Non breaking white space. There are a lot of characters in unicode that are used to get text to format correctly for Greek, or Korean, or english. The problem isnt any one language but having unicode work for every language makes it vulnrable. i.e

ψ N̸̹͐͝ő̴̤̗͔͘ř̴̙̅m̵̢̜͉͑̌͜͝á̶̅̓͜l̵͖̔͌͗̑͝ͅ ̸̙̺͉̬̬́t̴̛͈̓e̵͚̳̊̆x̷̤͌̐t̸̯̰̘͉̑̅͒̈́͌ ̵͓̼̤̃g̴͇̥͉̊͑́̚ò̵͕̪͎́̐e̸͓͗̀̀̕s̴̜̾̈́̃͘ ̷̦̪̥̤̅h̷̨̩̝̖̃̏͠e̵͕͎̦̽̀͗̄̏ͅr̷̟̳͈̋́̚e̴̟͘͜͠ ⛥

Although in this case its not unicode but a header in the protocol. The unicode thing was previous text sms bugs.