I’ve found a bug in the Android app: code snippets are being represented in a false encoding (in my case cp1252 instead of UTF-8 for, but maybe dependent on the system locale). If I look up the same chat via the web, everything is ok.
Hi and welcome to the Developer Forum!
What’s the best way to check the encoding?
I know how UTF-8 looks like when misrepresented as cp1252 and that’s what I see in the ChatGPT Android app once a code segment is complete. This mostly affects inline comments since the code is mostly ASCII. It only affects the representation of completed code segments, that’s not before ChatGPT adds the closing `` tag.
Can you come up with a prompt that generates one of these inlines? I’ll posts a screeny from my phone and see if it’s a localisation issue.
A prompt you could use is:
Write some code in Python and include inline comments in Russian and Chinese
1 Like
Yeap! It gets messed up once finished. That’s what has to be reported to the Android app developers.
Cool beans, I’ve moved this to the ChatGPT bugs section, hopefully it gets a few eyes on 
1 Like
I am also experiencing the same issue with the Android app. The strange thing is that it displays Korean normally while generating the content of the code snippet, but after the generation is complete, the Korean characters become corrupted, as if the encoding has changed.
The same happened for me that in the website, code with ‘(’ and ‘)’ have them show up as normal, but in the chatgpt Android app, both are displayed as an identical Chinese character but when I do “select text” and copy it, the character shows up as squares upon pasting it, probably because it’s from an encoding I don’t have on my Korean phone.
