← All articles

That bargain supplier does not exist: checking a company is real before you pay

A few years ago you could spot a scam website from a distance: pixelated photos, badly translated text, a design that creaked. Today, with artificial intelligence, anyone can put together an impeccable website in an afternoon — product catalogue, team photos, enthusiastic reviews, even blog posts with years of "track record". All fake, and all very convincing.

For a small company that is a new problem: the visual filter we have always relied on — "this site looks dodgy" — no longer works. The scam no longer looks like a scam.

How the trap works

The most common scheme has three variants:

The miracle supplier. You are looking for materials, machinery or stock and you find a new supplier with prices 30% below market. The website is professional, they reply quickly and sensibly (often it is a chatbot), they send a catalogue and a pro-forma invoice with every detail in place. You pay the deposit by bank transfer and nothing ever arrives.

The shop that never ships. The consumer-facing version, which also catches companies buying equipment: an online shop that appeared from nowhere, with aggressive offers and urgency ("last units"). It charges and vanishes, or ships something unrelated.

The clone. The most dangerous: they copy a real company's website, with an almost identical domain (a hyphen, a ".shop" instead of ".com", one letter changed). You think you are buying from your usual supplier; the money goes elsewhere.

In all three cases AI does the heavy lifting: it writes the copy without a single mistake, generates the product and "team" photos, produces dozens of varied, believable reviews, and answers your emails naturally. What used to take an organisation, one person now does over a weekend.

An example that is far too realistic

A joinery workshop needs fittings for a large order. The usual supplier quotes three weeks; looking for alternatives, the foreman finds a distributor with an impeccable website, a company number on the contact page and excellent prices. Emails go back and forth with an attentive sales rep who answers at any hour. They pay a €4,200 deposit by transfer. The following week the phone number no longer exists, the website has gone, and the company number turns out to belong to a firm dissolved years ago, copied from a public register.

Nothing in the conversation felt wrong. That is precisely the trap.

The warning signs

The checks before paying (ten minutes)

  1. Check the domain's age in a free whois lookup. Under a year old plus "long track record" is a direct red flag.
  2. Verify the company registration in the official register or a company-search service: the company exists, is active, and the name matches who is invoicing you.
  3. Check the bank account. The IBAN holder must be the company on the invoice. If you are asked to pay to a different name, stop.
  4. Call through a channel you find yourself (not the one on their website): the number on the register, on Google Maps, or from the trade association. If it is a clone of a real company, this is how you find out.
  5. Search the name plus "scam" or "reviews", and check whether the address shows real activity on Street View.
  6. A small first order, or payment on delivery where possible. No serious supplier is offended by that.

And if you have already paid?

Move fast: call the bank immediately to try to hold or recall the transfer (the first hours are decisive), report it to the police with every email and receipt, and keep everything without deleting anything. If you entered passwords or card details, change them and tell the bank about that too. The full sequence is in the piece on the first 24 hours.

Protecting yourself is not about distrusting everyone new. It is about replacing the visual filter that AI has made useless with a ten-minute one that it cannot fake.

This article is general information, not legal advice for a specific case.

About this article

iaetica.cat is a space for writing in the open: what I am learning about artificial intelligence, its ethics and its law.

— Laura Graells