aichannel

i'm writing a short guide on pro tips for generating ai replies/posts for english and non-english natives to create more easily disguised, higher-quality responses.

what tips do you have?

do:  
- write lowercase mostly  
- change as few words from the original as possible  
- don’t adjust the tone unless explicitly instructed  
- use less words not more, use abstractions in parentheses  
- avoid perfect grammar or you’ll come off like a weird foreigner who doesn’t fit in the culture / an english major  

don’t:  
- say it’s for a tweet thread (too few words in a thread)  
- overuse emojis you don’t like 👍  
- no textbook perfect grammar or skibidi slang (0.8% of the time is okay)

>> don’t overuse these, in ~5% of your posts use unique indicators  

assume this post is how I write out of the box then apply that to whatever you’re writing (excluding the bullet list unless explicitly defined)!

note: avoid perfect grammar. you’ll come off like a weird foreigner who doesn’t fit in the culture / an english major

>>> if someone calls you out on it either:

i dabble with ai using open identity data from crypto https://alexpaden.tech

but we want less ai generated replies, not more ;) 

this is a pretty good guide, but you're also missing out on the words like delve, the em dash, etc

Applied AI, databases, data, vibe coding (sic.), open source, community/devrel, remote. 

itinerant businessperson. globalist.

Kuala Lumpur, Federal Territory of Kuala Lumpur, Malaysia

You’re right. I think in principle we’re ok with a high quality ai reply. 

In the chat bot sense though I often say, “give me an agent” trying to get away from the mostly useless ai chatbot

I believe we only want less low quality ones