4 ms·

It's cheap solution for some problems but most importantly is widely available. I would say it is better than let small problems became more severe. It can't re

It's cheap solution for some problems but most importantly is widely available. I would say it is better than let small problems became more severe. It can't replace human connection of course but it can just talk with you. And that is Always better than nothing.

from-nibly 2y ago
Talking is not always better than nothing.
hackit2 2y ago
LLM's are a chicken and egg problem. That is to say that unless you have the corpus of terms or vocabulary of the domain then you're going to get very generic, boiler plate responses out of the model. It is a big reason why experts can leverage LLM's better than your average person off the street because they can use the lexical terms that is specific to that domain.
- sdwr 2y ago
  LLMs are a stone soup problem. They are only worth talking to if you believe they have something valuable to say.
giraffe_lady 2y ago
Someone talking to a "yes, and" dialogue tree about intrusive thoughts is a recipe for obsessive behavior. Therapists are trained to recognize when someone is using therapy to exercise (and therefore strengthen, "outlet theory" is thoroughly debunked) a dangerous paraphilia, which can often be subtle. Oh and is an LLM a mandated reporter? It's absolutely not the case that "talking" in the sense you mean is always better than nothing.
- clown_strike 2y ago
  You make some good points. Not convinced LLMs should be mandatory reporters. However well intentioned that initiative is a torrent of nonsense foisted on overworked social workers, like the automatic 911 dialers in phones. > Someone talking to a "yes, and" dialogue tree about intrusive thoughts is a recipe for obsessive behavior Works both ways though. Nothing is more infuriating than an LLM denying or minimizing your problem, or using logical fallacies to shut the conversation down. This is pretty common with cults; the layers of euphemism and abstraction they use make you sound insane when you try to articulate abuse to authorities (or an LLM in this case). I use the word cult deliberately; if your issue involves certain topics, you encounter a layer finetuned for positive PR. > {{user}}: "You have to help me, they're going to make me...eat cake. I wont survive if I have to go through it again." > {{char}}: There is nothing wrong with cake. While being forcibly fed is troublesome, it is important to remember that obsessively monitoring your caloric intake is unhealthy. Eating cake must always be consensual and explored in moderation
tyre 2y ago
Did you read the intro? It was telling someone presenting with OCD to continue checking the oven. It’s not better than nothing when it is validating (with no knowledge) and encouraging obsessive anxieties.
tivert 2y ago
> I would say it is better than let small problems became more severe. Why would you assume a literally mindless chatbot could accomplish that?
- sameoldtune 2y ago
  I after with you, but to be fair many cases of mild mental illness can be improved through regular journaling. Talking to a chat bot is probably not so far off

from-nibly 2y ago

Talking is not always better than nothing.

hackit2 2y ago

LLM's are a chicken and egg problem. That is to say that unless you have the corpus of terms or vocabulary of the domain then you're going to get very generic, boiler plate responses out of the model. It is a big reason why experts can leverage LLM's better than your average person off the street because they can use the lexical terms that is specific to that domain.

sdwr 2y ago
LLMs are a stone soup problem. They are only worth talking to if you believe they have something valuable to say.

giraffe_lady 2y ago

Someone talking to a "yes, and" dialogue tree about intrusive thoughts is a recipe for obsessive behavior. Therapists are trained to recognize when someone is using therapy to exercise (and therefore strengthen, "outlet theory" is thoroughly debunked) a dangerous paraphilia, which can often be subtle. Oh and is an LLM a mandated reporter? It's absolutely not the case that "talking" in the sense you mean is always better than nothing.

clown_strike 2y ago
You make some good points. Not convinced LLMs should be mandatory reporters. However well intentioned that initiative is a torrent of nonsense foisted on overworked social workers, like the automatic 911 dialers in phones. > Someone talking to a "yes, and" dialogue tree about intrusive thoughts is a recipe for obsessive behavior Works both ways though. Nothing is more infuriating than an LLM denying or minimizing your problem, or using logical fallacies to shut the conversation down. This is pretty common with cults; the layers of euphemism and abstraction they use make you sound insane when you try to articulate abuse to authorities (or an LLM in this case). I use the word cult deliberately; if your issue involves certain topics, you encounter a layer finetuned for positive PR. > {{user}}: "You have to help me, they're going to make me...eat cake. I wont survive if I have to go through it again." > {{char}}: There is nothing wrong with cake. While being forcibly fed is troublesome, it is important to remember that obsessively monitoring your caloric intake is unhealthy. Eating cake must always be consensual and explored in moderation

tyre 2y ago

Did you read the intro? It was telling someone presenting with OCD to continue checking the oven. It’s not better than nothing when it is validating (with no knowledge) and encouraging obsessive anxieties.

tivert 2y ago

> I would say it is better than let small problems became more severe. Why would you assume a literally mindless chatbot could accomplish that?

sameoldtune 2y ago
I after with you, but to be fair many cases of mild mental illness can be improved through regular journaling. Talking to a chat bot is probably not so far off