Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

I'm becoming AI-blind

cymerys.com

Post reply on HN
Print view View original

371–380 of 533 posts

Re: I'm becoming AI-blind

#371

Post by nnevatie »

causal wrote: post #30

There's some psychological mechanism by which my brain immediately recognizes AI generated text and just short-circuits to "there is no information here". And when I force myself to read AI-generated text I realize I'm making my brain do creative work to impart meaning to the words. It is exhausting because my brain is literally trying to do a just-in-time rewrite of the text into something valuable. Something is dee…

> Something is deeply wrong with AI generated output

For me, it is the endless maximalism and hyperbole. Almost if the output was driven through a radio-mix compressor - too loud for the reader/listener to be able to pickup any dynamics.

Re: I'm becoming AI-blind

#372

Post by LtWorf »

neom wrote: post #192
blakesterz wrote: post #4

"There's an ongoing discussion of whether humans are good at recognizing AI-generated text. While most research claims that humans don't really do a good job there, I disagree. " I wonder if humans that spend all day working in tech are good at recognizing AI-generated text, but people who spend all day doing jobs that don't involve computers aren't as good. And I wonder if those of us in tech are the only ones who r…

Most people do not realize when a personal message they receive was written by AI, study finds - https://theconversation.com/most-people-do-not-realize-when-... People who frequently use ChatGPT for writing tasks are accurate and robust detectors of AI-generated text - https://arxiv.org/pdf/2501.15654

They might not be able to pin what the issue is, but getting bored and starting to jump paragraphs will happen anyway.

Re: I'm becoming AI-blind

#374

Post by 10dpd »

I've taken to asking the engineers on my team to just share the original prompt, no judgement because its frustrating having to read through pages of AI generated technical documentation. Another approach is to feed the AI generated content into AI... who knows where this will end.

Re: I'm becoming AI-blind

#375

Post by rjmunro »

10dpd wrote: post #374

I've taken to asking the engineers on my team to just share the original prompt, no judgement because its frustrating having to read through pages of AI generated technical documentation. Another approach is to feed the AI generated content into AI... who knows where this will end.

Wouldn't "the original prompt" be pages and pages of back and forth conversations? That's how I've use AI agents to make documentation.

Re: I'm becoming AI-blind

#376

Post by prophesi »

rcxdude wrote: post #212
prophesi wrote: post #195

Earlier quoted context omitted.

I'm guessing whether you believe it possesses intelligence or not depends on your answer to Searle's Chinese room thought experiment[0]. I'd also recommend checking out the Peter Watts' book, Blindsight. [0] https://en.wikipedia.org/wiki/Chinese_room

The Chinese room is a good Rorschach test for this kind of thing (but not a good thought experiment, IMO, because it's obviously correct or obviously wrong depending on where you're already coming from), but also it's not really about intelligence per se, but more abstractly awareness and more adjacent to consciousness than intelligence, and these are not the same thing (though it does seem like a lot of people have…

It all comes down to semantics. And yeah, with the thought experiment, Searle presents three axioms of what could constitute intelligence. Then proposes the chinese room thought experiment as an approval of his third rule, "Syntax by itself is neither constitutive of nor sufficient for semantics." Which only makes sense when you take the other two rules together. Oddly enough, defining intelligence with our capability to derive semantics.

There are a lot of creative counter-arguments to look into on the thought experiment though.

Re: I'm becoming AI-blind

#377

Post by xigoi »

red75prime wrote: post #341
ThrowawayR2 wrote: post #93

Earlier quoted context omitted.

LLMs are pattern prediction systems with a large training data set. It is not surprising that they can predict patterns, particularly for a well structured field like mathematics that is also amenable to automated proof checking to help steer it.

"Pattern prediction" is a very broad stroke. What's something AI can never do that would astonish you if it did?

Perform the same tasks as a human can, given only the amount of training data and energy that a typical human has access to.

Re: I'm becoming AI-blind

#378

Post by AndyMcConachie »

Having just a little background in traditional linguistic discourse analysis help understand why this is.

The best basic explanation I've found is this video.

https://www.youtube.com/watch?v=ORgKY9AlybA

The real answer is to take a class in discourse analysis and learn how to pick apart sentences. Once you start understanding how language works in the brain you can start to see better how LLMs fail.

Re: I'm becoming AI-blind

#379

Post by lukan »

causal wrote: post #30

There's some psychological mechanism by which my brain immediately recognizes AI generated text and just short-circuits to "there is no information here". And when I force myself to read AI-generated text I realize I'm making my brain do creative work to impart meaning to the words. It is exhausting because my brain is literally trying to do a just-in-time rewrite of the text into something valuable. Something is dee…

Isn't that mostly hyperbole?

When I ask AI to research technical information about X and (include sources) - I get mostly solid information as response.

But poems or interesting fluff blog entries by LLM's? Not something I look for.

What disturbs me is all the "pretending to be human" all the personalizing language - that is clearly fake and I would much rather have a neutral robot language as response.

Re: I'm becoming AI-blind

#380

Post by dsign »

xpct wrote: post #136
dsign wrote: post #52

I don't like AI-generated text at work, at all. It feels lifeless and unfocused. But what I hate the most is that it is objectively better than what I had before. No typos, clear structure, and, regrettably, the verbosity and autistic obsession with detail of the LLM is more actionable and useful than the human guy who wrote lists of commands and URLs as documentation, without explaining anything. Or the colleague wh…

I'd say my experience is different. Even from people whose communication writing I didn't find that useful, they seem to have a better frame of mind than LLMs do. Though, I never encountered people like the examples you gave.

You have great colleagues. I wasn’t comparing frame of mind of my colleagues and LLMs, which is a loaded notion anyway, but now that you mention it…
Post reply on HN