Valvur SIGNAL · New

The sovereign information graph for digital trust

See the message. Reveal the scheme. Protect the audience. Disrupt the campaign.

An age-aware information-integrity, scam-prevention and audience-protection infrastructure layer for governments, platforms, games, schools, financial institutions and AI providers.

01

What message, offer or claim is being distributed?

02

What is hidden behind the link, image, QR or account?

03

How — and through which route — is it spreading?

04

Who appears to be behind or benefiting from it?

05

Which audiences are being targeted or exposed?

06

What proportionate protective action should be taken?

The problem

Modern threats move between systems. Defences don't.

Moderation examines posts. Banks watch transactions. Antivirus inspects devices. Fact-checkers investigate claims. Age-verification confirms age. Platforms only see what happens inside their own walls.

A child receives a "free skins" message in a gaming chat, follows a QR code to a fake influencer page, enters credentials into a cloned site and is nudged into a crypto payment. Each step looks harmless in isolation. The journey is the scam.

SIGNAL follows the full chain — from the first message to the final website, payment request, credential form, account, domain, narrative or beneficiary.

Misinformation & disinformation
Foreign information manipulation
Phishing & social engineering
Account & identity theft
Crypto & investment fraud
Fake giveaways & marketplace fraud
Impersonation of trusted brands
Age-inappropriate promotion
Architecture

Nine modules. One continuous signal.

Not a fact-checker. Not a keyword filter. SIGNAL is infrastructure — a layered pipeline from capture to intervention, with evidence preserved at every step.

01

SIGNAL Capture

Connects to public channels, platform APIs, games, forums, websites and authorised community environments.

Output → Normalised stream of messages, links, accounts, media and metadata.
02

SIGNAL Reveal

Resolves redirect chains, QR codes, visual text, speech, disguised wording and embedded promotions.

Output → The actual destination, offer, entity and action concealed behind the message.
03

SIGNAL Lens

Analyses claims, manipulation techniques, scam patterns, source reliability and audience suitability.

Output → Explainable content, fraud and information-risk assessment.
04

SIGNAL ScamShield

Detects social engineering, impersonation, credential theft, fraudulent payments and deceptive offers.

Output → Warnings, blocked journeys, risk signals and fraud intelligence.
05

SIGNAL Graph

Maps narratives, accounts, domains, wallets, payment destinations, communities and coordinated behaviour.

Output → Campaign, influence and scam-network intelligence.
06

SIGNAL Gate

Applies age, jurisdiction, institutional and platform policies.

Output → Allow, label, warn, add friction, age-gate, quarantine, block or escalate.
07

SIGNAL Proof

Preserves evidence, model reasoning, applicable policy and human decisions.

Output → Audit trails, regulatory reports, fraud reports and appeal records.
08

SIGNAL Feed

Produces signed, machine-readable information and threat packages.

Output → Feeds for AI, search, platforms, banks, media and public institutions.
09

SIGNAL Pulse

Measures exposure, emerging threats and intervention effectiveness.

Output → National, platform, school, financial-sector and campaign dashboards.
Scam Shield

A scam exists even when no known malicious URL does.

SIGNAL analyses the conversation, audience, persuasion technique and the requested action — not only whether a site is technically dangerous.

Threats detected

  • Credential phishing & account takeover
  • Fake game-item and marketplace offers
  • Fraudulent giveaways & promo codes
  • Impersonation of admins, banks, family, influencers
  • Investment, crypto & 'guaranteed return' schemes
  • Betting & gambling offers targeting minors
  • Fake shops, ticket sellers, subscription traps
  • Romance scams & emotional manipulation
  • Grooming, sextortion, private-image threats
  • Job, scholarship, grant & immigration scams
  • Tech-support & device-security scams
  • Money-mule recruitment
  • Recovery scams targeting prior victims
  • AI-generated voice, video & identity impersonation

Detection signals

  • False urgency, secrecy or emotional pressure
  • Requests to move to another platform
  • Requests for passwords, codes, images
  • Lookalike domains & cloned pages
  • Newly registered / rapidly changing domains
  • Redirect chains & link shorteners
  • Suspicious QR codes & embedded URLs
  • Crypto, gift-card and mule payment routes
  • Wallet, merchant & beneficiary intelligence
  • Impersonated branding & support accounts
  • Coordinated messaging behaviour
  • Offers implausibly attractive for the market
Risk passport

Every item receives a structured Information & Threat Risk Passport.

