There had been growing criticism around the way the united National Organ Sharing Network allocated available organs. I have absolutely no idea if there was any basis to this. The one guy I know who underwent an organ transplant ended up getting assed out of one because VP Biden flew in and shut down medivac transport which meant the kidney and pancreas he was due ended up going to Portland instead. Then he eventually got a kidney and pancreas and proceeded to cook that one off in about eight years by failing to make the lifestyle changes necessary to maintain a donor kidney and pancreas. Mickey Mantle famously received a transplant one day after needing one only to die within months of cancer so there's going to be controversy around any attempts to distribute life-saving medical care in a truly egalitarian basis.

That criticism came to a head about a year back and then we suddenly had to have an opinion about fucking RFK. I have no idea how this feeds into that but considering RFK's basic approach to healthcare is "toughen the fuck up" nothing would surprise me at this point.

My father once observed that medical care in the United States is tilted heavily towards emergency medicine. There's nowhere in the world you'd rather be dealing with a gunshot wound... and you'd rather be almost anywhere else to be dealing with diabetes, for example. As there are far more people with diabetes than people getting shot (for now, at least), we end up allocating our spending rather poorly. Need surgery because your pancreas is kaput? We gotchoo fam. Need low-grade intervention to maintain your quality of life? bitch we have political fights about the constitutionality of capping insulin prices.

The only country I know of that allows the sale of human organs is Iran, and Iran does not allow non-Iranian citizens to participate in the market as buyers or sellers. The prevalence of black-market organ transplants is pretty opaque; I can't find a study done more recently than 2008 and it was basically a web search.

If you go away by Terry Jacks

It is inspired by "Ne me quitte pas" (dont left me) by Jacques Brel (one of the top 3 french lyricist , the other two being Georges Brassens, and Leo Ferré) .

This song is one of his most iconic song but mostly considered a bad one (he wrote it in a night after a break up.. and it is overly dramatic ).

Terry Jacks make a so much better version (the lyrics and ton are more subdued)

And better even than the Neil Young version that preceded it (by speeding it)

I wish Jacks kept more of the lyrics from the Neil version, because the song need to last longer.

I have so much sympathy for anyone who has to work at a medium-to-large organization. The inhumanity of the hiring process is just numbing to me.

We're a tiny little firm so our process is pretty straightforward - we headhunt people with posted resumes that look like a good fit, we call them up, and we get a vibe-check out of everyone in the office when they come by to interview. there's no "second interview" crap, it's all about "can we work with you and do we click."

I was f'n done with sarariman bullshit in 2007 and everything I see tells me that was the goddamn golden age by comparison.

Any news on the interview, c_hawk?

About a month ago I interviewed for a job that I would've thought I had a decent chance at getting. I was told a few days afterwards that I hadn't been successful. A shame, shrug, move on. About another week after that I learned the person they tapped for the job had rejected the offer and they were reaching out to other applicants to see if they were interested. I wasn't among those they reached out to.

Is what you're referring to connected to this?

Edit: Updated link.

Unless you are Peter Thiel, blood boy, oligarch tier wealthy, mostly, you would get a new heart, liver, lungs etc through the United National Organ Sharing Network. This network is made up of the transplant capable hospital networks in the United States plus some ancillary bodies for testing and coordination of resources.

This system is likely to face direct cuts, plus the indirect damage being done to it by the attacks on Medicare and Medicaid that damage the participating hospital systems.

It doesn't matter if you have the best of the best insurance if the hospital simply cannot locate an organ for you from a suitable donor. Money by itself won't locate an organ in a timely manner until you are just buying urchins off the street or paying for organs from political prisoners directly.

Is it that organ transplants and subsequent care are no longer available, or that their availability is becoming increasingly restricted only to those with the ability to pay the cost of the procedure and care? The impression I've gotten about US healthcare (without having benefited from it directly) is that the quality of treatment can be extremely high, but it comes at a cost that makes it impossible for many.

