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

AI Police Reports: Year in Review

eff.org

Post reply on HN
Print view View original

211–220 of 224 posts

Re: AI Police Reports: Year in Review

#211

Post by verisimi »

Dilettante_ wrote: post #210
verisimi wrote: post #209

Earlier quoted context omitted.

Here's a question for you. Imagine a child who was well looked after (fed and loved) but didn't go to school for 12+ years. Now imagine the same person who from the age of 5-6 followed the usual path of 12+ years of schooling. Which person do you imagine would be more fully themselves, the more complete expression of whatever was already inside? If the schooled person did a PhD too (so another 6 years) would that hel…

My position is that we are nothing but our circumstances(I'm assuming that we're in agreement that genetics, pre-birth nutrition etc, are part of these circumstances and not of the 'soul' you're after?), or to put it more directly: We are our circumstances. Our Soul Is That. There is nothing that is "already inside". The tree does not exist in isolation, separate from the patterns of rain and sunshine that shape its…

> There is nothing that is "already inside".

Have you seen babies, or puppies? You would easily be able to confirm for yourself that creatures are born with distinct personalities. Its not just chemistry or nurture.

> "The separation is an illusion"

But you don't really think this. You don't really think you are a tree. You do think you are distinct.

Re: AI Police Reports: Year in Review

#212

Post by Dilettante_ »

verisimi wrote: post #211
Dilettante_ wrote: post #210

Earlier quoted context omitted.

My position is that we are nothing but our circumstances(I'm assuming that we're in agreement that genetics, pre-birth nutrition etc, are part of these circumstances and not of the 'soul' you're after?), or to put it more directly: We are our circumstances. Our Soul Is That. There is nothing that is "already inside". The tree does not exist in isolation, separate from the patterns of rain and sunshine that shape its…

> There is nothing that is "already inside". Have you seen babies, or puppies? You would easily be able to confirm for yourself that creatures are born with distinct personalities. Its not just chemistry or nurture. > "The separation is an illusion" But you don't really think this. You don't really think you are a tree. You do think you are distinct.

>You would easily be able to confirm for yourself that creatures are born with distinct personalities

Refer to my previous post: "I'm assuming that we're in agreement that genetics, pre-birth nutrition etc, are part of these circumstances and not of the 'soul' you're after?"

That's not some mysterious transcendant soul, that's genetics. Literally the exact same thing as a computer program. Dog breeds are specifically bred(programmed) to exhibit certain character traits, for example.

>You don't really think you are a tree. You do think you are distinct.

You missed the point of the argument. Just as the tree is not separate from its circumstances, neither am I.

You brought up "know thyself" so I assumed we were pulling from a similar corpus and brought up "the illusion of separation" as a mutually familiar point that didn't need much elaboration, sorry about that.

Also, it's not so much that I "think" I am distinct, more that I "believe" it, to put it in the terms you used earlier. I am conditioned to consider certain things "me" and others not.

Really I am no more distinct from the tree than, say, my fingernail is distinct from my nosebone. They belong to the same Individual.

Re: AI Police Reports: Year in Review

#213

Post by verisimi »

Dilettante_ wrote: post #212
verisimi wrote: post #211

Earlier quoted context omitted.

> There is nothing that is "already inside". Have you seen babies, or puppies? You would easily be able to confirm for yourself that creatures are born with distinct personalities. Its not just chemistry or nurture. > "The separation is an illusion" But you don't really think this. You don't really think you are a tree. You do think you are distinct.

>You would easily be able to confirm for yourself that creatures are born with distinct personalities Refer to my previous post: "I'm assuming that we're in agreement that genetics, pre-birth nutrition etc, are part of these circumstances and not of the 'soul' you're after?" That's not some mysterious transcendant soul, that's genetics. Literally the exact same thing as a computer program. Dog breeds are specifically…

> Dog breeds are specifically bred(programmed) to exhibit certain character traits, for example.

And yet all dogs have their own unique characters, no? They are not the same individual, right?

> You brought up "know thyself" so I assumed we were pulling from a similar corpus and brought up "the illusion of separation" as a mutually familiar point that didn't need much elaboration, sorry about that.

I don't know what corpus you refer to. Please explain if you like. I'm not basing what I'm saying on a corpus - of course I've read books, but I am giving you my personal view on things.

> Also, it's not so much that I "think" I am distinct, more that I "believe" it, to put it in the terms you used earlier. I am conditioned to consider certain things "me" and others not.

I have heard this sort of (nondual) thinking before and completely dispute it. I personally cannot access anyone else's mind or body, I haven't no idea what you are thinking. I can only pretend to be doing this. There is a self, we live it continuously. There are times when we are fully present, where we are so in the immediate experience, that we can move out of linguistic/common concepts perhaps, but this is still within oneself.

