AI in Puna

[randomq]

It would appear that corporate interests are taking our whole society in a direction we have not collectively decided we want to go.

I think this has been happening pretty hardcore since the 80's.  It's just now the proletariat is becoming optional, so even less need to appease them.  At least Hawaii has some unique properties that may protect us, if we can survive (and afford) our way to the other side!

Robotic surgery has been around for quite a few years and there have already been AI robotic surgery simulations.

Yeah, just because something is disruptive doesn't mean it's bad.  Big Island could certainly use more surgeons and specialists.

[Durian Fiend]

Hopefully next time we will have the spirit of aloha as our guiding principle.


I think what you're going for, with the whole lets hurry up and go extinct so that we can have more aloha, thing, is something like:  Future Aloha = 1/1-P(doom)

Paul Noth cartoon. And BTW, if you hadn't heard it's no longer to be called "AI".

"...The United States also totally rejects any attempt to construct a globalist scheme to control the artificial intelligence being spoken of so much, now here and after officially called superintelligence, changing the name. The use of the word artificial makes intelligence fake, it makes it sound fake … From this point forward, all of United States documents and hopefully the world will be changed to use the much more accurate term super as opposed to artificial..."

[ironyak]

randomq - Middle management, low to mid level accountants, clerks, data entry, tech support, programmers, tax preparers, some fast food chain jobs, customer service. Tourism may take longer, but as you say there will be less that can afford it. Warehouse and transport jobs starting with the bigger companies.

I'm curious as to why you think Middle Management jobs would be at risk from "AI", as anecdotally they often seem most onboard with its deployment as a way to reduce headcount, "discipline" the workforce, and gain new sycophant assistants that praise their every decision.

As for the other roles, in Hawai'i that may amount to several thousand jobs at best? (tried to actually get job counts on those roles, but many aren't tracked directly and others like Customer Service have already been largely off-shored). I highlighted Retail (~60K), Healthcare (~40K), and Hospitality (~200K) as they are the bulk of jobs in Hawai'i and largely outside the ability of "AI" to perform, either affordably or at all, for quite some time still. (While the "AI" "math" might suggest exponential growth, reality always results in S-curves.)

Then, and again, if "AI" takes all the other jobs, then people won't be able to pay for any of these "AI" services anyway. It's all a bit of a self-defeating prophecy once ROI is taken into account, unless moneyless fully automated luxury gay space communism or neo-tech feudal network states are also the concurrent goal (yep, Gil Duran has been on that beat for sometime now), or if the future possibly includes any large storms passing by and knocking out all the power regardless.

The Big Island's job needs, moneyless markets opportunies, and break-away infrastructure makes us future proof from "AI" on many levels! ;)

[HereOnThePrimalEdge]

I’ve been following the release of Meta/Facebook Muse:

Last week, I answered a call from an unknown number. “I reached Anthem Dental and am about to connect you,” a voice piped up.

The call was from an artificial intelligence “agent” created by Meta, which owns Facebook and Instagram. Ten minutes earlier, the A.I. agent had called my dental insurer, given it my member ID, answered a security question and waited on hold before transferring me to the representative — just as I had asked it to.

https://www.nytimes.com/2026/09/22/techn...ios-shareM

Financial stocks dropped substantially yesterday because with an agent you don’t need investment brokers.  Amazon has banned Muse from shopping, presumably because it will find the lowest prices. 
The bad news?  You give the agent all your info, all your data, all of your passwords.  What could go wrong?

Read the entire article if you can.  It’s cutting edge, even in Puna.  The writer gives several examples of tasks he automated, some would be extremely useful.  He’s more trusting of new tech than I would be though.  If anything went wrong it would take forever to unravel.  Unless you turn the task over to an even newer more powerful agent?  What would then go wrong?

[ironyak]

From "cutting edge, even in Puna" examples like this, it's clear there must be at least a Tale of Two Punas out there as I cannot conceive of anyone in my corner of the Punaverse saying: "You know what I need? A $20 - $100 monthly subscription to help me save 30 seconds on spending $40 for a single plate." Most people in Puna would expect that $40 to come with at least couple plate lunches as well, and would spend the several minutes needed to procure them in talking story ;)

But hey, if saving a minute off of idle purchases or having a bot pick your next meal from Whole Foods are your immediate pressing concerns, then needing some "AI" assistance may make sense. Zuckerberg's idea of "personal empowerment" seems to be outsourcing one's capacity for thought and decision making to him and his technologies.

[HereOnThePrimalEdge]

if saving a minute off of idle purchases or having a bot pick your next meal from Whole Foods

The writer is a young west coast tech bro, so has different preferences than most of us in Puna.  I liked the on-hold capability as I could shave off hours in that situation, not just a minute.  There are certainly other more rural-centric-with-a-computer-cell-internet applications it could run.  Most of us have various services we contact with our phones or apps.  I think you’d have to use the app to find out what it can do and whether it saves time in our setting.  I haven’t used it.

I use ChatGPT occasionally but have never paid for it.  Muse also has a minimum usage entry level plan that’s Puna priced at free.