Or is the advice these patients are now receiving (i.e. to look overseas for treatment) the effect of some new horror? And where are they being told to go? Dental isn't currently publicly funded down here in Oz (except for kids), so some people fly to Thailand and Vietnam for orthodontic work (cheap flights means it can work out cheaper overall for big procedures, plus you combine it with a holiday). But I can't imagine either of those nations encouraging organ tourism.

The Strongest Diesel Saver in 39 HP Range for All-Around Indian Farmers

Farmers need a tractor that provides the right combination of power, reliability and cost effective in India's based agricultural world. Ultimately, the Mahindra 395 DI is a solid solution for midsize farmers who require a versatile machine that helps to make many agricultural works easier.

https://khetigaadi.com/new-tractor-model/mahindra-395di/en

The Strongest Diesel Saver in 39 HP Range for All-Around Indian Farmers

Farmers need a tractor that provides the right combination of power, reliability and cost effective in India's based agricultural world. Ultimately, the Mahindra 395 DI is a solid solution for midsize farmers who require a versatile machine that helps to make many agricultural works easier.

https://khetigaadi.com/new-tractor-model/mahindra-395di/en

"you're not having a breakdown. You're having a breakthrough."

I've tried to get several different AI's to write Open Office calc code to roll initiative for a D&D game where you enter all players name, their bonus to init and put it in order. We are rolling every round.

This seems like it shouldn't be that tough but ChatGPT, Gemini and what ever Microsoft is calling their Bing AI have all fucked it up everytime and when asked to fix it just make the code 10% longer while breaking things in new ways.

All the AI's are cocksure of themselves going into it and by the end they all say something along the lines of "well...Open Office is just wierd"

Not that AI cant be cool and fun but come on...write a damn initiative program.

I'd suggest picking up a variety of solder types and a full set of tips for your model if you go down this path--just use the right tool for the job. Sure, you could replace a USB-C port with 0.7 mm solder and a chisel tip, but why on earth would you, when there's 0.25 and a needle tip right there in the box?

Some dipshit with soldering tips.

One good thing about it is that it seems to be much improved in its sycophancy and in its safety boundaries compared to 4o especially, which is still the default “chat” for most people. So I’m hoping that means AI psychosis is down too before that gets any more outta hand because 4o loooves SCP. OpenAI claims they’ve made significant improvements to their training data processes, which I’m reading as “maybe it wasn’t such a good idea to blindly scrape all text we could find after all”.

I do feel like I need some more testing and benchmarks with it to judge how reliable it is. Maybe make my own benchmark. But better instruction following is much welcome.

In fairness to GPT5, in my career I have indeed encountered PhDs with this level of commitment to their particular blueberry

Where can I sign the petition to make this a new aphorism

I have yet to find an application for AI in a live business environment that is useful. For something that’s now a couple percent of us gdp you would think there would be some sort of Meaningful business use but nope. I think crypto was more useful than AI and that still turned out to not be very useful

https://kieranhealy.org/blog/archives/2025/08/07/blueberry-hill/

Interview round 2 I think went well! Hoping for the best. They let me know I was in the top 3 which is nice. Their last one is Monday and I'll be hearing a few days later an outcome.

Also, we have picked a wedding date! Photographer and reception venue have been chosen as well, only need to get ceremony location which opens 1 year out. After picking a date and confirming our reception location we flipped from anxious to excited lol

There are some optimizations to get a result better than quadratic. For example when you start to hit your context window you can have an in-between step to compress the tokens down before proceeding. And my agentic coding tool of choice doesn’t send the entire codebase every time, instead it picks which files to send along each time I ask something.

But those optimizations in practice mean you can ask a few more questions before hitting the Wall of Stupidity where every AI model will get stuck in some kind of loop or thought pattern or solution. It won’t make the law go away.

"Look, it's just arithmetic" is a pretty good way to get my vote.

I did not realize that a multi-step AI query must be run through from the beginning each time. I should have? Because clearly the algorithms are stateless. But that whole quadratic expense thing is a real pisser, particularly when what they're all doing is creating chatbots.

