dsubburam

Interesting. I think this also explains in part why I prefer CDs to a streaming service for listening to music.

Would you side with the adjacent statement "expertise is not the bottleneck"? Didn't we recently discuss "LLMs reward expertise"?[1]

I suspect Discovery Loop will have to hire experts in each area they are targeting, to supervise and prompt their system effectively, much like the Terence Tao conversation with ChatGPT the OP cited[2].

[1] https://news.ycombinator.com/item?id=49161518 [2] https://www.seangoedecke.com/llms-reward-expertise/

Article looks substantially AI written, and I find it difficult to get past that, especially on a topic like ethics[1]. We already have the following guideline for comments:

"Don't post generated text or AI-edited text. HN is for conversation between humans."

I for one will value a similar guideline for submissions. Maybe we can have a poll. @dang.

[1]maybe an AI written article is fine if it is e.g. documentation for some new gadget.

How about speed and speedometers in cars? For some reason I prefer digital readouts of speed while I prefer analog faces for watches/clocks.

> They are not vocal because any political activism is not encouraged in China . Check jack ma story .

This is an under-rated comment. "Nice" seeming places in Asia might be so because the governments tightly control the narrative and brook no dissent. Citizens end up minding their own business and become apolitical. Society looks neat and organized; but if you don't conform, you get hammered down.

I do not know what fraction of LLM-processed messages are due to posters not being confident in their written English. To the extent that it is meaningful, we can

a) be welcoming and indicate that we're OK with broken English (content and intent typically do get through), and/or

b) allow posting in languages other than English (does this even work right now?).

Above probably ahead of its time, but thought I'd just make the suggestion.

A lawyer told me they might have felt they had no choice because enforcing a no-AI policy might be impractical.

Maybe we can distinguish craftmanship from creativity. This case can then be cast as one of deploying creativity without embodying the traditional craftmanship (ability to play guitar, sing low notes). I don't see that as illegitimate, so long as no false credit is taken about the said guitar playing, low note singing.

Can an artist be good if they can't draw a good circle by hand? Yes. Except they can't take credit for the goodness of circles that appear in their work, if not drawn by them.

[Edit: "responsibility" -> "credit"]

Her recent article about her illness and how she lived through it was quite touching and very sad. On the New Yorker, and posted on HN[1], though it did not get much discussion here then.

[1] https://news.ycombinator.com/item?id=46024709

Music is an interesting case. You can't slow down the consumption of music (you have to let it play at the speed the performer intended), but you can dial up the attention you give it. Listening with headphones, eyes closed, and phone+doorbell etc. switched off would be close to max. Sitting at a live concert (I am thinking classical) is up there too, because you've given yourself permission to not think of/work on anything else in that time. For music, we can say that the default settings are too LOW.

And similar to the point OP made, you get more out of it when you attend more closely. And similarly, most music does not withstand this level of scrutiny.

Another way lawyers can get displaced is if the rule of law breaks down. Then, whether you can get, say, regulatory approval for a deal, will depend more on if you can bribe someone than on carefully drafting and negotiating the terms.

This makes it in the lawyers' interest to stand their ground when established legal principles are under attack.

Persuasion that happens in good faith is a two-way street. You explain your position, but also truly listen to theirs. If you are prepared to change your own position based on what they say, then you can hope that they might change theirs based on what you say.

If it is truly two way in this sense, including your best efforts to extract from the other party their strongest, potentially unexpected, arguments for their position and give them your due consideration, it shouldn't feel like manipulation.

I do not know how "general intelligence" is defined, but there are a set of features we humans have that other animals mostly don't, as per the philosopher Roger Scruton[1], that I am reproducing from memory (errors mine):

1. Animals have desires, but do not make choices

We can choose to do what we do not desire, and choose not to do what we desire. For animals, one does not need to make this distinction to explain their behavior (Occam's razor)--they simply do what they desire.

2. Animals "live in a world of perception" (Schopenhauer)

They only engage with things as they are. They do not reminisce about the past, plan for the future, or fantasize about the impossible. They do not ask "what if?" or "why?". They lack imagination.

3. Animals do not have the higher emotions that require a conceptual repertoire

such as regret, gratitude, shame, pride, guilt, etc.

4. Animals do not form complex relationships with others

Because it requires the higher emotions like gratitude and resentment, and concepts such as rights and responsibilities.

5. Animals do not get art or music

We can pay disinterested attention to a work of art (or nature) for its own sake, taking pleasure from the exercise of our rational faculties thereof.

6. Animals do not laugh

I do not know if the science/philosophy of laughter is settled, but it appears to me to be some kind of phenomenon that depends on civil society.

7. Animals lack language

in the full sense of being able to engage in reason-giving dialogue with others, justifying your actions and explaining your intentions.

Scruton believed that all of the above arise together.

I know this is perhaps a little OT, but I seldom if ever see these issues mentioned in discussions about AGI. Maybe less applicable to super-intelligence, but certainly applicable to the "artificial human" part of the equation.

[1] Philosophy: Principles and Problems. Roger Scruton

Also, a fountain pen can be held at a smaller angle to the paper. Unlike the other kinds of pens, it even tends to write better that way. I find the smaller angle more comfortable to hold.