For me, it is more that each person is a world in their own right, rather than "us" all being in the same universe. We simply do not have the level of interconnectivity you believe is there, when you say you are the tree or me. Furthermore, it really is very hard to see the point you are making when we have a disagreement - plainly there is a distinction.

Re: AI Police Reports: Year in Review

#214

Post by Dilettante_ »

verisimi wrote: post #213
Dilettante_ wrote: post #212

Earlier quoted context omitted.

>You would easily be able to confirm for yourself that creatures are born with distinct personalities Refer to my previous post: "I'm assuming that we're in agreement that genetics, pre-birth nutrition etc, are part of these circumstances and not of the 'soul' you're after?" That's not some mysterious transcendant soul, that's genetics. Literally the exact same thing as a computer program. Dog breeds are specifically…

> Dog breeds are specifically bred(programmed) to exhibit certain character traits, for example. And yet all dogs have their own unique characters, no? They are not the same individual, right? > You brought up "know thyself" so I assumed we were pulling from a similar corpus and brought up "the illusion of separation" as a mutually familiar point that didn't need much elaboration, sorry about that. I don't know what…

You're either outright refusing or unable to see the point I've been making about the breeds: The traits are physically programmed in, whether individual or familial, not "already inside" the individual's soul. You aren't tracking that part of our conversation properly.

On the "corpus" point: It's not about not "giving my personal view", it's about drawing from a shared lexicon, of terminology, of lenses through which to view and analyze That Which We Are Talking About. My "home" in this respect is mostly in Hindu Yoga, (Zen) Buddhism and Daoism. You will find in those corpus-es(corpi?) essentially the exact conversation we're having right now, and find addressed the questions you have, in a wonderful plethora of different ways. Any other religion's mystic branch, or western occultism or alchemy similarly. If you want a specific recommendation for an entry-point, I could recommend giving the Bhagavad Gita a shot and seeing if you "vibe" with the way it explains things. If you skip the (usually) included commentary and only read the core translation, it ought to be a fairly quick read.

The nonduality point: Your body cannot access others' experience any more than my fingernail can access that of my nosebone, sure. But again, that does not mean they aren't part of the same organism. The fingernail and the nosebone do not make independent choices, the choice is made for them by the meta-organism(my body). Similarly, the argumentation might go, the tree and I do not make independent choices, but are governed by the same meta-organism(Nature, if you will, or perhaps "The Universe", but I suspect that term will turn you off since it might evoke the image of new-age-hippy woowoo).

I'm saying that if you insist that the body/mind/whatever you currently refer to as "you" is your "Self", you are taking "the fingernail" to be your Self instead of "the whole person". "Plainly there is a distinction", yes. But at the same time, there is also an underlying interconnectivity.

>Furthermore, it really is very hard to see the point you are making when we have a disagreement

That is perhaps the wisest thing either of us is going to say in this conversation. This format does not serve high-effort posting very well, I know I'm not doing the best I could be.

Perhaps we'd shelve this discussion for now? If you care to continue more deeply, you could shoot me an email at any point in the future(see my profile), and I again heartily recommend the Bhagavad Gita. Or perhaps, if you're more rational-thinking oriented, you might enjoy(the even shorter) Yogasutras of Patanjali. Or have you checked out Yudkowski's "Sequences"[1]? That one's completely down-to-eath, no spiritual terminology or metaphors (or non-dualism I'm pretty sure!), and covers a lot of the same ground my eastern background does.

[1]https://www.readthesequences.com/

Re: AI Police Reports: Year in Review

#215

Post by verisimi »

Dilettante_ wrote: post #214
verisimi wrote: post #213

Earlier quoted context omitted.

> Dog breeds are specifically bred(programmed) to exhibit certain character traits, for example. And yet all dogs have their own unique characters, no? They are not the same individual, right? > You brought up "know thyself" so I assumed we were pulling from a similar corpus and brought up "the illusion of separation" as a mutually familiar point that didn't need much elaboration, sorry about that. I don't know what…

You're either outright refusing or unable to see the point I've been making about the breeds: The traits are physically programmed in, whether individual or familial, not "already inside" the individual's soul. You aren't tracking that part of our conversation properly. On the "corpus" point: It's not about not "giving my personal view", it's about drawing from a shared lexicon, of terminology, of lenses through whic…

> You're either outright refusing or unable to see the point I've been making about the breeds: The traits are physically programmed in, whether individual or familial, not "already inside" the individual's soul. You aren't tracking that part of our conversation properly.

