GOAL we want to achieve:
Hope to release a Chinese version of AIPRM,Includes prompt words in Chinese.
HOW we could do it:
Just convert all the existing prompt words into Chinese.
IMPACT for using AIPRM:
For those who like AICHAT’s tools but are not familiar with English is a hard need.
I am not an employee of AIPRM, but even I can see that’s not really a description of HOW AIPRM could do it.
AIPRM don’t get to just wave some magic wand and somehow, without effort, costs, translators, the interface all magically transforms into a perfectly working second version with a wholly rewritten interface in Chinese (is there still a distinction between Mandarin and Cantonese btw?).
No, HOW something like this could be done is to hire a whole new team of developers who in addition to having all the high skills needed to do the software engineering, are also fluent in both languages (with the much, much higher wage demands you’d expect of such extra skills). You’d need a team of developers, rather than one translator, because AIPRM has to adapt and re-code things very quickly anytime that OpenAI changes, updates, tweaks how the API works, etc.
Remember also that the interface is sometimes sized around where things fit. A button big enough for the way one language expresses something may be far too small for the way it needs to be said in another, or conversely, too large to look right.
AIPRM then need to hire additional Chinese fluent customer support staff for the emails from this whole new market that cannot speak English and if the software is only in Chinese expect customer support to have the same capability.
This will change in the future, I am certain. One of the great promises of AI lies in translation, in real time, and Translators are rated as one of the jobs/careers most exposed to the risks of being completely replaced by AI. But living in the ‘right now’ machine translation is still unreliable, creating mostly laughable, but sometimes truly dreadful and offensive results.
The costs of this look to me (not an employee) to be extremely high, and ongoing, for a market that is just one country rather than global, and that will probably also produce prompts that only serve a second market, not the one universal marketplace for sharing prompts envisioned.
Do you see any realistic and business-safe alternate paths around all those huge extra costs for only a marginal return?