--- title: Frequently Asked Questions — Noah Predict description: How Noah Predict works, what it predicts, how the models are validated, who uses it, and how it compares with traditional political risk products. url: https://noahpredict.com/faq/ source: Noah Predict — Worldwide AI Media Ltd last_updated: 2026-04-17 licence: CC-BY-4.0-with-attribution-and-no-redistribution --- Home›FAQ # Frequently asked questions What Noah Predict does, how it works, and how to use it responsibly. Noah Predict is a predictive risk intelligence service used by insurance, reinsurance, broking, shipping, security and government teams. This page answers the questions we are asked most often. For full methodology, see methodology. For defined terms, see glossary. ## About the product ### What does Noah Predict actually predict? Noah Predict produces probabilistic assessments of how risk is moving in the near term. Its core output is a composite risk score across 7-day, 4-week and 3-month horizons for political violence, civil unrest, kidnap and ransom exposure, terrorism, maritime chokepoint disruption and sovereign instability. It does not predict single events in isolation; it identifies directional change in the underlying drivers of risk before that change is reflected in official ratings, broker consensus or market pricing. ### How is Noah Predict different from a traditional political risk rating? Traditional political risk ratings answer the question *how risky is this place today?* — a state measurement, usually updated quarterly. Noah Predict answers a different question: *how is this risk moving, and how fast?* It is a change detector, built on real-time narrative signal analysis, and refreshes every 15 minutes. The two are complementary rather than substitutes. ### Who is Noah Predict for? Political violence XL, CPRI and trade credit underwriters; treaty and facultative reinsurers; hull, cargo and war-risk underwriters; K&R and specialty crisis-response carriers; brokers preparing placements; corporate security and duty-of-care teams; sovereign analysts; and government risk functions. The free tier is aimed at individual analysts trialling the engine. ### Is Noah Predict a regulated advice service? No. Noah Predict is a decision-support tool. It is not authorised or regulated by the Financial Conduct Authority, is not an insurance intermediary, is not a credit-rating agency and is not an ESG-ratings provider. Regulated decisions must be taken by qualified professionals who bear the associated authority and liability. See /terms. ### What horizons does the engine cover? 7 days, 4 weeks and 3 months. These reflect the horizons over which directional narrative signals have historically had measurable predictive value. Longer horizons are available to enterprise customers under bespoke engagement. ## How the engine works ### What data does Noah Predict use? Noah Predict analyses global public and licensed information at scale — including news, official statements, regulatory filings, sanctions listings, maritime tracking data, court records, social signal, academic output and specialist reporting. It does not access paywalled content outside the terms of the relevant licence, and it does not train on customer inputs without written agreement. Full source categories and licensing position are described in /methodology. ### How many sources does Noah Predict monitor? Approximately 1.6 million sources, refreshed on a rolling 15-minute cycle. ### What external benchmarks does Noah use? For country-level risk outputs Noah attaches six independent external anchors to every bundle: UK FCDO travel advice, US State Department travel advisories, the Institute for Economics & Peace Global Peace Index, the World Bank Worldwide Governance Indicators, OFAC and UK HMT sanctions registries, and ACLED armed-conflict event data. For insurance-underwriting outputs, Noah produces the structure of a Marsh-style World Risk Review — WRR-adjacent peril tables, peer comparison, district-level verdicts — without licensing Marsh WRR data. ### What is a "signal fire"? A signal fire is the moment at which one or more underlying drivers cross a calibrated threshold indicating a directional change in risk. Signal fires are timestamped, source-attributed and auditable. Noah Predict fires approximately 1,847 such signals in a typical 90-day window across all verticals; positive predictive value on the composite score is currently 0.71. ### What is "narrative lead"? Narrative lead is the elapsed time between a Noah Predict signal firing and the same directional change being reflected in broker consensus pricing. Median narrative lead across verticals is currently 6.4 hours. Narrative lead is measured retrospectively against documented pricing moves and published alongside each vertical's backtest. ### How are models validated? Through rolling out-of-sample backtests against historical events, held-out windows for each vertical, and continuous calibration of confidence bands. The equal-source-vote doctrine is