Foreign word and negative meaning check

How the check works

AI reads each word of the mark, and the mark as a whole, in the world's major languages, including Spanish, French, German, Italian, Portuguese, Mandarin Chinese, Japanese, Korean, Vietnamese, Tagalog, Arabic, Hindi, Russian, Polish, Dutch, Turkish, and Hebrew, and in English slang. It looks for:

• Foreign wording: words from another language, their English translation (and transliteration for non-Latin script), and a draft translation statement.
• The same word with another meaning in another language.
• Split readings: the mark read as separate words, such as NOVA read as NO VA ("doesn't go" in Spanish).
• Sound-alikes: how the mark sounds when spoken aloud.
• Slang and offensive words hidden inside longer words.

It then runs web searches to confirm what it found. Each finding must be backed by a page those searches actually returned. Links that were not really found are removed, and a negative meaning with no confirmed source is not reported at all. Each meaning is rated by severity (offensive, negative, or awkward) and by how likely real speakers are to make the connection. Some famous examples, including the NOVA story, are mostly myth.

Why it matters

• Translation statement: the USPTO requires an English translation of non-English wording in a mark.
• Doctrine of foreign equivalents: foreign words are generally translated into English to decide whether they are generic or descriptive, and to compare them with other marks for likelihood of confusion.
• Negative meanings: since Matal v. Tam and Iancu v. Brunetti, disparaging or scandalous meaning is generally no longer a bar to registration. It is mainly a branding and market risk, especially for goods sold where that language is spoken.