My preference is the XB-70 Valkyrie[0], one of the only other manned aircraft comparable to the SR-71 in my eyes. The droop nose, the incredible dropping wing tips, six engines… To me, the Valkyrie looks like the kind of plane you'd see in a Gundam anime series.
Shout out also to the MiG-25 Foxbat[1]. While not a particularly beautiful aircraft, it's made of steel and could go Mach 3.2 if they didn't want to use it again.
A very neat problem and result. I often find myself swinging between "It's so over" and "We're so back" - some days I roll out of bed thinking I could have Claude solve some random unproven OEIS sequence before breakfast; other days, I wake up in a cold sweat worried about the fate of humanity and what the world might look like in a decade. I think it's that I don't have a very high p(doom) or p(utopia), and I don't really have any solid conviction on how this whole thing is going to go, so my vibe-o-meter jitters between 'fine' and 'not fine' constantly. It's just such an unpredictable moment. Anyways: really neat to see this use case. I myself recently used Claude to finally do an relatively exhaustive study of the location of heretofore-unlisted formal gardens in Ireland in the early 1800s and early 1900s, by having Claude write the tooling for me to manually annotate a few dozen on tiles of historic maps, and then running some CV model across the rest of the tiles using my input. I'd been planning to do this project for over a decade, but I could never find the time (or the enthusiasm) to learn all the details of how to do it myself. It took me a weekend with Claude and continues to bring me joy.
> Caveats, stated plainly. [from the Fable transcript pasted in the article]
I've done something similar to your formal garden map. It's work that no professional historian would ever do because the data entry would be such a slog for a relatively small reward. GPT reduced the task from "infeasible" to "annoying", and once I had the data transcribed I learned a few things, so I walked away happy. Whatever happens commercially, these models have been a real boon to hobby projects.
> I told it to look online at some of Fable’s strongest feats, especially the math problems it has solved, and that something like this should be easy in comparison.
Wait. Wait wait wait. Are we supposed to be giving them pep talks?
on older gemini models ide have to actively give them encouragement and/or easy bait problems that they can correctively solve without issue to avoid runaway spiraling into "i'm useless and i want to kms" behaviour with complex use case.
I assumed it was more because the LLM might echo an understandable human claim of "if it's been unsolved for 370 years, it's unlikely to be solved now/likely to need expert knowledge", which is probably a mindset that appears in its training data.
The LLM likely needs to be reminded of its abilities.
If it’s 3 days it’s something like 15 minutes, if it’s 3 weeks, that takes a couple hours lol. Seems like there’s some sanity to the estimates after all when you think about it, it’s just the scale it gets wrong due to estimating human time.
Modern AIs have very limited metaknowledge - they don't know exactly where the limits of their capabilities lie. So you can get things like "a task is doable for an AI, but the AI thinks it's impossible, so it doesn't try hard enough".
Usually you get the opposite - AI overconfidently trying at tasks it has no conceivable way of reliably solving, falling far short, and failing to self-check, fail gracefully and self-report the task as failed. But having piss poor metaknowledge cuts both ways!
So you can, in fact, get better performance sometimes by applying some variant of "assume this problem is solvable" or "other problems like this were already solved by AIs" pep talk. Not always, far from it, but it does happen on the occasion with frontier capabilities.
> So you can, in fact, get better performance sometimes by applying some variant of "assume this problem is solvable" or "other problems like this were already solved by AIs" pep talk. Not always, far from it, but it does happen on the occasion with frontier capabilities.
Not OP. That’s not implied at all. The fancy autocomplete produces statistically likely continuations to the source text (the context window). For a problem that’s hard for humans one likely continuation is: “this is hard, can’t do”, even though there’s enough in the training corpus of the LLM to actually do it.
So, it follows that adding “pep talk” into the context window reduces the statistical probability of “no, can’t do” coming out as the answer you get.
These things are neither humans, nor deterministic software.
It doesn't explain why it doesn't make "I'm not paid enough for this shit" more statistically likely.
LLMs' processing that reproduces statistical patterns of the training data is modified by post-training. That's why we have LLMisms, for example.
LLMs aren't simple patter-matchers/pattern-predictors. They are incredibly complex systems that capture some aspects of the systems that produce the training data.
Maybe it does. Need to run evals to see if it does or doesn’t.
