CircleID

More technically, it transliterates to Cyrillic characters that resemble "PY", but are actually distinct characters, with different Unicode positions.  Similarly, the old "USSR" didn't really transliterate to "CCCP" as widely believed by English-speakers, but to Cyrillic characters that resembled those Latin characters.  Computers have no problem keeping them distinct as they are completely different characters in their character sets; humans have more trouble distinguishing them, and this has been used in phishing scams.  (Even pure Latin character sets include some lookalike characters; depending on the font, an uppercase I, a lowercase l, and a number 1 might resemble one another, as do a letter O and a number 0.