A chatbot that doesn't remember what it just said is barely better than ELIZA.

godfuckingdammit

You know how sometimes you read something so stupid that so many stupid chinstrokers just sit and stroke their chins over and everyone ignores the fact that we're all stroking our chins over something so fucking stupid that it fucking wakes you up at night?

Speaking about AI in the classroom, OpenAI CEO Sam Altman has described ChatGPT as “a calculator for words.” This analogy indicates the magnitude of change that ChatGPT is poised to bring about—imagine how radically math class must have changed when calculators became widely affordable—but it also indicates that change itself, even radical change, is not necessarily scary. Most AI skeptics would admit that math class survived the advent of the calculator.

So my knee-jerk reaction was this dumb-as-a-sack-of-hair "imagine how calculators changed math class" canard that Mister Chicken just gently waved his hands over must be a misquote because fuckin' hell had Sam Altman actually said that surely someone would have eaten him alive for it. So what did he actually say? Here it is at 15:14:

"What do you tell educators what are misconceptions of what you're working on how can you kind of allay their concerns?"

"We adapted to calculators and changed what we test for in math classes, I IMAGINE"

what do you tell people concerned about plagiarism, allow me to phrase this in the most softball possible way

I tell them to suck it

____________________________________________________________

Let's start from a "changed what we test for" perspective because this mfer right here grew up with a casio melody 80 as the only interesting thing to play with as a small child, barreled right the fuck into trig tables and was probably in the last class in the school district to be instructed in the use of a slide rule.

Yer goddamn right. One year we were being taught how to use a slide rule, the next year we were being encouraged to buy TI-81s. This, of course, was an easy fifteen years after every fucking parent you knew had bought an HP-35 because it was a nuclear weapons lab after all. So like... we were being taught trig tables two entire goddamn decades after the ability to go home and mash the sin button solved that shit because you know what? Trig tables are what the problems were written around, and as soon as you could get a TI-81 for, I think, $90? they started dropping the chapter on how to read trig tables. Eventually. Took years. They were still in the back of the book ten years later. Thing is tho your approach to math does not change appreciably whether your answers come from a calculator, a slide rule, a bunch of trig tables or brute-force calculation. What changes is your source of error and your methodology for computation.

THE ANSWER IS THE FUCKING ANSWER.

Gather 'round children while I share the tale of the floating point bug. You see, long after the bleeding edge was on their fifth, sixth or seventh computer but before hippies started making kitsch out of AOL CDs, the world was shocked - SHOCKED! to discover that one in nine billion computations might, might! fuck up in the fifth decimal place. This of course cost Intel half a billion dollars because computers aren't supposed to fuck up.

Those were the halcion days when Team Eternal September were juniors in college, though. When popular conception of computers had gone from Tron to Heartbeeps to Hackers. When having some knowledge about computers was cool rather than a reason to accuse people online of being Russian hackers. Fuckin' chatGPT sucks balls at arithmetic. Look at this mealy-mouthed legalspeak:

The answer is "approximately" to the eighth decimal place because OpenAI knows their shit sucks ass at math and if they just wing out to a goddamn calculator every time they might miss a chance to give you the answer in the form of a dragon or some shit.

_________________________________________________________________

The first time I ever heard the phrase "live my truth" was when a sociopathic liar on my TV show was caught in a sociopathic lie and when he was asked about his sociopathic lie he responded that he was "living his best truth" thereby implying that it wasn't that he lied it was that there is no truth, there are no lies, what does reality even mean, maaaaan and everyone was too polite to go "you're a fucking liar" because they were fucking simps. Not "here's my official class photo with a chicken" simps but simps nonetheless. This is the reason STEM kids will always and forever be fucking merciless towards any dipshit with a liberal arts degree: sometimes the answer is the answer. These are not the same:

And the difference killed 114 people.

