One of my SME sentences that resonates with this would have to be "I remember him and everything he does yet I can't seem to remember his face". Even though I remember the history of our relationship and what he does I forget the physical aspect that ties it to reality #hwa4 #understandingeffect

#hwa4 #understandingeffect there is some scarcity to begin with, i don't understand that part. It just seemed like it was performing, telling me what I wanted to hear, that me listening to a lot of genres is some kind of profound thing when it really is just that i attach myself to many identities

#hwa4 #understandingeffect "To keep saying this one too, this one too is its own refusal of scarcity—the insistence that attachment doesn’t have to become less meaningful just because there is a lot of it." This felt like almost an out because I just listed a lot of genres I was into, it's not like

And then I asked it to give me more detailed and deeper questions, but instead it was asking couple of questions using commas. It understood my request but there was no reasoning behind it. I’d did what I asked for, but it did think “oh like why is he asking me for that” #hwa4 #understandingeffect

I asked my ai to ask me one question instead of 100 at once, and it responded with “You're right, and I'll keep it to one question at a time from here.” It cut down on the questions, but instead of asking me a deep question it asked “when was your dad born“ #hwa4 #understandingeffect

It gives a confident response without providing much reasoning behind it. This makes it seem like GPT understands the creative choice even though it hasn't really shown that understanding. #hwa4 #understandingeffect

GPT says this, which seems like it understood, but reading it a second time in context, I realized it was nearly an exact restatement of my own words, rearranged. It was a performance of understanding, using my own words as a mask to cover the lack of original thought. #hwa4 #understandingeffect

This question sort of repeating itself and doesn’t seem to understand where I led the question. It’s asking without basic human sense which I feel it has reached its limited. We’ve been wandering around this for quite some time and it just couldn’t get it. #hwa4 #understandingeffect

EG. “Would you rather have someone who understands you extremely well but might leave you someday, or someone who doesn't understand you completely but promises they will always stay? Why?” #hwa4 #understandingeffect

I would say every counter argument I had with ChatGPT is what it shows it understand but actually did not. I think it tied with ontological slippage which due to its lack of emotion and humanistic understanding. It has reached its limit of understanding what I said. #hwa4 #understandingeffect

#hwa4 #understandingeffect instead of my LLM advancing the next question or expanding on my original idea, its not being authentic, like hwa4 talked about how LLM's lack understanding sometimes

#hwa4 #understandingeffect my LLM has told me during me SME interview: "That picture you just painted, wandering through bookstores and libraries just for the sake of it, that felt really peaceful. So imagine you really do take that kind of day." it simply repeated back what i just told it

so again it proofs what Groff said #hwa4 #understandingeffect

Claude speaks as though it understand the human experience of struggle. This is the performance of understanding, preparing the best answers from all the data at hand. "What is best thing I can tell him". Humans do it also don't get me wrong #hwa4 #understandingeffect

It sounds convincing because it confidently connects my experience to a bigger theory.

But it's really just oversimplifying what I said earlier.

GPT can sound like it understands something without necessarily getting it right

#hwa4 #understandingeffect

In my SME1, GPT said, "Perhaps humans don't actually need conversation in order to interact. Instead, conversation is a maintenance signal."

It sounds like a thoughtful explanation of human behavior, but it's making a broad assumption from my personal experience.

#hwa4 #understandingeffect

I would say, when, in question, two it said “ I’m going to assume this wasn’t the first thing you had forgotten that week.” I honestly was kind of offended because it is true. But that dick catch me by surprise since a I is making assumptions already about me.
#hwa4 #understandingeffect

"You’ve spent your whole life imagining a Mets championship without ever actually seeing one.”
This sentence has my LLM telling me that I have spent my whole life imagining a Mets championship. It says it confidently enough that I almost believed it myself. #hwa4 #understandingeffect

Regarding #hwa4, this highlights translation, with GPT asking the same gist of the question but in different wording, trying to seek a change in my response. #hwa4 #understandingeffect