies both temporal boundaries.

Current truth

Return the version valid now using the latest accepted system record.

Known at decision time

Return only versions that had entered the system by the selected historical timestamp.

Revised historical truth

Ask what is now believed to have been true on a past date, including corrections learned later.

Implementation checklist

How to evaluate a bitemporal memory layer.

Preserve every version

Corrections should close validity windows, not destroy prior records.

Test late-arriving facts

Backdated information ingested later must not appear in earlier decision replays.

Link retrieval to evidence

Each result should retain source identity, provenance, and the temporal window that made it eligible.

Verify deterministic replay

The same as-of query should produce the same eligible memory set and evidence package.

Related: Lookahead bias in agent memory → · AI agent audit logs → · LoCoMo memory benchmark →

Test historical recall with real temporal boundaries.

Apply for a design partnership →