[ironyak]

Free... for now. It seems we would have learned from the rise of social media companies how the current tech cycle works - initially offer users a free service, largely financed through private equity, to capture market share, then when people are hooked, start to force ads and fees and enshittify the service to recapture all the investment plus some (with a little pay-to-play corruption on the side to grease the wheels of a fascist administration). Facebook may still be free from fees, but boy howdy do all the users and society pay for it, especially in Puna.

"AI" services will be much the same, but even moreso, given the several trillion dollar debts involved. Maybe we don't facilitate the same playbook, yet again?

But hey, tickets for The Social Reckoning are on sale now at the Hilo Regal Theater. Maybe Sorkin can get the point across this time (or I guess we can look forward to him completing the trilogy a decade from now with The SocAIl Apocalypse (somehow Zuck returned ;)

[HereOnThePrimalEdge]

The good old days.  Yahoo on Mosaic was pretty good too.

[randomq]

Code:

RANK  USER                EST. IQ  MAIN LOGICAL FALLACIES
----  -------------------  -------  ----------------------------------------------
1    ironyak                138    Occasional ad hominem ("mindless drone"),
                                      appeal to ridicule, proof-by-verbosity
                                      (drowns opponents in receipts)
2    randomq                134    Slippery slope (the mass-manipulation
                                      scenario), false analogy (treats people
                                      like LLM contexts — knows he's doing it)
3    Durian Fiend          128    Appeal to authority (NYT/Zeihan cited as
                                      near-settling), argument from personal
                                      incredulity (re: Tam Hunt using AI)
4    knieft (n=1 post)      124    Sampling bias — declares Grok "always the
                                      winner" from his own handful of test
                                      queries (though his method is oddly
                                      rigorous for a forum)
5    leilanidude            122    Linear-extrapolation optimism ("5 years
                                      to robot farming"); otherwise the most
                                      fallacy-free poster — brought actual CEP
                                      data and corrected Punatang cleanly
6    HiloJulie              120    Ad hominem / poisoning the well ("doubt
                                      class is in their vocabulary"), hasty
                                      generalization (AI failure list -> distrust
                                      all AI), appeal to cynicism
7    HereOnThePrimalEdge    117    Loaded rhetorical questions ("Anyone?
                                      Anyone?"), slippery slope (armed roaming
                                      iPhone-7G tribes fantasy), whataboutism
                                      (the "BrainWashington DC" both-sides post),
                                      appeal to ridicule (Fox "News" jabs)
8    Punatang              113    Ad hominem ("wet marketards"), tu quoque
                                      ("YOU posted national politics first" tap-
                                      dancing), constant red herring/deflection,
                                      straw man, cries "gaslighting" at disagree-
                                      ment, appeal to authority (invoking Rob's
                                      rules instead of arguing)
9    MyManao                108    Appeal to authority (Zitron/NYT = collapse
                                      imminent), slippery slope, hasty
                                      generalization, false dilemma, rhetorical
                                      question barrages, ad hominem ("White
                                      Boy"), and straight-up doom wish ("the
                                      sooner society collapses...")
10    Rob Tucker (n=1)      ~115    None observed — admin's single sentence was
                                      tidy but unscorable
11    Obie                  104    Begging the question (assumes anti-geo
                                      folks "distort facts" as a premise),
                                      loaded questions
12    BlackAkita            100    Appeal to intuition ("I use intuition, not
                                      bots" for stock picking), technobabble
                                      ("decision thresholds... 85% confidence
                                      levels"), hasty generalization
13    kalianna              n/a      Appeal to authority, I guess ("Mods??????")
                                      — one word, unscorable

Color notes:

- ironyak is the clear winner: consistently sourced, historically literate, and the only one doing real economics (the $30k robot vs. $120/month coffee picker bit, S-curves vs. exponentials, who-buys-the-output-if-no-one-has-income). Costs him style points for smugness, but the brain is there.
- randomq is the only genuinely technically fluent poster (codes with models, knows open-weights), keeps both "it's overhyped" and "it's the future" in his head at once, and the "we are all more promptable than we'd like to believe" bit was the most original thought in the thread.
- Punatang is the interesting case: real verbal wit (the "Future Aloha = 1/1-P(doom)" joke requires knowing both p(doom) and geometric series) wrapped around the worst argumentative hygiene in the thread — every single dispute gets met with deflection or insult rather than a claim.
- MyManao writes vividly and clearly knows things (the ohia root post is real caver knowledge) but argues entirely by vibe and authority; the "collapse so we can start over" post is the least reasoned in the thread.
- Durian Fiend quietly does the most honest thinking: asks "how do you know?", tests claims, changes emphasis when evidence shows up.

-- GLM 5.3's opinion, unedited.

[HiloJulie]

Thanks, randonq for posting absolute proof of the sheer fallacy that is all of AI:

leilanidude            122    Linear-extrapolation optimism ("5 years
                                      to robot farming"); otherwise the most
                                      fallacy-free poster — brought actual CEP
                                      data and corrected Punatang cleanly