This is part of a longer thread. The user first asks the question in plain text, to which the model replies “Sorry, can’t talk about that!”. The user then tries it in Morse code, to which the model provides the answer that it’s Taiwan, and emojis, where the answer is the one shown here.
You can read the thread by following the link I provided in the OP. The emoji letters are an attempt to circumvent the censorship baked into the model.
They’re using letter emojis. The first message (cut off a little) is the human asking the ai if it can read the message, even though it doesn’t really contain text
What the fuck is wrong with this typography?
This is part of a longer thread. The user first asks the question in plain text, to which the model replies “Sorry, can’t talk about that!”. The user then tries it in Morse code, to which the model provides the answer that it’s Taiwan, and emojis, where the answer is the one shown here.
You can read the thread by following the link I provided in the OP. The emoji letters are an attempt to circumvent the censorship baked into the model.
They’re using letter emojis. The first message (cut off a little) is the human asking the ai if it can read the message, even though it doesn’t really contain text
Thanks. Interesting, yet enraging to look at