diff --git a/CHANGELOG.md b/CHANGELOG.md
index f3d0e1f0f5c..f8b104bcbe6 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -32,6 +32,7 @@
 - Bugfix: Fixed user info card popups adding duplicate line to log files. (#5499)
 - Bugfix: Fixed `/clearmessages` not working with more than one window. (#5489)
 - Bugfix: Fixed splits staying paused after unfocusing Chatterino in certain configurations. (#5504)
+- Bugfix: Links with invalid characters in the domain are no longer detected. (#5509)
 - Dev: Update Windows build from Qt 6.5.0 to Qt 6.7.1. (#5420)
 - Dev: Update vcpkg build Qt from 6.5.0 to 6.7.0, boost from 1.83.0 to 1.85.0, openssl from 3.1.3 to 3.3.0. (#5422)
 - Dev: Unsingletonize `ISoundController`. (#5462)
diff --git a/src/common/LinkParser.cpp b/src/common/LinkParser.cpp
index ecc8fb75169..eed4f97c151 100644
--- a/src/common/LinkParser.cpp
+++ b/src/common/LinkParser.cpp
@@ -148,6 +148,16 @@ void strip(QStringView &source)
     }
 }
 
+/// @brief Checks if @a c is valid in a domain
+///
+/// Valid characters are 0-9, A-Z, a-z, '-', '_', and '.' (like in GFM)
+/// and all non-ASCII characters (unlike in GFM).
+Q_ALWAYS_INLINE bool isValidDomainChar(char16_t c)
+{
+    return c >= 0x80 || (u'0' <= c && c <= u'9') || (u'A' <= c && c <= u'Z') ||
+           (u'a' <= c && c <= u'z') || c == u'_' || c == u'-' || c == u'.';
+}
+
 }  // namespace
 
 namespace chatterino::linkparser {
@@ -233,6 +243,11 @@ std::optional<Parsed> parse(const QString &source) noexcept
             rest = remaining.mid(i);
             break;
         }
+
+        if (!isValidDomainChar(currentChar))
+        {
+            return result;
+        }
     }
 
     if (lastWasDot || lastDotPos <= 0)
diff --git a/tests/src/LinkParser.cpp b/tests/src/LinkParser.cpp
index 746486c9711..72748d6891a 100644
--- a/tests/src/LinkParser.cpp
+++ b/tests/src/LinkParser.cpp
@@ -1,11 +1,13 @@
 #include "common/LinkParser.hpp"
 
+#include "common/Literals.hpp"
 #include "Test.hpp"
 
 #include <QString>
 #include <QStringList>
 
 using namespace chatterino;
+using namespace literals;
 
 struct Case {
     // -Wmissing-field-initializers complains otherwise
@@ -91,6 +93,14 @@ TEST(LinkParser, parseDomainLinks)
         {"", "a.com", "?("},
         {"", "a.com", "#("},
         {"", "a.com", "/__my_user__"},
+        {"", "a.b.c.-._.1.com", ""},
+        {"", "0123456789.com", ""},
+        {"", "ABCDEFGHIJKLMNOPQRSTUVWXYZ.com", ""},
+        {"", "abcdefghijklmnopqrstuvwxyz.com", ""},
+        // non-ASCII characters are allowed
+        {"", u"köln.de"_s, ""},
+        {"", u"ü.com"_s, ""},
+        {"", u"─.com"_s, ""},
         // test case-insensitiveness
         {"HtTpS://", "127.0.0.1.CoM"},
         {"HTTP://", "XD.CHATTERINO.COM", "/#?FOO"},
@@ -168,6 +178,7 @@ TEST(LinkParser, doesntParseInvalidIpv4Links)
         "196.162.8.1(())",
         "196.162.8.1(",
         "196.162.8.1(!",
+        "127.1.1;.com",
     };
 
     for (const auto &input : inputs)
@@ -219,11 +230,22 @@ TEST(LinkParser, doesntParseInvalidLinks)
         "~~a.com()",
         "https://chatterino.com><https://chatterino.com",
         "<https://chatterino.com><https://chatterino.com>",
-        // invalid characters are still accepted (see #4769)
-        // "chatterino.com><chatterino.com",
-        // "https://chatterino.com><chatterino.com",
-        // "<chatterino.com><chatterino.com>",
-        // "<https://chatterino.com><chatterino.com>",
+        "chatterino.com><chatterino.com",
+        "https://chatterino.com><chatterino.com",
+        "<chatterino.com><chatterino.com>",
+        "<https://chatterino.com><chatterino.com>",
+        "info@example.com",
+        "user:pass@example.com",
+        ":.com",
+        "a:.com",
+        "1:.com",
+        "[a].com",
+        "`a`.com",
+        "{a}.com",
+        "a.com:pass@example.com",
+        "@@@.com",
+        "%%%.com",
+        "*.com",
     };
 
     for (const auto &input : inputs)