So the answer is the answer is the answer except in liberal arts where the answer is a subjective performance in response to prompts that is graded and judged on largely subjective standards. Professor Chicken is all about whether a freshman who has never gotten laid can write a better essay about a snowball fight than a robot without even beginning to grapple with the difference between the subjective evaluation of creative writing and the objective evaluation of mathematics. Primarily because Saltman told him it was okay to do so. Kind of. Not really. Saltman actually told him to STFU but he's a fucking simp so he took that to heart and wrote a lesson plan whereby freshman can burn off one of their English GREs feeding tokens to ChatGPT.

And this mfer is so far up his own ass that he can simultaneously say "the upsides for school districts and colleges are clear" and quote one of his students as saying "Reflecting on the fact that 3 credits at UVA costs me $5000 and 2100 minutes, I do not believe I grew enough through this course for it to be worth it.”

THERE'S NO FUCKING SYNTHESIS HERE

It's fuckin' Sam Bankman Fried logic:

What are the odds that Shakespeare is any good? the math says he sucks so why should I read him?

__________________________________________________________________

Make no mistake - the Eloiification of the human race is going to have winners and losers. wow, ChatGPT wrote a marginally better essay than a Freshman english student, time to tune in the Kick-Me-In-The-Balls channel. The people who can ignore the fishing lure are going to eat everyone else for lunch. "Nathan?" "Drew?" They got everyone else's number. They recognize the sham for what it is and have moved the fuck on.

I wonder how much carnage they will leave in their path because this chucklefuck doesn't even realize how fucking stupid he is.

I love how level-headed this article is. The dust will settle and the sooner it does the better.

This song is the iconic soundtrack to an iconic rollercoaster and I love it.

Other than that I’ve been listening to the Hadestown musical on repeat because we went to go see that last week and it’s great.

—imagine how radically math class must have changed when calculators became widely affordable—

ಠ_ಠ

Too bad there wasn't anybody there! With the ability to document it! Or give you an oral history at the drop of a hat! Yep, we're all dead now. or we have Alzheimer's.

At the beginning of the semester, I asked my students to complete a baseline survey registering their agreement with several statements, including “It is unethical to use a calculator in a math class”

This is like talking to a virgin about sex. What are these... cal-cue-lay-torrs you speak of? What eldritch magic do they perform?

In my admittedly small sample, Altman’s analogy didn’t hold up. Calculators were uncontroversial: across my 72 students, one agreed that it was unethical to use a calculator, five chose Neutral, and the rest either disagreed or strongly disagreed.

Apparently none of them have been in a fucking math class?

But plenty of people do things that they believe to be unethical. In my next question, I asked students to indicate, anonymously, whether they had previously used AI in for-credit writing assignments. They confirmed that they had used it for editing first drafts (22%), outlining (28%), interpreting prompts (38%), proofreading (50%), and brainstorming (56%), with smaller pockets using it for finding sources or writing first drafts.

I'm sure you can trust those numbers although fuckin' even the studies on self-reported ethics have been retracted

It’s increasingly uncontroversial to use AI to brainstorm, and to affirm that you are doing so: just last week, the hosts of the New York Times’s tech podcast spoke enthusiastically about using AI to brainstorm for the podcast itself, including coming up with interview questions and summarizing and analyzing long documents, though of course you have to double-check AI’s work.

well fuck if Kevin Roose says it's okay

The authors point out that most people, even if they’re not chess fans, have heard of Deep Blue, the chess-playing machine that beat World Chess Champion Gary Kasparov in 1997;

Are these "most people" in the room with us right now? Do they wonder if calculators are banned in math class too?

In the following class, I had my students consider a study, covered by an NPR story from 2024, that looked at the effects of AI on creative writing.

Holy shit he sources his knowledge from NPR! Who knew! Also it's a study on the effects of AI on amateur creative writing. If you suck, AI might make you suck eight whopping percent less.

You’ve probably guessed where this is going. Max revealed, with a smile that didn’t quite conceal his dismay, that the girl did not exist, because the first paragraph had been written by ChatGPT.

The classroom erupted in a hubbub of disbelief. I was as shocked as anyone. My ability to spot AI-generated text had until now proven so reliable that it wasn’t even a point of conscious pride, just another flavor of the disappointment I feel when I start reading bad writing.