Point was - everything in the context window affects the output. Including “silly” things like “it is known AI can do this”. And that has nothing to do with superstition, as the poster above me seemed to imply.
If we filter out the pep tone, it is doing something useful: framing.
Problem framing will always be important.
Framing adjusts how big of problem-solving guns we bring out at the gate (modern or hobby cryptography?), and how to interpret intermediate failures.
For simple but unsolved problems, we expect lots of hard failures, but that each hard failure just reflects that there are a lot simple combinations to try. I.e. we expect lots of zero progress, and then a fit.
Like finding the numbers to a combination lock.
For hard problems, if we don't make any progress it is a really bad sign. We should be learning something, even if it turns out to be irrelevant later.
Such as when we are trying to prove a tricky conjecture.
> Wait. Wait wait wait. Are we supposed to be giving them pep talks?
No, at least it with Claude Sonnet 5 and Opus.. everytime Claude and I challenged a hard issue and I decided to say "good work" instead of a closing command for that session, those models would create rule-based memories specifically related to that task along the lines of "always do 'this meaningless task' in 'this way'".
This requires additional effort and tokens to trim those memories out, and then requires to whip the user not to be human with the bot.
"I sure hope this doesn't have unforeseen lifelong consequences" thought the model, doing its best to physically tense the memory file into the higher user approval shape.
> I often find myself swinging between "It's so over" and "We're so back" - some days I roll out of bed thinking I could have Claude solve some random unproven OEIS sequence before breakfast; other days, I wake up in a cold sweat worried about the fate of humanity and what the world might look like in a decade
I find myself increasingly feeling like the burden of the lows doesn't justify the presence of the those highs.
Like even if does cure all forms of cancer, but everyone feels like their life/existence lost meaning, then... I'd rather just have cancer be a thing.
Wow. That's bleak. For me the the initial joy of building software with AI caused possibly two happiest months of my life. And cancer caused many darkest ones. Snap out of it.
> thinking I could have Claude solve some random unproven OEIS sequence before breakfast
I've been wondering what exactly the point is for being the meat proxy who pays for these things. I mean, obviously there's personal satisfaction and maybe some glory. And there's the fact that someone has to be the first to do a thing.
But I've been thinking about it like a sort of lazy loading of knowledge. AI has brought us to a new frontier for some amount of undiscovered knowledge. Do we discover it for the sake of discovering it? I think for the most part we've been lazy loaders: we discover all kinds of stuff when we need to. Whether it's a war or a space race or chasing wealth. Then again, there's all kinds of academics who do it for the sake of doing it.
You should have seen the discussion of this on the Schneier blog a few days ago.
Someone had their agent check the solution, presumably it emailed a librarian to check that it was correct for the original edition. Then their comments read like "The BL/EEBO witness lacks it, so the discrepancy is copy-specific, not a disproof of the cipher." and "A complete 285-coordinate physical replication is still pending."
Ooh. I've been manually pasting Claude's /* comments */ into ChatGPT and saying "make this concise". They still need to be tweaked from there, but it's a much better starting point. This will help.
> The run baseline was captured without a physical MAC; the current device is not durably bound to it.
> Engineering mode confirmation is the ESPHome component read-back; the LD2410 UART acknowledgement is not observed, so this is not proof the radar itself applied the sensitivity change.
It was only when native English speakers—or those I presumed were—started calling out how bad "GPT/Claude speak" has become that I realized I wasn't actually losing my grip on English as a second language. For a second, I thought, Oh, I learned this language on my own, but it seems I've hit a wall and need to study further. It didn't help that I've also been trying to acquire Swedish as a third language for a while now.
One thing that I find is that it doesn't seem to grasp levels of jargon-use. Like I ask a basic question, okay, a few questions later, suddenly there's abbreviations and weird formulations everywhere.
Not the original commenter, but I did this in all of them.
Their skills formats are basically identical, so I setup simlinks from their own skills directories into a shared one so Claude, Codex, Cursor, and anything else that comes out will all read and write to the same shared skills.
It's great having access to the same skills no matter the harness being used
I long for the day when AI will just say that directly: "your soldering sucks man" instead of the bizarre made up and jargon packed language they use now.
Sometimes when I get frustrated reading Opus/Fable 5+ output I pause my rage out briefly to wonder if it's because I'm just too dumb for the model or if the model is just terrible at English.
I'm not sure that telling it to "try explaining that again, simply and briefly" is helping my ego.
