Atlas Librarian · CMI v2

Adaptive recall asks which memory is useful for this cognitive state—not merely which text looks similar.

The next generation of the Conscious Memory Index introduces an additive adaptive-recall path for Atlas Librarian. It evaluates a wider set of eligible memory candidates, considers their cognitive role and current lifecycle, and assembles a bounded Working Memory for the task. The established fast-recall path remains available while CMI v2 is evaluated and strengthened.

State-awareRecall can respond differently to continuation, temporal relevance, task scope and operational risk instead of applying one ranking posture to every request.
Authority-awareMemory importance does not override governed business authority. Rules, provenance and authoritative memory status remain distinct from contextual history.
Context-budgetedWorking Memory is assembled against an explicit context budget so persistent project knowledge does not become permanent prompt clutter.

Algorithmic direction

The memory architecture now has a policy layer that can optimize what becomes active.

Segmented memory provides structure. CMI v2 adds a selection policy across that structure. The public description identifies the signal families and observable behaviour, while the exact ranking recipe remains proprietary.

Cognitive role

A rule, procedure, handover, decision, event and durable fact are not treated as interchangeable passages. The memory type contributes to its fitness for the current task.

Authority and risk

Higher-risk tasks can require a stronger preference for governed or authoritative memory, while lower-risk reasoning can admit more contextual material.

Lifecycle and supersession

Freshness, thermal state, actual use and supersession help distinguish currently useful knowledge from material that remains valuable only as history.

Temporal and continuation state

A due follow-up or a session restart can change which memory classes deserve priority even when the underlying text has not changed.

Evidence relationships

Links among memory items can contribute context across cognitive segments without collapsing them into a single generic store.

Verified outcome evidence

When an operational result is explicitly verified, that evidence can inform future recall as one bounded signal rather than allowing retrieval frequency alone to define usefulness.

Value per unit of context

Persistent intelligence is useful only if active cognition stays selective.

CMI v2 treats Working Memory as a scarce reasoning resource. Candidate memory can be considered without all of it becoming active context, and only the final selected set is counted as checked out by the adaptive path.

Wider consideration

The Librarian may inspect a larger eligible candidate pool before deciding what should participate in reasoning.

Bounded assembly

Optional memory competes for a finite Working Memory budget. Mandatory governed context can be preserved even when it is not the cheapest context to load.

Cleaner telemetry

Candidate consideration is separated from final selection so the system does not teach itself that every inspected memory was genuinely used.

Outcome before popularity

A frequently retrieved memory is not automatically a good memory. Verified successful, failed or rolled-back outcomes can provide stronger evidence about future usefulness.

Implementation status · 5 September 2026

CMI v2 exists as an additive internal evaluation path; it is not being presented as a finished universal benchmark winner.

The current CMI v2 alpha implements state-aware candidate scoring, authority and risk sensitivity, lifecycle and supersession handling, relationship signals, verified outcome evidence and bounded Working Memory assembly. Speedway is retaining the existing recall path in parallel while comparative evaluation expands.

Implemented

Adaptive candidate ranking, bounded context assembly, mandatory authoritative-rule handling, supersession suppression, temporal and continuation signals, outcome evidence and selection-only checkout telemetry.

Under evaluation

Broader cross-segment graph traversal, longer-horizon consolidation policy, richer outcome learning and comparative public benchmark coverage remain active development areas.

No inflated claim

The first AMES result is a synthetic contract verification suite. It confirms intended behaviours but does not by itself prove superiority over another memory architecture.

See the evidence record →

Protected implementation

Speedway discloses the cognitive objective and evidence—not the recipe required to clone it.

Exact equations, weights, thresholds, tie-breakers, private prompts, routing recipes, graph traversal policy, consolidation triggers, internal schemas, private evaluation corpora and security-sensitive topology are intentionally withheld as proprietary or security-sensitive intellectual property.

Public

Signal families, architecture, observable behaviour, version status, benchmark methodology, sanitized evidence, limitations and known failure classes.

Private

The enabling implementation details that would permit direct replication, reverse engineering, attack optimization or appropriation of Speedway's proprietary methods.

Proof without surrender

Our transparency standard is that material technical claims should become measurable and versioned, not that the implementation itself must be disclosed.

CMI v2

Structure tells Atlas what a memory is. Adaptive recall decides whether it should influence this decision now.