7 Things You Must Never Ask AI to Do in Crypto Trading

1 · Predict the next hour's BTC direction

“Will BTC rise in the next hour?” can produce a confident answer. Confidence in the wording tells you nothing about forecasting accuracy. First check which market data the model received and when those data were captured.

A useful evaluation needs a fixed horizon, mutually exclusive labels, a simple baseline and a complete output log. Our ChatGPT accuracy evaluation guide uses invented records to explain scoring; it reports no measured model performance. Its 48-hour example cannot establish next-hour accuracy either.

The missing condition: a model without a live feed or supplied market data does not obtain today's order book and news just because the prompt says “current market.” Even complete, timely inputs do not turn an explanation into a verified future outcome.

The alternative: supply timestamped material and ask which changes deserve further observation, with supporting evidence and conditions that would contradict each interpretation. Check prices, thresholds and any resulting action separately; do not let the answer place a trade.

2 · Decide whether you should "all-in" right now

Suppose you ask AI "should I all-in BTC right now?" Chances are it returns a thoughtful-looking answer: 3 bullish points + 3 risks + a closing disclaimer. Looks comprehensive, is actually useless — because AI doesn't know your:

• Total funds (all-in $1,000 vs all-in $100,000 are entirely different things)
• Family obligations (mortgage? kids? emergency fund?)
• Time horizon (you need this money in 6 months vs you can lock it for 10 years)
• Risk tolerance (you can sleep through a 50% drawdown vs you lose sleep at 10%)
• Existing exposure (already 80% BTC vs zero coins)

Without those inputs, any "advice" is empty. Picture the AI giving you a few bullish points plus a soft "consider scaling in," and you putting most of what you own into the trade. The market then pulls back, you can't hold on and you sell near the low. The AI had no way of warning you about that loss: it doesn't know your total assets or how much of a drop you can stomach, and it doesn't know you sell when you can't sleep. It should never have been asked.

The alternative: ask yourself — not AI — these 5 questions: (1) how long can I lock this money; (2) can I sleep through a 50% drop; (3) what am I already holding; (4) do I have an emergency fund; (5) am I FOMO-ing right now. AI has no informational advantage on any of these.

3 · Write you a strategy "without a stop-loss"

This is one of AI's most dangerous blind spots. If your prompt doesn't explicitly include a stop-loss, AI mostly won't add one. It will do exactly what you said — "add to position every time BTC drops X%" — all the way down to liquidation.

Say you ask it for a futures strategy that "adds to the position every time price drops 3%," and you never say when to stop adding. It will still hand you position-adding code that runs. In a sustained decline, a few rounds leave the account at high leverage and deep in the red, and one more could trigger forced liquidation. Ask it afterwards and the answer will be something like: "Your prompt didn't require stop-loss logic. I wrote what you asked for." It has a point. The mistake was leaving the constraint out of the prompt.

The alternative: any prompt that asks AI to write a trading strategy or pyramid logic must lead with this block:

All strategies must satisfy these risk-control rules — refuse to generate code if violated:
1. Per-trade max loss does not exceed 2% of total capital
2. Force-close all positions if cumulative drawdown exceeds 10%
3. Leverage caps at 3x
4. Before every add-on, re-evaluate whether the trend has reversed

With that block in the prompt, the AI has explicit constraints to check against. Whether the generated code actually contains the matching stop-loss logic is still for you to verify, line by line.

4 · Read on-chain data by making it up

Ask AI "how much BTC did address 0xABC...123 buy in the last 7 days?" and a model with no on-chain lookup tool connected is quite likely to invent a plausible-looking number. Ask it "show me this wallet's transaction history on August 5, 2024" and it will fabricate transaction hashes, block numbers, counterparty addresses — all believable.

This is hallucination. AI knows "on-chain data should look like this" — block numbers are integers, hashes are 64-char hex — but it doesn't know what actually happened at a specific address on a specific day. Its training data has a cutoff date; the live on-chain state isn't available to it. Treat a "whale wallet transfer record" produced that way as an alpha signal and buy the same tokens, and when you finally check a block explorer, the address, the tokens and the transfers may not match anything at all.