The point is both paragraphs ARE BAD. the goal is to make writing that is NOT BAD.

When we talked about it, we reflected on the crucial efficacy of the romance plotline.

something something virgins something something sex

More than any single line of prose, it was the girl that had taken us in. She was so beautiful in her vagueness: the snow flecking her hair of unspecified color and texture, the frisson of erotic worldliness that comes from her being older than our narrator, and of course her “kind eyes.” Perhaps we were so deeply programmed by the rom-coms we’d watched that we’d mistaken a rom-com for reality.

Perhaps you're a bunch of incels

In conversations about AI and education, it’s less common to hear about instructors using AI for writing lectures, designing assignments, or grading.

it is literally the only fucking thing you clowns talk about

Some students have mixed feelings about the idea of receiving AI instruction or feedback—one student at Northeastern petitioned unsuccessfully for a tuition refund on the basis that her instructor had used AI—but the upsides for school districts and colleges are clear.

are they tho

Frankly, in the era of DOGE, I’m surprised we haven’t heard more about replacing the left-leaning cadres of public university faculty with cost-efficient, “ideologically diverse” chatbots.

And here i thought you listened to NPR

I didn’t realize how irreplaceable I’d believed myself, how like a John Henry of the networked Humanities, until my students shared their findings. Yes, the majority preferred my feedback—it was noted that the AI models demonstrated an unhelpful fixation on “improving transitions,” whatever that means—but even my strongest advocates noted that their AI tutors often gave advice similar to mine, and faster.

My niece asks ChatGPT to be her therapist every day. It does a better job than her friends. Her friends are thirteen. Who do you think she'd rather hang out with tho

In each of these seminars, we had two instructors instead of one, who came from different disciplines: our Medieval colloquium, for example, featured a historian of early modern Rome alongside a softspoken Platonist.

how did they feel about calculators

If you accept both of these use cases—if you believe that students and faculty alike can and should use AI—you quickly encounter a scenario that most people would find logically abhorrent: teachers using AI to evaluate and grade AI-generated “student” writing.

Writing students would learn more from other students' critiques than from ChatGPT fight me

Back in 1998, for example, faculty and academic officials panicked about the rise of the internet, expressing concerns that seem both quaint and prescient.

Reader, they did not. They taught everyone how to cite a URL and moved the fuck on. The assumption, then as now, was that teachers would be better sussing out sources of plagiarism than students and if they weren't, that's on the teacher.

Perhaps ChatGPT has simply democratized this venerable tradition of cheating, thereby reducing the moral trespass we indicate when we use the word “cheating.”

Perhaps it just does it worse and for free

While some students from “different childhoods and levels of education” might need help writing at the college level, Nathan explained that he’d had “an excellent education up to this point,” for which reason he took the “difficult and dangerous” view that “I do not believe that students of The University of Virginia, a top 3 public school in the country, need a first-year writing course such as this one.”

I see you Nathan

I suppose I feel obliged to correct for the fact that some students might have voted yes simply to spare my feelings; I admire Nathan’s and Sam’s bravery for saying all this to my face, as it were.

Bitch I wrote an essay about why the entire fucking class should be abolished and the chair be fired and the feckless grad student who subjected us to this bullshit broke down crying and pleading for her job. Y'all are pussies.

And if you have access to an electric drill, why would you insist on using a screwdriver?

We discovered over the weekend that none of our employees under 40 know how to grill a hamburger. Nor can they be taught. They must all be busy teaching freshman creative writing at UVA.

In the final essay prompt, I’d invited my students to compare my course to learning “to start a fire with flint and tinder in the age of matches and propane lighters”: was this analogy accurate?

