Multilingual Again

[anthony]

What people have to understand about final AI programming, is that the more you give it, the worse it can be. The less you give it, the worse it can be. You have to find a balance and let it think for itself on some things, use best judgement, which is not always best, yet the same applies to humans. Why the similarities? Because it is trained on everything human documented on the internet, books, journals, etc. Anything we have written, it is using that information to form an identity and provide that information back to us.

 

[Movingforward10]

@anthony , I think I like that style of programming. I suppose the only way to find out is to try?
I do find the current one too agreeable and it's making me suspicious. The humans on here I trust more and challenge me more than the current AI.
For growth, we do need challenging. And I go back to: it is up to us how we engage with other humans on here (there are some I don't engage with) and AI (I can decide not to engage with the AI).
But also, how would the AI know how not to retraumatise? Is that a concept it understands?

 

[anthony]

Ok, you can test this in the testing topic, call @NoBS bot, you can copy and paste your diary stuff into it, as that is a member forum, for testing purposes and all.

 

[Ecdysis]

Ok, you can test this in the testing topic, call @NoBS bot, you can copy and paste your diary stuff into it, as that is a member forum, for testing purposes and all.

I've started a testing thread with it. After calling it initially with an @ tag in the first post, it responded, but then in the thread you have to either quote it or tag it with an @ again, every time you reply. If you just respond normally to the thread, it doesn't realise you are talking to it

 

[Ecdysis]

Also, just questioning the language again...In your testing thread @anthony it says stuff like

spill—what's the core itch here?

I still don't get why such teeanger speak is considered appropriate for trauma contexts or why the AIs choose that style when they've been asked to use simple language?

 

[anthony]

If you just respond normally to the thread, it doesn't realise you are talking to it

I have set it that way, the same as @ therapist is... you have to call it.

I still don't get why such teeanger speak is considered appropriate for trauma contexts or why the AIs choose that style when they've been asked to use simple language?

Tweaking is essential. Finding the right words and phrases takes time.

 

[Survivor3]

@anthony The atnobs ai is quoting someone else 's text stating it is mine. I never wrote that.

 

[OceanSpray]

Also, just questioning the language again...In your testing thread @anthony it says stuff like


I still don't get why such teeanger speak is considered appropriate for trauma contexts or why the AIs choose that style when they've been asked to use simple language?

To be fair, I am one who would appreciate the slang speech. When it’s formal, it feels “too” AI, too therapisty, and would put me off it completely. I like the informality.

 

[OceanSpray]

@anthony The atnobs ai is quoting someone else 's text stating it is mine. I never wrote that.

Because that whole thread was the other users whole conversation with it. . .

 

[Survivor3]

Because that whole thread was the other users whole conversation with it. . .

Well ok, I understand what your saying but it's a "members testing" thread and it can't differentiate between different members. Surely that's an issue?

 

[Ecdysis]

Well ok, I understand what your saying but it's a "members testing" thread and it can't differentiate between different members. Surely that's an issue?

No, you're supposed to start your own thread.
It's fine tho, you can keep that one now.

 

[anthony]

@anthony The atnobs ai is quoting someone else 's text stating it is mine. I never wrote that.

Yer, make sure you start a new post, otherwise it will read stuff you didn't post. I can set it to what it can see, and right now, it can see a lot.