I think it really depends on the use case. Otherwise, OpenAI would not make you fill out a form describing what you are looking to use a fine-tuned GPT-4 for.
If you are a high-revenue generating Enterprise client, then this may work in your favour but as for the regular developer I think it is highly use case dependent. As OpenAI has stated itself, the fine-tuning process for GPT-4 appears to be more complex and requires more effort to achieve meaningful changes/improvements, so to me personally it makes sense that they take a case by case approach.