Unwanted injection into new GPT-4o model system messages by OpenAI

API Bugs
23

I’m seeing this in mini as well as gpt-4o-2024-08-06.

24

Yes, still gpt-4o-2024-08-06 as reported earlier…

in json_schema, for example…

{your system}

# Response Formats

## response_output_name

{"type":"object","properties":{"reproduction_string":{"type":"string"}}}

You are trained on data up to October 2023.

It seems while justification was provided for mini, we get to live with it in the larger model.

They’re giving away free input tokens, though. If you go by the API usage returned. :smile:

25

I’m getting this bullshit response in my function that translates a message between languages, when GPT is commanded to translate from the same language to the same language.

Instead of just returning the message, it retrns this message, weird bug.