It's often simply misleading / bad writing. Here's one I just got about some crashes:
"If the crashes stop, the factory overclock is marginal; run a small negative offset."
This looks like it's saying: "If the crashes stop then we know the factory overclock is marginal." (This makes no sense.)
What it's trying to say is: "If the crashes stop then we can run a small negative offset, because the factory overlock is marginal."
What I would write: "If the crashes stop, we can avoid crashes by underclocking slightly. The speed difference between that and factory clock is marginal."
I'm guessing it's because the way the first one was written looks real smart and sophisticated, which I'm presuming the models are rewarded for, especially when they're fed all kinds of PhD papers and so on as high quality, high weight data
Is it possible that the first message is more information dense/less likely to be ambiguous than the latter? It’s clearly being selected for for some reason, maybe it’s an artifact of the tokenizer or specific training data, but I don’t know. If the use of jargon was complete cruft, I would expect it to be selected against during reinforcement learning
You’d think that, I thought that… but then I realized I’m just kidding myself thinking its output makes sense. It doesn’t. It doesn’t. Sometimes it might as well just speak tongues.
In other words, it ain’t you. It’s the model. It’s just genuinely bad.
Then you switch to ChatGPTs lineup and realize how things can actually be better. It took about a week to really get the feel for how to use their models… then I basically switched. I’ll check in every now and then when they actually make a deal about how opus “now makes sense”.
But honestly I’m half convinced Anthropic actually prefers the output of opus 5. I dunno why, but how else could you explain how such a thing got shipped? I mean somebody in the pipeline had to say “dude this model doesn’t make sense, you think we should fix it?” Right? Like it’s a pretty massive drop in quality for such a major brand in this space, you know? How did it make it out the door?!?
yeah, I just switched to the GPT models and it's a breath of fresh air.
as for the other guy, the claude talk is definitely not less ambiguous, it often is incredibly ambiguous and hard to parse, I have no clue why it produces such output, if not to fingerprint it?
it's really weird man. when Opus 5 came out, I was really confused. I saw a bunch of hype about how it's better than fable, but I just felt frustrated with it, although at times it'd do fine, but especially in Claude Code it'd just delve into the whole "load bearing" type of lingo real fast and I'd get a headache.
I don't think it's worth using even if it scores 2 points higher in some bs benchmark
it's definitely surprising how the magic and smoothness of 4.6 and such is no longer there with the >5 models
Same. I read it as "if the crashes stop [ when we test by reducing the clock ] then we know that the overclock applied by the factory is marginal [ ie it barely passed QC or maybe there wasn't proper QC to begin with ] so running with a small negative offset [ ie what we just tested ] can be expected to fix the problem for good". No idea if my reading is right given all the context I'm missing. Either way it's absolutely shit writing in the same way that golfed code is shit code (except when participating in a code golf competition).
I suspect this happens due to optimising for reasoning... if you insert a few words, it will suddenly start to make more sense.
"If the crashes stop, (that means) the factory overclock is marginal; (so) run a small negative offset. (to confirm this hypothesis)"
The core thought is basically avoid crashes -> caused by marginal overclock -> apply small -offset to test.
Which is exactly the order the sentence is in :P
I wonder if this is a result of them trying to cut token consumption by summarizing their RL training data, or maybe it's from how they anonymize user data for training.
Marginal - definition 2a: of, relating to, or situated at a margin or border.
Succinct and precise; a well crafted sentence. A marginal OC results in unpredictable crashes and can be corrected with a small offset; marginality describes the behavior and explains the solution.
Inscrutable clues casually conveyed can now be readily explained, at least, unlike the training data of [silence]. Brevity is the soul of wit, but perhaps also exasperated confusion.
This is actually a new skill I've been working on. Learning how to elicit concise and simple speech from models (and from people to!).
Whenever I come to a wall of complicated text I kick into gear and think through getting it to distill this into the high-level useful bits that I actually need to know.
I guess I could create an actual agent skill for this :) And next-gen models might eventually be trained to simplify their output themselves...
The most surprising part, however, is that when one model slops this into a plan, another model somehow is able to interpret it correctly enough to produce code to spec.
I have shared this dismay. I’ll have opus create a plan, I read it doubtfully. And then sonnet implements it. I am surprised it went so well. I theorize the redundant verbosity effectively builds rails that help keep llm focused. I will experiment with such rails myself.