empirically anchored: upweighting tier-one sources reduced Noah's accuracy on resolved Polymarket markets by 30% (Brier score). Validation methodology, held-out set composition, precision/recall characteristics and known failure modes are published on /methodology. ### Can Noah Predict be wrong? Yes. Predictive output is probabilistic. Every score is published with an explicit confidence band and an audit trail linking back to the drivers and sources behind it. Every output must also name the strongest case against its own headline — the contrarian read — and the specific corpus gap that might invalidate it. Users are expected to treat output as one input into professional judgment, not as a binding forecast. ### Is every output auditable? Yes. Every Noah Predict conclusion is traceable to the underlying drivers, thresholds and source documents that produced it. This is why the provenance metric on the homepage reads 100%. Audit trails are exportable at Professional and Enterprise tiers. ## Coverage and verticals ### What risk verticals does Noah Predict cover? Political violence and conflict; terrorism and targeted attacks; civil unrest and protest dynamics; kidnap and ransom exposure; operational and access constraints; maritime and chokepoint disruption; sovereign and regime instability. Additional exploratory verticals are available through Noah Labs. ### Does Noah Predict cover every country? Coverage is global but depth varies. Benchmark-grade coverage applies to the territories where we have the densest source base and longest backtest history. Territorial coverage maps are available to Professional and Enterprise customers. ### What is Noah Labs? Noah Labs is the exploratory arm of the platform. It exposes capabilities that use the same engine but are still in early stages — for example sanctions and compliance reads, single-company watch, macro probability and markets. Noah Labs output is clearly labelled exploratory or experimental and must not be used for regulated, contractual or consequential decisions without independent validation. See /labs. ## Integration and access ### How do I access Noah Predict? Through the web workspace at noahpredict.com, via authenticated REST and webhook APIs on the Professional tier, and via custom feeds, on-premises deployment and treaty/cat-model integration at Enterprise tier. ### Is there a free tier? Yes. Free-tier users get three analyses per day with Google sign-in, full narrative lead output, and web interface access, with no card required. ### Does Noah Predict integrate with cat models? Yes, at the Enterprise tier. We support treaty and cat-model integration under a separate signed agreement. ### Where is data hosted? In secure cloud infrastructure primarily within the United Kingdom and European Economic Area, with encryption in transit (TLS 1.2+) and at rest. ## Compliance, ethics and safety ### Is Noah Predict GDPR-compliant? Yes. Worldwide AI Media Ltd is the data controller for personal data processed through the service and processes data in accordance with the UK GDPR and EU GDPR. See /privacy. ### Will Noah Predict profile individuals? No. The service is designed to analyse patterns of events, not to profile named natural persons. The Acceptable Use Policy expressly prohibits the use of the platform to surveil, score or make automated consequential decisions about named individuals without appropriate human review and lawful basis. ### Can I use Noah Predict output in reinsurance placements? Yes, subject to the Terms of Business. Output may be embedded in internal underwriting files, treaty documentation, risk registers and placement material. Raw output may not be redistributed or resold externally without a separate licence. ### Does Noah Predict use customer data to train models? No. Customer inputs are used to deliver the service and, in aggregated or de-identified form, to improve model quality. Customer data is never used to train third-party models without an explicit written agreement. ## Pricing and trials ### How much does Noah Predict cost? Free tier is £0/month. Professional is currently free during beta, standard price £490/month. Enterprise is bespoke. Full terms are on the pricing section of the homepage. ### What happens at the end of the beta? Existing beta users are given at least 30 days' notice before the Professional tier reverts to its standard price, and are offered a continued-use discount where available. ## Company ### Who operates Noah Predict? Worldwide AI Media Ltd, a company registered in England and Wales, based in London. Worldwide AI Media Ltd also operates Noah Wire (noahwire.com), a news intelligence service from which the Noah Predict engine evolved. See /about. ### How can I contact you? By email at contact@noahpredict.com, or through the form at /contact. ### Try the system. Three free analyses a day. No card required. Start free → Read the methodology --- ## Citation Cite this page as: "Noah Predict, FAQ, Worldwide AI Media Ltd, https://noahpredict.com/faq/, accessed [date]."