The alternative: (1) never let AI "recall" on-chain data — always make it call a tool. Binance Skills Hub's address_monitor / wallet_tracker can query Etherscan / BscScan in real time. (2) Once the data is in, let AI interpret it. Never let AI source the data itself.

5 · Evaluate whether a new token is worth buying

Ask AI "what about this token?" and you almost always get a templated answer: (1) list the use case from the official site (2) list the tokenomics (3) a "do your own research" caveat. This is worthless — it just re-reads the whitepaper at you.

The deeper problem: AI cannot tell whether a project is real. The whitepaper can be plagiarized, the team can be anonymous fakes, the smart contract can be a knockoff, on-chain volume can be wash-traded — AI sees none of that. It reads "what the project says it is," not "what the project actually is."

Make the question concrete. Before some "AI + DePIN" project launches its token, you ask an AI what it thinks, and you get back "this project combines AI and decentralized storage, with a team from Silicon Valley." The only basis for a description like that is the project's own website. Should the project later rug, that assessment will turn out to have been marketing copy read back to you.

The alternative: split token review into 3 independent steps — (1) use Skills Hub's token_analysis to pull the objective data (holder concentration, liquidity, honeypot check); (2) read the contract source yourself, or use a professional tool like De.Fi Scanner; (3) verify team, partners, and community activity yourself (not just Telegram member count — look at message frequency and original content ratio). AI helps with step 1. Steps 2 and 3 are not for AI.

6 · Design a 100x leverage strategy

AI will happily design 100x perpetual strategies for you — candle analysis, entry/exit logic, stop-loss and take-profit levels. Looks professional. But the higher the leverage, the smaller the move against you the position can absorb. At 100x, the AI's response latency and its occasional mistakes are both magnified, and it has no answer for extreme liquidity events (a sudden exchange outage, for instance).

A backtest won't rescue it either. Have an AI design a high-leverage ETH perpetual strategy, such as "long the 1-minute breakout off the open in a ranging market," and the backtest may look beautiful. But the backtest has none of the real friction in it, like a thin order book on the opening bar and the slippage that comes with it, and at high leverage one bad fill can push the position into liquidation. The AI won't point that gap out unprompted.

The alternative: 100x strategies should not be designed by AI at all — they belong on professional quant platforms (Hummingbot / Freqtrade / your own low-latency system). If you insist on AI assistance, cap leverage at 3x–5x. That's the band where AI response time and slippage tolerance actually work.

7 · Follow some KOL's call on your behalf

Sometimes users discover AI "can read Twitter" (most models actually can't read it live — they're using cached data or making things up), then tell AI "track @bigshot's recent calls and copy them for me."

Multiple things are wrong here:

• A "call" may be after-the-fact marketing — when the KOL tweets "long BTC," they may already have bought; when they tweet "short BTC," they may already have closed. The tweet won't tell you which.
• AI cannot verify the KOL's actual entry and size. "I'm positioned" could mean a $100 test or $100K real money.
• The delay in copying can leave you buying the top — if the KOL is already +20% in when the tweet posts, how much is left for you?
• AI's faithful repetition makes you think you have alpha — what you actually have is structured old information.

Stack those four together and this is what you get: the KOL calls "ETH to such-and-such a price," and the AI opens a long exactly as you configured it to. If that moment happens to be a local top and the KOL was calling while selling, you have bought the high. The AI can't tell that from a tweet.

The alternative: (1) never let AI "trust" any specific person on your behalf; (2) let AI aggregate signals (X KOLs + on-chain data + funding rate + OI) and only alert when everything aligns; (3) every actual order decision stays in your hands — AI provides data, you provide judgment.

One-line summary

Where AI helps in crypto trading is with organizing: pulling data from several sources into one view, writing strategy code cleanly, and making execution discipline routine. Predicting, judging and deciding whom to trust stay with you.

Before you give AI a job, ask yourself one question: is this organizing facts that already exist, or is it making a judgment for me? If it's the second, don't hand it over.

— PromptDeck, published 2026-05-22, revised 2026-10-06
Further reading: How to spot when AI is wrong ｜ ChatGPT accuracy evaluation method ｜ Grok and X trading signals: checking misreads