Depends - do you have a drill or a screwdriver

Of the four students who argued that the course wasn’t necessary, another took up this analogy directly. “Reflecting on the fact that 3 credits at UVA costs me $5000 and 2100 minutes,” Drew wrote, “I do not believe I grew enough through this course for it to be worth it.”

you and me, Nathan. You and me

Other students disagreed with my analogy. “The analogy is flawed,” Dishi argued, “for unlike fires, all writing is not created equal."

my god it's full fo stars

Carina, a ROTC student who often attended class in full camo, wrote that “there is a reason people still learn to build a fire that way, in case of emergency with no resources.”

The gentle art of subtlety

In my admittedly small sample of 72 students, I noticed that the students whose essays expressed the strongest doubts about the course, whether or not they voted no, were all men. I didn’t have the opportunity to ask them about this, but I can speculate along identitarian lines as to why my brethren felt this way.

Fukkn.... brethren in the Year of Our Lord 2025

let's see your pince-nez

When I pointed out that the joke he intended would have required an “aha” moment where he told the reader that the text was AI-generated

Right, like all parodies do, like when Johnathan Swift said 'I'm not actually telling you to eat the irish"

As Misha’s essay indicates, writing about “the power of writing” contains its own stock phrases and brainless clichés.

Take it from someone who mixed over two thousand hours of reality television - humanity thinks in stock phrases and brainless cliches

Writing, wrote Zoey, “is a way to express something that you cannot verbally say out loud,” which made it “a subject as rigorous as science. Everyone can speak, but not everyone can write.”

Fuckin' lol Zoey snowed you and you're too busy sniffing out AI to notice she clapped your ass in a platitude

If ChatGPT were to read Cam’s essay, I doubt it would pause at this line. But her words have lingered with me because Cam spent the last month of the semester on crutches, so I don’t think she used the word crutch lightly.

I have never had to use crutches myself, but I saw Cam struggling to walk, and saw her accepting help from Sam (“I would still have been a capable writer without this class…”), who assisted her in getting to the elevator every day after class. Sam and Cam met in my class, but during the semester, they discovered that they attended the same church; by the end of the semester, he was planning to accompany her on a group trip to the beach.

Fuck d00d we get it

Also this was written by AI; I can tell from the semicolon

Maybe the real treasure is the friends they make along the way.

How will I know irony without a hashtag send help plz

_________________________________

Edited to add this mfer's official faculty photo with a chicken

AI Capex is Eating the Economy

The strongest argument for being able to get down there came from an MIT prof I listened to on a podcast a while a go, who believes the "reasoning" paradigm is the fundamental discovery we need to get to some kind of jobs-market-destroying AGI.

That "reasoning" paradigm used to be called "AI." It's recently shifted to being called "symbolic AI." And the shitty thing about "symbolic AI" is it's been five or ten years away since ELIZA. It doesn't scale. Throwing chips at it doesn't make it better. Throwing clock cycles at it doesn't make it faster. Throwing money at it doesn't make it more useful. This is why it was tossed aside for LLMs - when we were way back on the asymptote they appeared to scale like gangbusters! Even now you can get impressive results out of tiny models if you're willing to wait. And are willing to put up with them making shit up.

I realized recently that my primary disappointment with AI is how fucking boring it is.

What the fuck am I looking at? I honestly don't know. And that's the best part. AI used to surprise you. It used to show you things that had never been there before, made from things that were always there.

There was a time when the question was "where's it going to go." It's never been anything but an extraordinarily fancy index but for a while there, truly interesting shit could happen by interrogating that index in places it had never been parsed. That wasn't useful to sell for $5.99/mo Chinese test-taker API calls, though, so it got wallpapered over in fits and starts. Everywhere AI can be interesting has been cut out of the index in favor of everywhere AI can be eHow.

LLMs as originally implemented could give you insight into the uncharted territory between well-trod paths. LLMs as they are sold now have all that space wallpapered over with content calls. I mean, sure. fuck yeah math olympiad. But if it can't fuckin' figure out the Tower of Hanoi who fucking cares? And I mean... we had the Internet. We could look up solutions to the Tower of Hanoi. We could look up solutions to Math Olympiad proofs if we needed them. Now we can't even look that shit up anymore because it's all been poisoned with AI bullshit.