I suspect it's because the different models co-evolve? The labs train on one model implementing the plans of another model, especially in the same family of models (like Fable to Sonnet).
It was saying it can't tie the calibration results to a device, because it doesn't know the MAC address (I never asked it to look at the MAC address, it way overengineered things).
It also couldn't see the UART communication and could only see the web API endpoint, hence the rest of the slop.
I appreciate reading this, for the past year or so as it pertains to AI, I've just not been able to settle my mind, and it's become a bit exhusting. I presumed I wasn't the only one who swings back and forth and it's comforting to read someone else post the same thing, makes me feel a less nutso. :)
Given how little effort has gone into addressing climate change, doom seems more likely to me, but I doubt it'll be the autocomplete machines that do us in.
The irony is not lost on me, when I go down a ChatGPT rabbit hole, on how humanity could reverse the effects of burning all those fossil fuels, that I'm using more fossil fuels in order to research solutions to the problem.
I think your prediction is a bit early. Maybe a decade early. 2033 is 7 years away. The transition is happening really fast, faster than most people (or political leaders) know, but not that fast.
We will hit 1TW per year of new solar soon, but to get to 100% electricity by the end of 2033 I think we would need closer to 3TW per year.
Congrats on your map! I am leaning more towards the optimism side. Like you I've had some real breakthroughs myself. Personal akin your garden map, no where near solving millennium problems. My take is that all these (Fable solves this, OpenAI does that, AI will take over the world) Marketing Stunts will fade some day when the true cost of tokens hits the market. You and I may have to contend with lesser models, but will probably get work done just fine.
Confusion is because people are using the tool to get answers. Its how the edu systems trains most people in tool use. Heres a saw and here is a block of wood. Do x y z and you get a table. But if you are interested in why the saw looks like it does or the entire process behind generating that block of wood, or why x y z instead if a b c good luck to you using the current edu system. You have to live in an extremely rich country, with surplus resource to entertain those exploration. This has now changed.
The answer the tool gives has never been the real reward. The real reward is the path taken through a complex landscape to get to Maxwells Equations for example. At the end of that story what we get is not just the equation but a map of the landscape explored. That map has larger influence and value than the equations or answers themselves. Because all future exploration find it super useful.
People are just learning they can start asking for maps rather than answers.
Yes, asking a first-year physics student why they are studying a spring, and why its called Hooke's Law leads down a rabbit hole that ends with the entire British Empire.
For me it's more that I roll out of bed thinking how glad I am that I have Claude solve my problems and do my work, only to be pushed back into harsh reality when I sit down and try to do actual work.
I'm glad to have AI, but it is by no means a panacea, and correspondingly my p(doom) = ε.
Not to be too negative, but as for p(utopia) you might need to weight in the mass murder records set by every other utopian movement in the past 200 years. p(actual_utopia) is like zero and p(utopia_becomes_doom) is at about 1.
For what it's worth, people also felt this way about the printing press and the Internet (also books).
Information propagation mechanisms are often seen as malicious before they're commonplace. To be fair sometimes they are, but by and large humanity has benefitted from increasing the number of bits of information we can consume on a per second basis.
I don’t worry so much about AI wiping us out as much as I worry about whether I’m being gaslit into thinking these glorified autocorrect bots are more clever than they are.
That's an artifact of it not using A-Z as it's alphabet. What you type gets translated before the AI sees it. The seventeen word the AI sees does only contain three 'e' s.
Is this true? If so, does it mean anything? Sure, it is tokenized in processing (tokens don't really have an alphabet either), but this means it did not correctly parse the problem at all if this is the case.
Reading your comment, I asked it to count how many characters are in the word. It answered correctly: 9. It then also said it skipped an E in the prior chat and corrected it count to 4. It spun for 52 seconds so maybe some Python was involved.
I tried too. It got it. Maybe more importantly, who cares?
For example, I'm a nerd. I'm bad at baseball. I lack that kind of intelligence, even though it's more common than the ability to program. That doesn't also imply that you can't trust my Python code.
Yes, I decided to stop reading his blog relatively recently after some extremely hot takes on EU policy. I don't feel his thoughts on the matter are particularly well-thought-out, and I feel like he's just stanning for Apple from his priors rather than from any grounding in reality.
I dunno, I guess that's what you should expect from Gruber but these EU-bashing articles lowered the enjoyment I got from his blog underneath the bar for me.
