Fake AI trading bots are one of the most common scams in crypto and forex. They look professional, they use the language of artificial intelligence, and they promise returns that no honest trader would promise. The pitch is always similar: deposit money, let the AI trade for you, and withdraw profits every week. The reality is usually different. The bot either does not exist, does not trade, or is designed to drain your account through fake profits and blocked withdrawals.

The good news is that most of these scams share the same warning signs. If you know what to look for, you can spot them before you pay anything. This guide walks through the red flags, the checks to run, and the safe way to test any bot you are considering.

Guaranteed profits are always a lie

No trading strategy can guarantee profits. Markets move unexpectedly, strategies have losing streaks, and even the best systems lose money in some months. Any bot that promises a fixed daily return — 1 percent a day, 5 percent a week, 300 percent a month — is lying. The same applies to claims of 95 percent win rates or zero losing trades. Real trading results include losses. A provider that hides its losses is hiding its real performance. If the marketing promises certainty, treat the product as a scam until proven otherwise. Legitimate bot providers talk about risk, drawdown, and losing periods. Scammers talk only about profits.

Check for a verifiable track record

Ask for proof that the bot has traded real money over a meaningful period. A verified track record on a third-party platform, a public trading account with a full history, or a live demo you can watch over several weeks is a reasonable request. Screenshots are not proof. Anyone can fabricate a screenshot in minutes. Videos of a dashboard can be staged. Be suspicious if the provider refuses to share a track record, shares only a few days of results, or shows results from a demo account presented as live trading. Also check whether the results include fees and drawdown. A track record without drawdown figures is incomplete by design.

Look at the company behind the bot

A real product has a real company: a registered business, named founders, a physical address, and a support team that answers questions. A scam often has none of these. Check the website for a company registration number, a privacy policy, and terms of service. Search the company name plus words like scam, review, and complaint. Look for the founders on LinkedIn and see whether their profiles look real and consistent. Be wary of websites registered only weeks ago, anonymous teams, and contact details that are only a Telegram username. If you cannot find out who runs the bot, you should not trust it with your money.

How the money flow reveals the scam

Pay close attention to how the bot asks you to deposit. Legitimate bots usually connect to your existing exchange account through an API key, and your funds stay on the exchange. Scams often ask you to deposit directly to their own platform, wallet, or exchange, where they control the money. That is a major red flag. Other warning signs include pressure to deposit quickly, bonuses that lock your funds until you trade a huge volume, withdrawal fees that appear only when you try to withdraw, and minimum deposits that keep rising. Test withdrawals early with a small amount before you deposit more. If a small withdrawal is delayed, blocked, or requires extra payment, stop immediately.

Fake reviews and social proof

Scam bots invest heavily in fake social proof: paid reviews, fake testimonials, Telegram groups full of bots posting profit screenshots, and influencers paid to promote the product without disclosing it. Look for signs of fabrication. Are all the reviews five stars and written in similar language? Do the testimonials use stock photos? Does the Telegram group delete critical questions or ban users who ask for proof? Search for independent reviews on forums and review sites, and give more weight to detailed negative reviews than to vague praise. A product with no independent criticism at all is more suspicious than one with a mix of honest reviews.