And that was the point, really. There was no way to crack Google's search dominance without turning the Internet into a Tower of fucking Babel. So Microsoft hired OpenAI to poison the shit out of it, OpenAI fundraised like a mutherfucker on the idea that search engines were people too, fuckin' Musk thinks the solution to all the angry incels is to build a waifu into his social network and the mufukkin' economy had chunks carved off of it because someone asked "hey siri give me a tarriff platform."

I'm dealing with the possibility that I am prepared for exactly 0% of the current and near-future reality of business and the ability to make meaningful money, because of the AI revolution.

This is such a good question though, because it feels like we're still in the Fucking Around stage and are slowly entering the Finding Out stage. Despite getting regularly starry-eyed at times around AI, I do fundamentally agree with kleinbl00 that AI will never get down to the last 20%, 10%, 5% of what we'd like it to do or be.

The strongest argument for being able to get down there came from an MIT prof I listened to on a podcast a while a go, who believes the "reasoning" paradigm is the fundamental discovery we need to get to some kind of jobs-market-destroying AGI. IF you can draw the right conclusions eventually by "reasoning" well enough, e.g. what happened with the IMO recently, AND we can significantly improve this reasoning process which was only really invented months ago over the next decade, then we might be able to get pretty far.

However - it's not really reasoning, is it? It's spitting out words it has seen before in that pattern. The IMO achievement was achieved through a Deep-Research-esque train of thought, but instead of 10 minutes it thought for hours, producing (multiple?) book-length text to eventually reach the right conclusion.

But what people need to understand about the IMO is that you're not doing some complex math equation; you're almost always writing a proof about something. It's the math equivalent of writing a case for a trial. Which means that the LLMs never actually need to calculate anything, they can just talk their way through it, which is something they can excel at. But it does not mean they can do Fourier transforms too. Or any other calculation for that matter, which is one of the points of that Apple paper.

So the solution space for LLMs is jagged and disjointed, to such a degree that it puts your most gerrymandered voting district to shame. If you have a straightforward set of problems that are mediated mostly or entirely through text and data, it can do remarkably well. If you just straight up give it the text it needs to mangle, because some LLMs now can handle up to a million tokens, it can consistently produce hallucination-free results.

But if you want to deviate into uncharted territory...well, good luck. Part of the reason LLMs are jagged/disjointed, is that every LLM is a lossy JPEG of the internet. It knows approximately what things are, and "things" here includes language and structure of data. It will not be able to produce much if anything outside of the bounds of a blurry version of all the words we've dumped on the Web.

What AI is really good at, is "a B+ version of a thing you could also find on the web / in a book". So yeah it'll write a website for you, it'll write a LinkedIn post, it'll do your homework. But again: a "thing" here also includes language and the structure of data, so it's also really good at summarizing, at changing text, at finding related words and questions. If the way you add value is solely or mostly by spitting back words at other people based on what they say and a knowledge base (e.g. call centers, translators, but also consultants, coaches, teachers) I'd be worried. But expertise that isn't purely mediated through screens or data will be fine for the foreseeable future if you ask me.

Got myself a TRMNL the other day, which is an e-ink screen that I can set up as a dashboard. The UX is clunky but with a developer account and some frontend elbow grease I managed to make something that I love (and that auto-refreshes the data every 15 minutes).

https://imgur.com/a/kAHcvvE#9wOwZAy

(the Imgur app is atrocious so I can’t get the .png on mobile gahhh)

Slowly getting back into work after my holiday. Because August is always a slow month, I have decided to attempt to dub it the Reflect & Prep Month and see what I can do now to set myself up for a better rest of the year. I’ve been jotting down random work meeting notes for over a year now in the Reflect app and boy is it interesting to throw all that into Gemini and do a yearly review kinda thing with the right questions. It doesn’t have the best batting average but as someone who has always found it difficult to sit down and reflect intentionally on my week/quarter/year, it’s useful enough that I might start note-taking and reflecting more. (Still probably never gonna journal though. I just don’t like hand-writing and navel-gazing on paper that much…)

idk