Explore our new Cybersecurity Terms Glossary, explained by our experts.
Cybersecurity Terms Glossary
An IDN homograph attack registers a domain containing look-alike Unicode characters so that it renders identically to a trusted brand in the browser address bar and in email.

What Is an IDN Homograph Attack?
An IDN homograph attack registers a domain name that contains characters from a non-Latin script which look identical, or near identical, to Latin letters. The result is a web address that a person reads as apple.com or barclays.co.uk but which resolves somewhere else entirely. The deception happens in the reader’s eye rather than in the protocol, which is what makes it so effective against people who have been trained to “check the URL”.
Internationalised domain names exist for a good reason: the internet is not written only in ASCII. IDNA, defined in RFC 5890 and its companions, encodes non-ASCII labels into an ASCII form called Punycode, prefixed with xn--. A Greek-script domain travels across DNS as an xn-- label and is rendered back in its native alphabet by the browser.
Unicode contains many characters that are visually indistinguishable from Latin ones at normal font sizes. The Cyrillic small letter a (U+0430) and the Latin a (U+0061) are separate code points that render the same in most typefaces. So do Cyrillic е, о, р, с, and х, Greek ο and ν, and a long list of Armenian, Cherokee, and mathematical alphanumeric symbols.
The best-known demonstration came from the researcher Xudong Zheng in April 2017. He registered xn--80ak6aa92e.com, built entirely from Cyrillic characters, obtained a free TLS certificate for it, and showed that Chrome and Firefox displayed it in the address bar as apple.com complete with a padlock. The browsers were patched within weeks.
Modern browsers do not display a Unicode label blindly. Chromium and Firefox apply a set of rules that fall back to the raw Punycode form when a label looks suspicious:
Registry policy matters just as much. Nominet and most responsible registries publish per-script character tables and block the registration of labels that mix incompatible scripts. Several also operate variant bundling, where registering one form reserves its confusable siblings.
Neither layer is complete. Whole-script confusables—a label made entirely of Cyrillic characters, which breaks no mixing rule—are the hardest case, and defences differ between Chrome, Firefox, Safari, and the many applications that render links without a browser engine at all.
The browser address bar is now the best-protected surface. The weak points are everywhere else:
Defence is mostly about visibility and normalisation. Monitor Certificate Transparency logs for certificates issued to names that normalise to your own, which catches a homograph domain at the moment it becomes usable over HTTPS. Feed newly registered domain data through a confusable-character comparison against your brand portfolio rather than a simple string match—the Unicode Consortium publishes the confusables table that makes this straightforward.
On the receiving side, configure mail gateways to convert incoming links to Punycode before applying reputation checks, and to flag any sender domain containing non-ASCII characters where none is expected. Defensive registration of the highest-risk variants is worth the renewal fees for a small number of names, though it cannot scale to every combination.
Finally, remove the dependency on visual inspection where you can. Hardware-backed multi-factor authentication using FIDO2 binds a credential to the real origin, so a passkey registered for your genuine domain simply will not offer itself on a homograph clone, no matter how convincing the page looks. That is a stronger control than any amount of awareness training.
Our Social Engineering Services include look-alike domain simulations that show how your staff and your filters handle a name they cannot tell from your own.
Copy the address into a Punycode converter, or look at the certificate details, where the subject name appears in its xn-- form. On the command line, a DNS lookup will show the encoded label. If a domain you expect to be plain ASCII comes back with an xn-- prefix, every character in it needs checking. Browsers will also reveal the Punycode form when their mixed-script heuristics are triggered.
No. A certificate proves that the holder controls that exact domain name, and a homograph domain is a legitimately different name, so a free certificate is trivially obtained for it. The padlock therefore confirms encryption to the attacker’s site rather than the identity the user assumed. Certificate Transparency logs are useful for detection, but the certificate itself offers no protection to the visitor.
They are related but distinct. Typosquatting relies on the user making a mistake, registering domains such as gogle.com in the hope of catching mistyped addresses. A homograph domain requires no mistake at all: the user clicks a link that displays correctly and arrives at the wrong place. Homograph names are generally more convincing and harder to spot in a log file.
For a handful of high-value names, yes. The number of possible Unicode substitutions across popular top-level domains makes exhaustive registration impossible, so treat it as targeted rather than comprehensive. Spending the same budget on Certificate Transparency monitoring, newly registered domain feeds, and phishing-resistant authentication usually buys more protection per pound.
You May Be Interested In These
We use a few cookies that are needed to make this website work, and they are always on. We would also like to set analytics cookies, to see which pages are useful; functional cookies, so that our live chat works; and advertising cookies, to measure our campaigns. We will not set any of those until you say yes, and you can change your mind at any time from the Cookie Settings link in the footer.
Choose the cookies you are happy for us to use. Strictly necessary cookies are always on; everything else is off until you turn it on.
These cookies keep the website working: they hold your security checks, your form submissions and the choice you make here. The site cannot run without them, so they are always on.
These let our live chat open and remember your conversation, and let the newsletter sign-up work. Switch them off and the chat window will not appear.
These tell us which pages are read and how people find us, so that we can improve the site. The reports are about groups of visitors, never about you personally.
These let us measure our advertising and show you our adverts on other websites, such as LinkedIn, Google and Meta. Switch them off and you will still see adverts, but they will be less relevant.