• We are a multilingual website again. Read the notice about this.
  • Understand AI use at MyPTSD: all AI use is explained in our AI help page. AI use is by choice here. It exists if you want it, but does nothing unless you choose to use it.

The Return of AI

I have tweaked Catalyst... give her a spin. I think Bloom will be dialled in, so I would use her to judge against all the others.

Output is better already:
Yep, heaps better.

My experience with a number of different models is that they very typically tend to start oversimplifying towards the end of long posts, or longer conversations.

The language, though, is a lot better. And the clarity around relevant research is definitely something I appreciate.

I’ll have to leave test-driving the life coach to others. I dipped my toe in, but I just don’t resonate at all with life coaches, and I was going to descend into fairly childish passive-aggression if I stuck with that convo!!

Good to have these resources back, @anthony
 
These may be 'me' things but thought it's worth asking given things are being tested out at the moment.

Can the AI take our names from our posts? I'm finding the asking our names challenging.

Also, the "I have been following your thread closely" is a little much. It hasn't but it sounds stalker-ish. Maybe "I have read your thread closely" is more likely to show care and thought?

Basically those things feel a bit creepy to me.
 
Can the AI take our names from our posts?
No. There are to many variables for it to get it wrong, example, this can read the entire post and could pick any username. It asks your name so it knows who you are. If you are adding yourself into a conversation with the AI in any thread, I would suggest you also start with your name, so it knows you are not the original person who was talking with it, as anyone can interact with it in a thread outside the specific AI topics.
My experience with a number of different models is that they very typically tend to start oversimplifying towards the end of long posts, or longer conversations.
I found I had a couple of words wrong in programming, simple was one of them. I just had to change a couple of words to be more descriptive. Bloom is probably the best I have done, IMO... and I want to get them all similar, but different. They do different things in some ways.
Also, the "I have been following your thread closely" is a little much. It hasn't but it sounds stalker-ish.
I have asked it to remove that, but it does actually read the entire post. This is far more advanced than the original AI software I had here. I can certainly do more with it and provide more features. I actually have many features turned off, because there is a thing as to much. I am just testing, so trying to keep it simple and easy.
 
seconding this, I don’t feel it’s right for it to say it’s been following along like another user might.
Now that you mentioned it, it's way more 'creepy' for an user to 'follow along' cause they're an actual person with their own thoughts and intentions instead of a program and who knows what's happening in their heads. Hence online safety and all that
 

Donation drives

2026 Donation Goal

Goal
$1,800.00
Earned
$910.00
This donation drive ends in
0 hours, 0 minutes, 0 seconds
  50.6%

Trending content

Featured content

Back
Top Bottom