Technology

The model is not the only black box.

blkbx is an evidence infrastructure layer for autonomous systems. Our core abstraction is the evidence receipt — but underneath every receipt is a local-to-global reconstruction engine.

The local-to-global problem

A real AI incident is not caused by a single output. It emerges from a chain of local evidence fragments that must agree: the prompt, the retrieved documents, the policy constraint, the model output, the agent plan, the tool call, the human review, the system mutation, and the downstream consequence.

The hard problem is not merely recording each fragment. It is determining whether those fragments agree. When they glue into a coherent global record, the action is reconstructable. When they do not, the system surfaces an obstruction.

Evidence cover

U₁

prompt context

U₂

retrieved documents

U₃

policy references

U₄

model output

U₅

agent plan

U₆

tool call

U₇

human review

U₈

downstream action

Sheaf-native evidence

Why sheaves?

Insurance depends on reconstructability. AI systems create distributed evidence. A sheaf is the mathematical object for exactly this situation: local data, overlap rules, gluing conditions, and obstruction when gluing fails.

This is the difference between observability and insurance evidence. Observability asks what happened? blkbx asks can what happened be reconstructed, verified, and defended?

Grounded in sheaf-theoretic BLT work — local sections, learned restriction maps, cohomological diagnostics, sheaf Laplacian spectra, stalk-level inconsistency, and obstruction classes.

Local section

Evidence fragment from one part of an AI action

Hook window

Capture point in a model, agent, workflow, or system

Restriction map

Rule describing how two fragments should agree

Coboundary

Measured disagreement between fragments

Global section

A coherent, reconstructable event story

H⁰

Space of globally consistent explanations

Irreducible obstruction in the record

Sheaf Laplacian

Consistency and robustness operator

Stalk defect

Local contradiction at a specific decision point

Obstruction energy

Fraction of the event not explained by coherent evidence

From AI action to claim-reviewable record
STEP 1

Capture

Capture the consequential action at the model, agent, API, workflow, or system boundary — claim denial, loan approval, refund, coverage recommendation, account mutation, exception approval.

STEP 2

Issue receipt

Create a signed evidence receipt with model, policy, input, output, tool, human-review, and action metadata.

STEP 3

Build evidence cover

Decompose the action into local evidence regions U₁…U₇.

STEP 4

Test consistency

Check whether overlapping regions agree — does the output match the policy? does the tool call match the approved action? does review cover the actual consequence?

STEP 5

Surface obstruction

If evidence does not glue, identify the obstruction: missing review, policy mismatch, version conflict, undocumented tool call, retrieval gap, unauthorized action.

STEP 6

Export insurance packet

Produce artifacts for underwriting, audit, incident response, claims review, broker submissions, carrier review, and reinsurer analysis.

Consistency diagnostics

Not decorative math — insurance questions.

global-section existence

Can the event be reconstructed?

H⁰ consistency dimension

How much of the record is coherent?

H¹ obstruction count

Where does the evidence disagree?

spectral gap

Is the workflow insurable?

stalk defect

Which control failed?

obstruction energy

Is the record complete enough for claims review?

Insurance export layer

From technical evidence to insurance-facing artifacts.

  • Underwriting control exhibits
  • Broker submission packets
  • AI incident reports
  • Claims reconstruction files
  • Audit trails
  • Regulatory review packets
  • Reinsurer exposure summaries
  • Renewal evidence