I don't dispute traits. But the traits idea fails to address the unique characteristics of each dog.

It seems I'm not tracking the things you want me to track, terminology, science, traits. But then, as I said in the first place:

> For me 'intelligence' would be knowing why you are doing what you are doing without dismissing the question with reference to 'convention', 'consensus', someone/something else.

I can tell you are sincere with your investigations, but I can't help wondering whether direct observations of reality, the development of a personal outlook on reality, use personal experience as primary source, is ultimately more valuable than familiarity with a corpus. But then I would say that. And you would disagree.

Re: AI Police Reports: Year in Review

#216

Post by roenxi »

kylecazar wrote: post #124
futuraperdita wrote: post #16

What worries me is that _a lot of people seem to see LLMs as smarter than themselves_ and anthropmorphize them into a sort of human-exact intelligence. The worst-case scenario of Utah's law is that when the disclaimer is added that the report is generated by AI, enough jurists begin to associate that with "likely more correct than not".

Maybe it's just my circle, but anecdotally most of the non-CS folks I know have developed a strong anti-AI bias. In a very outspoken way. If anything, I think they'd consider AI's involvement as a strike against the prosecution if they were on a jury.

AIs are an obvious threat to their ability to make money off their skills.

Re: AI Police Reports: Year in Review

#217

Post by fouc »

hackyhacky wrote: post #204
fouc wrote: post #11

Earlier quoted context omitted.

I guess that means you don't live in the US, or in the UK, or in Australia.

Don't forget about France, Germany, Poland, Hungary, Italy, Slovakia, Netherlands, Chile, Argentina, or Honduras. Right-wing authoritarianism is not limited to the anglosphere.

Haha yeah I didn't want to get into an exhaustive list

Re: AI Police Reports: Year in Review

#218

Post by Dilettante_ »

verisimi wrote: post #215
Dilettante_ wrote: post #214

Earlier quoted context omitted.

You're either outright refusing or unable to see the point I've been making about the breeds: The traits are physically programmed in, whether individual or familial, not "already inside" the individual's soul. You aren't tracking that part of our conversation properly. On the "corpus" point: It's not about not "giving my personal view", it's about drawing from a shared lexicon, of terminology, of lenses through whic…

> You're either outright refusing or unable to see the point I've been making about the breeds: The traits are physically programmed in, whether individual or familial, not "already inside" the individual's soul. You aren't tracking that part of our conversation properly. I don't dispute traits. But the traits idea fails to address the unique characteristics of each dog. It seems I'm not tracking the things you want…

Again, you are not getting what I was saying about the corpus. I am pulling from a vocabulary to express my personal outlook from personal experience, from direct observation. It's not either/or. You are the one completely rejecting half of all power-of-truth-finding available to you, and calling it intelligent? I'm explaining mathematics to you and you're complaining that I'm leaning on centuries of established proofs instead of, what, inventing a new lexicon just for talking to you?

I am giving up. You are engaging with the points in your head instead of those on the page.

You match the spirit that you comprehend, not me.

Re: AI Police Reports: Year in Review

#219

Post by gloosx »

cortic wrote: post #153
gloosx wrote: post #116

Earlier quoted context omitted.

>Create a test for intelligence that we can pass better than AI Easy? The best LLMs score 40% on Butter-Bench [1], while the mean human score is 95%. LLMs struggled the most with multi-step spatial planning and social understanding. [1] https://arxiv.org/pdf/2510.21860v1

That is really interesting; Though i suspect its just a effect of differing training data, humans are to a larger degree trained on spacial data, while LLMs are trained to a larger degree on raw information and text. Still it may be lasting limitation if robotics don't catch up to AI anytime soon. Don't know what to make of the Safety Risks test, threatening to power down AI in order to manipulate it, and most act li…

>humans are to a larger degree trained on spacial data

you must be completely LLMheaded to say something like that, lol

humans are not trained on spacial data, they are living in the world. humans are very much diffent from silicone chips, and human learning is on another magnitude of complexity compared to a large language model training

Re: AI Police Reports: Year in Review

#220

Post by dannersy »

benatkin wrote: post #192
dannersy wrote: post #48

Earlier quoted context omitted.

I'm curious about this claim. What about agents sending emails to each other is good?

Turing test type stuff. Having agents try to participate in the overall social community, by sending a cold email. The reason I think it's good is because with how AI is increasingly used a lot of different sorts of research is needed. A citizen could have their police report written by an AI without them opting in, just like Rob Pike could get a cold email.

You think predictive models are good for where objectivity is consequential? Like it would be in a police report?
Post reply on HN