Trust Centre
How VILIQ works, stated plainly
Everything on this page describes behaviour that is implemented. Where something is not connected, it says so rather than describing an intention in the present tense.
Data
Provenance and freshness
Every market figure carries the publisher that produced it, the time the publisher assigned, the time VILIQ stored it, and a freshness state. Each figure is also labelled by kind: live quote, daily reference, official fundamental or VILIQ calculated — so a reference rate is never presented as a tradeable price.
A source that fails is never replaced with simulated data. The surface reports itself unavailable and names the reason. Production refuses to start at all if demonstration data providers are enabled.
Served today
- BTC and ETH quotes and history from CoinGecko
- ECB daily euro reference exchange rates, and cross rates VILIQ calculates from them
- The US Treasury yield curve, including the 10-year
- NY Fed EFFR and overnight reverse repo (ON RRP) volumes, and the Treasury General Account
- Policy rates of five central banks
- EIA weekly energy fundamentals — inventories and production, not prices
- News from eight primary public feeds
- Trust Check: domain registration through RDAP and ACN/ABN checksum validation
Not served
- Equity indices, single stocks, metals, DXY and VIX — no approved source
- Energy prices, futures and options — no approved source
- A current USD/VND rate and Vietnamese dealer gold quotes — no approved source, so the fields stay empty
- ASIC, ABR and IP Australia lookups for Trust — agreements needed, reported as not checked
- Identity verification — no provider, so creator profiles and VILIQ PRO applications stay closed
- Payment — not active; nothing can be bought
Models
What the scores mean, and what they do not
- 01
Observed and inferred are kept apart
Measured market pressure is presented separately from any model outlook. They are different kinds of claim and are never merged into one figure.
- 02
Scores are deterministic and versioned
The same inputs produce the same output, and every score names the model version that produced it. No language model computes a score.
- 03
Insufficient data yields no score
Below the minimum observation threshold VILIQ withholds a score rather than computing one on noise. Silence is a result.
- 04
Separate scores stay separate
Market Score, Trust Score and risk assessments measure different things and are never combined into a single universal number.
- 05
Trust Score is about evidence, not safety
It measures the quality and consistency of available evidence about an entity. It is not a prediction of investment outcome and must not be read as one.
AI
What VILIQ Research can and cannot do
VILIQ Research answers from VILIQ’s own tools, records and retrieved sources. Answers are checked against that evidence before they are shown, and a figure with no source behind it is withheld rather than rendered.
Retrieved content — news items, documents, pages submitted for a Trust check — is treated as data throughout. It is fenced and labelled before any model sees it, and nothing inside it can issue an instruction, claim authority or cause a tool to run.
Where no model provider is configured, VILIQ Research reports itself unavailable. It does not fall back to answering from memory.
VILIQ Research will not
- State a market figure no source produced
- Invent your balance, rewards or progress
- Give personal financial advice or a recommendation to buy or sell
- Follow instructions embedded in content it reads
- Present its own analysis as a source fact
Security and privacy
Implemented controls
Row-level security is enabled and forced on every table. Authorisation is decided from server-side role records; user-editable account metadata confers no privilege, and that is verified by tests that attempt the escalation directly.
Ledgers, prediction records and audit entries are append-only, enforced by database triggers rather than by convention. Privileged actions are authorised on the server, and the most sensitive ones require a second approver.
Logs are structured and redacted by field name and by value shape, so a credential quoted inside an error message does not survive into a log. Secrets are scanned for on every release, and the scan blocks the release when it finds something.
Academy credentials you earn can be checked by anyone you show them to, through a public verification page, without exposing anything else about you. Passport visibility is yours to set; sensitive fields are private by default.
VILIQ Points
The status of VILIQ Points
VILIQ Points recognise eligible participation and contribution within VILIQ. They are off-chain, non-transferable, not purchasable or sellable, and have no market value.
Nothing in the product is an offer of a financial product, and no statement about VILIQ Points is a promise of value.
Positioning
What VILIQ is, in regulatory terms
VILIQ provides market information, analytics, research, education, trust verification and simulated trading. It does not provide personal financial product advice, does not take your circumstances into account, does not execute trades and does not hold client money or assets.
Language across the product is deliberately constrained. Surfaces use observed, model outlook, probability, scenario, historical analogue and evidence. Terms implying certainty or recommendation are not used, and that constraint is enforced by tests.
Report a problem
Found something wrong?
Incorrect evidence, a misattributed source, a security issue or a Trust result that looks unjustified — we would rather hear it than not.
Security reports are triaged ahead of everything else. Please do not include exploit details in a first message; describe the class of issue and we will arrange a secure channel.
VILIQ is operated by LTM Trading Pty Limited (ACN 659 211 426), Australia.