> I’m still trying to understand what is the biggest group of people that uses local AI (or will)?

Creatives? I am surprised no one's mentioned this yet:

I tried to help a couple of friends with better copy for their websites, and quickly realized that they were using inventive phrases to explain their work, phrases that they would not want competitors to get wind of and benefit from; phrases that associate closely with their personal brand.

Ultimately, I felt uncomfortable presenting the cloud AIs with their text. Sometimes I feel this way even with my own Substack posts, where I occasionally coin a phrase I am proud of. But with local AI? Cool...

This parallels the debate about free will and determinism. If you were in the determinist camp, believing that all that one does was predetermined by prior environmental causes, could you still hold people responsible for their actions?

Hobart makes a convincing argument that you can: "Fatalism says that my morrow is determined no matter how I struggle. This is of course a superstition. Determinism says that my morrow is determined through my struggle. There is this significance in my mental effort, that it is deciding the event." [1]

i.e., he is a "compatibilist", thinking that you can believe in free will and determinism too.

If you find Hobart persuasive, time-blindness or no, it does make sense to reproach someone for being habitually unpunctual.

[1] https://philarchive.org/archive/HOBFWA

Let me tie it with:

ts;du

(Too short; don't understand.)

I didn't know of that paper, and thought the title was a riff on Karpathy's Unreasonable Effectiveness of RNNs in 2015[1]. Even if my thinking is correct, as it very well might be given the connection RNNs->LLMs, Karpathy might have himself made his title a play on Wigner's (though he doesn't say so).

[1] https://karpathy.github.io/2015/05/21/rnn-effectiveness/

Another way to say this is truth matters and should have primacy over e.g. agreeability.

Anthropic used to talk about constitutional AI. Wonder if that work is relevant here.

Unlike when focusing on a movie screen, your eyes have to turn inward to direct the pupils to converge at the physically near iPad. This can cause muscular eye strain (it does for me).

You can get clever and order a prismatic prescription that bends light out, so your eyes don't have to turn inward. I tried it too, but it gave me nausea.

The intended interpretation for clocks IMO is "we are now past this time".

Like a mile marker when you are running down a jogging track.

It doesn't matter if the granularity is one-day or even one-year, like the ball drop in Times Square on New Year's Eve.

I challenge what I read in textbooks and hear from lecturers by asking for contrary takes.

For example, I read a philosopher saying "truth is a relation between thought and reality". Asking ChatGPT to knock it revealed that statement is an expression of the "correspondence theory" of truth, but that there is also the "coherence theory" of truth that is different, and that there is a laundry list of other takes too.

> This raises important and difficult questions about how edtech vendors can handle risk management. The most likely denouement is that vendors will press schools to get even more consents from parents in a blizzard of back-to-school paperwork, a resolution with its own disadvantages. IXL will also need to monitor/audit its school customers to ensure the consents are obtained, another drain on its resources.

I understand the broader concern but the conclusion above is oblivious to another workaround: Just have your revenue model not depend on stuff that requires parents' consent, such as in the intro (“collected and monetized the data of millions of school-age children who used the IXL platform without parental consent,” in violation of the ECPA and state law.)

> The "world model" of a human, or any other animal, is built pursuant to predicting the environment

What do you make of Immanuel Kant's claim that all thinking has as a basis the presumption of the "Categories"--fundamental concepts like quantity, quality and causality[1]. Do LLMs need to develop a deep understanding of these?

[1] https://plato.stanford.edu/entries/categories/#KanCon

Tip: It is Dean Phillips. Not Dean Philips.

I made that mistake too recently.

The opinion says at the time the warrantless search occurred, one appellate court had already held "that no warrant was required in those circumstances" (p 42). Only a year after the search occurred, did another appellate court rule the other way.

This is the main argument that the search met the good faith exception to the exclusionary rule (i.e. the rule that says you have to exclude evidence improperly obtained). This exception is supported in the opinion (at p41) with several citations including United States v. Ganias, 824 F.3d 199, 221–22 (2d Cir. 2016)

> Translation: “The Self hidden in all beings does not shine forth, but it is seen by subtle seers through their one-pointed and subtle intellect.”

That doesn't look sound to me. If the "seers" are seeing "The Self", are they beyond and separate from "The Self"? If they do so with their "subtle intellect", is that intellect outside of "The Self"?

If affirmative, then "The Self" is something external to the seer, making the term a misnomer. And furthermore, there is something outside of The Self (that which is seeing The Self), which remains to be explicated.

How would you respond to the argument that the burden of proof lies with the other side, i.e, the side that says there is too little AI doom risk to worry about?

Agreed. Also citing Graham's "lack of experience in this area".

Even in the steelman case of the Hollywood celebrity promoting a cause they are clearly not qualified to evaluate, I think we ought not belittle the speech act itself. Better to just present your counter-arguments if you disagree.

If everyone felt free to share their opinions, the better the quantity and quality of ideas that surface after due examination.

I re-learnt how to write as an adult, after a hand injury, using this handwriting manual from the ~1920s[1]. Shows positions for holding your hand, arm, etc. as well as exercises for practicing the various strokes involved (for English).

[1] https://thepalmermethod.com/