Not a simplistic "fake / safe / scam" label — a machine-readable dossier of claim, source, evidence, distribution, audience, commercial character, manipulation techniques, technical risk and jurisdiction, with explicit confidence and uncertainty.

passport / v1 · signed
conf 0.87
Claim status
Verified, supported, disputed, unsubstantiated, false, outdated
Source transparency
Established publisher, anonymous, impersonated, newly registered
Evidence quality
Primary, independently corroborated, circular, absent
Provenance
Original, edited, synthetic, impersonated, unknown
Distribution behaviour
Organic, spam, coordinated amplification, targeted infiltration
Audience suitability
General, 13+, 16+, 18+, restricted, prohibited
Commercial character
Ad, affiliate, undeclared sponsorship, scam, political ad
Requested action
Read, share, disclose, download, transfer money, provide credentials
Financial risk
Merchant, wallet, payment method, beneficiary risk
Manipulation technique
Urgency, fabricated authority, secrecy, fear, targeting
Technical risk
Malicious files, credential forms, cloned interfaces, redirects
Jurisdiction
National law, platform policy, institutional rules
Audience-aware

The response depends on who is exposed — not only on the content.

SIGNAL never needs a person's full identity. Valvur Identity supplies a privacy-preserving attribute — under 13, 13–15, 16–17, adult, verified professional — and Gate applies the appropriate policy.

Fake skin giveaway → child
Prevent credential entry, explain the impersonation signals, alert the platform.
Betting affiliate → verified minor
Block, explain the restriction, preserve compliance evidence.
Betting journalism → minor
Allow with context — do not classify as gambling promotion.
Suspicious investment offer
Strong warning, identify pressure techniques, add payment friction.
'Parent' requesting money
Recommend verification through a known channel before any transfer.
Possible sextortion
Preserve evidence safely, discourage payment, offer trusted-help pathways.
Unverified political claim
Add source context and alternative evidence — do not automatically block.
Coordinated foreign influence
Alert authorised analysts, create a campaign evidence package.
Scenario

One intervention. A whole network exposed.

  1. 1

    A 14-year-old receives an Estonian-language message in a gaming group offering free skins.

  2. 2

    SIGNAL identifies urgency and an imitation of a known game brand.

  3. 3

    SIGNAL Reveal reads the QR code and follows a multistage redirect.

  4. 4

    First destination: a cloned login page designed to steal credentials.

  5. 5

    After login, the user would be redirected to an offshore betting affiliate.

  6. 6

    Valvur supplies an anonymous 'under 16' attribute.

  7. 7

    SIGNAL Gate blocks the login form and explains the page is not operated by the game publisher.

  8. 8

    The child receives simple instructions to protect the account and tell a trusted adult.

  9. 9

    The platform receives a moderation alert.

  10. 10

    SIGNAL Graph identifies the same domain infrastructure in 27 youth communities.

  11. 11

    ScamShield links accounts, QR codes, affiliate identifiers and payment destinations.

  12. 12

    SIGNAL Proof creates an evidence package for the platform and relevant authorities.

  13. 13

    SIGNAL Feed distributes a Threat Capsule to schools, banks, media, platforms and AI assistants.

Valvur ecosystem

SIGNAL stands alone. Integrated, it becomes an operating system for digital trust.

Identity

Privacy-preserving age & role verification

Shield

Real-time protection in games, communities & devices

ScamShield

Scam, phishing & social-engineering detection

Dossier

Consent-controlled safety settings & history

Parent

Meaningful alerts, explanations & recovery

Schools

Media, financial & cyber literacy

Pulse

Institutional & national risk dashboards

PQC

Protected evidence & long-term cryptographic resilience

AI Companion

Personalised safe-decision support

Financial Connect

Risk signals for banks, insurers & PSPs

Safeguards

Government-grade — with visible limits on itself.

A protective infrastructure only earns trust when it also restrains its own reach. SIGNAL is built with explicit safeguards against surveillance creep and political misuse.

  • No classifying political disagreement as false merely because it contradicts an official position.
  • Facts, opinions, predictions, satire and value judgments kept separate.
  • Multiple-source requirements for high-impact decisions.
  • Visible confidence levels and reasons for interventions.
  • Human approval for politically sensitive or campaign-level actions.
  • Special protection for journalism, research and democratic speech.
  • User, publisher and platform appeal mechanisms.
  • Independent model, bias and human-rights audits.
  • Public methodology and transparency reporting.
  • On-device analysis for private communications where appropriate.
  • Strict authority and purpose limitations.
  • Aggregate reporting wherever individual identification is unnecessary.
First pilot

Hidden betting, scam & manipulation campaigns targeting minors.

Initial languages: Estonian, Russian, English. Initial channels: public social & gaming communities, one integrated Discord or game environment, public websites and redirect networks, text, images, QR and shared links.

Core KPIs

Precision & recall
by language and threat category
Median detection time
from first appearance
Concealed destinations resolved
% of hidden endpoints uncovered
Underage exposures prevented
count over the pilot
Credential / payment journeys interrupted
count over the pilot
False-positive & appeal-overturn rate
trust and fairness
Cross-channel campaigns identified
graph-level detections
Repeated domains / beneficiaries
network reuse insight
Valvur

Age-aware information integrity for the AI and cross-platform era.

Relevant to ministries, cybersecurity agencies, law enforcement, regulators, game developers, social platforms, banks, insurers, schools, fact-checkers and AI providers — deployable as a standalone national, institutional or platform-level solution.