MODERN INTELLIGENCE. CLASSICAL DISCIPLINE.

The market prices the headline.
Ailanum prices what comes next.

Complex public information, structured. Ailanum uses AI-assisted extraction and scoring to turn hard-to-track changes across company narratives and relationships into institutional-grade signal inputs.

SIGNAL INPUT
Built from public company disclosures and relationship context that most teams do not structure consistently.
PROPAGATION MODEL
Signals do not stop at the originating company. The model traces where exposure and second-order effects can travel.
RESEARCH READY
Designed for firms that want differentiated inputs they can test, combine, and use inside existing workflows.

Signal output,
ready for integration.

Ailanum structures narrative and relationship signals into a format that can move directly into research, risk, and systematic workflows.

signal_family: narrative and network signals
coverage: public-company universe
delivery: flat file, API feed
cadence: event-triggered

TICKER   SIGNAL   SCORE PROPAGATION
XOM supply chain stress ↑ +0.34σ 3rd order
CAT earnings language delta ↓ −0.18σ 2nd order
DE supply chain stress — +0.07σ 1st order
LMT warranty accrual delta ↑ +0.21σ 2nd order
CVX litigation flag ↓ −0.29σ 1st order
updated: 2026-04-07T09:14Z  |  public company coverage  |  relationship-aware scoring
01  —  EXTRACT
Reads the filing.

Narrative, disclosure, and operating context are converted into structured signals. The focus is on material shifts, not just headline metrics.

02  —  PROPAGATE
Traces the network.

Signals propagate through supplier, customer, and competitive relationships so first-order events can be evaluated as broader second and third-order exposures.

03  —  DELIVER
Delivers the signal.

Institutional-grade outputs structured for direct use in model research, monitoring, and production pipelines.

Built for systematic strategies, quant hedge funds, and multi-manager platforms.

What the data is.

Ailanum processes large volumes of public company disclosures and relationship data, then structures that material into signal families that can be tested, monitored, and integrated systematically.

AI supports the extraction and classification layer, while the broader system focuses on turning qualitative changes and inter-company relationships into disciplined, research-ready outputs.

SYSTEM DESIGN

The objective is not to summarize documents. It is to identify meaningful shifts, map where those shifts can travel, and produce signals that stand up inside institutional workflows.

Six signal categories.

CATEGORY 01
Supply Chain Stress

Language shifts in supplier, sourcing, and logistics disclosures. Propagated through the relationship graph to identify downstream exposure before it surfaces in financials.

CATEGORY 02
Earnings Language Delta

Year-over-year NLP scoring of Management Discussion and Analysis sections. Deviations from prior-period language are the primary signal.

CATEGORY 03
Litigation Flags

Emerging or escalating legal exposure extracted from legal proceedings disclosures. Scored for materiality and temporal change relative to prior filings.

CATEGORY 04
Leadership Change Signals

Proxy and 10-K language changes indicating executive transitions, board composition shifts, or compensation structure modifications.

CATEGORY 05
Warranty Accrual Delta

Warranty and product liability disclosure changes. Early indicator of product quality issues or liability exposure, typically leading financial statement impact by one to two quarters.

CATEGORY 06
ESG & Regulatory Exposure

Environmental, safety, and regulatory language changes across risk factor and operations disclosures. Scored for direction and magnitude of shift.

What the dataset emphasizes.

The emphasis is signal quality and propagation-aware context rather than breadth for its own sake. The dataset is designed for teams that want differentiated inputs they can evaluate alongside their existing research stack.

CoverageBroad public-company universe with relationship-aware mapping
Input typesPublic disclosures, textual changes, and inter-company relationship context
Update cadenceEvent-triggered refreshes as new information becomes available
Signal score formatNormalized sigma (σ) relative to universe, by signal category
Propagation depthUp to third-order via the corporate relationship graph
Research useDesigned for factor research, monitoring, and model development
Delivery formatFlat file (Parquet, CSV) or REST API feed

Where Ailanum sits relative to alternatives.

Most adjacent products either stop at document processing, focus on generic alternative data, or present information in a research interface. Ailanum is positioned around structured signals plus propagation logic.

CATEGORY 01
Document Intelligence Tools

Strong at search, summarization, and workflow assistance, but typically not built as a propagation-aware signal dataset.

  • Useful for analyst workflows and ad hoc review
  • Often focused on documents one company at a time
  • Usually stops short of systematic signal packaging
CATEGORY 02
Traditional Alternative Data

Broad and often expensive, but not always designed to capture subtle narrative shifts or how those shifts propagate through company relationships.

  • Often emphasizes breadth and raw coverage
  • Can be harder to connect to qualitative change
  • May not model second-order transmission explicitly

Designed for existing infrastructure.

Ailanum is built for firms that already have research, risk, or systematic infrastructure in place and want differentiated signal inputs rather than another front-end workflow tool.

Built for institutional use.

Ailanum is shaped by experience building decision systems around messy, incomplete, and fast-moving information. The objective is not to produce more commentary. It is to produce signals that are structured enough to be evaluated rigorously.

The propagation model is central to that effort. Rather than treating an event as isolated to one company, the system is designed to trace how exposure can extend through related firms and adjacent contexts.

PRINCIPLES
  • Public information, structured with discipline
  • AI used as an enabling layer, not the product claim
  • Propagation-aware modeling over isolated event detection
  • Built for research teams, not casual browsing

What gets overlooked is often where the signal begins.

Ailanum focuses on information that is public, visible, and often under-structured. The edge is not in claiming secret access. It is in identifying changes that matter, classifying them consistently, and capturing where they may lead.

AI helps with extraction, organization, and scoring, but the value of the dataset comes from the system around it: signal design, propagation logic, and the discipline to make outputs usable inside institutional workflows.

Early access is exploratory. Share a few details in the private form and we will follow up directly. The contact email is intentionally not shown on the page.