Guide

How to evaluate if an AI trading bot is legitimate

AI trading products often blur marketing and market data. This guide is a diligence checklist you can apply to any bot — then see how Stratium maps to each item.

Open live deskCompare modes

What should you check before funding an AI trading bot?

Start with signals you can verify in minutes without depositing. If a product fails the first three checks on a public demo, treat funding as optional — not urgent.

Diligence checklist vs Stratium
CheckWhy it mattersStratium
Named price venueAnonymous “proprietary feeds” are hard to verifyCoinbase Exchange public REST + WebSocket
Paper vs live labelsStops demo P&L from being mistaken for brokerage fillsPaper / backtest / live modes labeled on the desk
Reasons per fillGreen curves without thesis are weak evidenceNarratives with side, size, confidence, and why
Risk limits before depositCapital should not precede guardrailsMax %/trade + daily loss stop; live unlock at 20 USDC
No guaranteed returnsGuarantees are a trust red flag in cryptoExplicit disclaimer: paper ≠ future returns

Why does a named market-data venue matter?

If a product will not say where the mid price comes from, you cannot sanity-check the tape. Stratium prints Coinbase USD pairs (BTC-USD, ETH-USD, SOL-USD, XRP-USD) sourced from Coinbase Exchange so you can compare against Coinbase’s public market pages within a few seconds. API surface area for candles and products is described in the Coinbase Exchange API docs.

Small differences can appear from timing, last-trade versus mid, or network lag. Large persistent gaps between a bot’s printed mid and the named venue are a red flag — regardless of how polished the UI looks.

How should paper, backtest, and live be labeled?

Legitimate products state what is simulated. On Stratium, backtest and paper fills do not move real balances; live trading sizes from deposited USDC and updates the ledger on realized P&L. See /guides/backtest-vs-paper-vs-live for a full mode comparison table.

What red flags should make you walk away?

  • Guaranteed returns, “risk-free AI,” or equity curves presented as brokerage-like fills without a paper/live distinction.
  • Unnamed or “proprietary only” price feeds with no public venue you can cross-check.
  • Deposit required before you can see how decisions are explained.
  • Pressure to fund immediately after a short demo win streak.
  • No risk limits (max position size, daily loss stop) before live mode unlocks.

How does Stratium map to this checklist in practice?

  1. Open / with $0 at risk — confirm Coinbase mids and paper labels on the desk.
  2. Read a few fill narratives (side, size, confidence, reason) before considering funding.
  3. Open /?mode=backtest or /backtest to inspect historical markers without capital risk.
  4. Review timing specs in /guides/agent-timing-lab-note so pacing is not mistaken for edge.
  5. Only then consider sign-in, risk limits, and USDC funding (live unlock at 20 USDC).

What should you do after the checklist?

Diligence is a sequence, not a single screenshot. Watch paper, run a backtest, read mode labels, then decide whether funding is worth your risk tolerance. Product overview: /how-it-works. Guide index: /guides.

  1. Watch Stratium paper fills on the live desk with $0 at risk.
  2. Run a backtest replay to inspect historical markers and narratives.
  3. Only then consider sign-in, risk limits, and USDC funding.

FAQ

Frequently asked questions

Can an AI trading demo be legitimate if it starts in paper mode?

Yes — starting in paper with a named price venue is a positive signal. Legitimacy still requires clear live labeling, risk disclosure, and no guaranteed-return claims.

Does Stratium guarantee profits?

No. Stratium is not financial advice. Paper fills and backtests are for demonstration. Crypto trading with real funds can result in partial or total loss.

Where can I compare Stratium’s modes?

See /guides/backtest-vs-paper-vs-live and /how-it-works for product detail, or open the desk at / and /?mode=backtest.

Where do Stratium prices come from?

Stratium uses Coinbase Exchange public REST and WebSocket APIs for BTC-USD, ETH-USD, SOL-USD, XRP-USD. Compare printed mids to Coinbase’s public market pages when verifying a session.

What timing specs should I know before judging a Stratium demo?

See /guides/agent-timing-lab-note: ~12s fill cooldown, ~1.6s decide-then-execute, and backtest replay ~140 ms/bar at 1x. Pacing is product UX — not a performance claim.

Continue on Stratium

Use the desk to verify behavior on Coinbase prices — then fund only if it earns your trust.

Open live deskCompare modes