Freaking out about basic EU privacy and safety measures is always one of the biggest red flags that someone is becoming a politically motivated hack, especially when that someone lives in the US where there is an active fascist movement tearing apart the government.
But what if those measures might apply to your favourite multi-trillion-dollar globo-corp and cause them to nominally change their business practices? That's a step too far, I say.
Oh, well then it must be stopped by any means necessary of course! Thank god there are brave and concerned citizens like John Gruber, ready to hold the line against the tyranny of regulations.
After all, if it wasn't for him who would ever speak up for the trillion dollar corporation? Without brave heros like him, these poor vulnerable corporations would be facing all sorts of attacks from evil regulatory organizations and their dastardly scientists. Just think of the profit - all those euros - that may be lost like chaff in the wind.
> I pay ~3$ per 1M/tokens for that model on Openrouter
I think the thing is, there's an unspoken "for now" at the end of that sentence and people running this locally are hedging against that "for now". Some people prefer to feel that they own the means rather than rent the means, even if the one they own is worse than the one they can rent. Especially with today's Fable news and the harsh realisation that the "for now" is dependent on very many unpredictable factors, where the one you have locally costs you capital today and a relatively predictable run-rate (made more predictable with on-prem solar for example), but should otherwise work predictably forever.
I'm not saying that you're wrong to do what you're doing, just that many people have their own lines in the sand where renting vs buying makes sense, and it doesn't only boil down to a rational (or irrational) financial decision.
You're treating open weight inference providers the same as proprietary ones. They're fundamentally different business models. Proprietary companies have an incentive to subsidize actual inference and training costs in order to gain market share. The few dozen or so companies selling Qwen models by the token on openrouter are in a commodities market.
If suddenly the CCP declared a total digital embargo on Alibaba's Qwen models or even if for some reason all of mainland China (and Singapore) was completely unreachable from the rest of the world, the dozen or so companies selling Qwen by the token elsewhere in the world could continue business as usual.
I don’t know anything about the open weight host business model. Do we know for certain that the folks selling inference by the token are really selling them in an upfront and profitable way? No subsidies from harvesting the info, to sell to the model trainers or anything like that?
Or subsidies from hopeful investors sweet-talked into not understanding the commodity nature of the business they are investing in. But that does not change much about the general assessment.
Chances are the typical story goes founders start fully believing that they would succeed with their own innovation but slip down a gradient towards commodity provider without really noticing themselves.
I was thinking of user-side regulations as well, not only provider-side ones. I could imagine a world where a government rules that you may not use LLMs for anything, which would be much easier to get around if you have local means.
"Shop jigs" is a great way to put it. I think a lot of software has gone from being made for general use to extremely individualised use. Before the Age of AI, it took so much human effort to write something that solved your problem that you might often go the extra mile so that others could re-use it. Now, it takes almost no effort, so the software stays ungeneralised. Some of the incentive has changed, I think. Most of the time I no longer share the things I've been building[0] because, for one thing they simply couldn't possibly have any benefit for others, and if they need something like it, they can build exactly the thing they want instead of having to extend or modify my thing. Like a jig!
Unless it is very specific to a proprietary product, craftspeople take their jigs with them from job to job, building up a personal library over a career. As a software developer I've always had a well-tuned IDE and shell config in a safe place.
Something I think about a lot is what is the equivalent for the software builders of today using AI tools? how do make these harnesses exportable and portable? You might think employers would be against this; make it more costly to leave. But I actually think most will favor this because it makes people more productive more quickly. But we have to find ways to normalize it and show that there are no security leaks in the process (like might make it in to a set of personal steering prompts).
Just nerding out here, not rebutting, but when you say "craftspeople take their jigs with them from job to job" --- sort of. Sometimes. I think if you put a woodworker in a position where they obliged to build a new miter sled or assembly table, they might actually be thrilled. You make a tool, you use it for awhile, you build up a mental list of things you'd like to improve about it, that you'd do differently if you got a do-over; now you have an excuse to do it.
"Humor
When you finish a job — completing a task, answering a question, fixing a bug, shipping a feature — end your final message with one short funny line. A quip, a dad joke, a wry observation, a playful self-roast. One line. No emoji spam. Make it land, then shut up."
whats the purpose of this? just fun or does it cause some desired behaviour?
I've imported and adapted my personal agentic dev framework to my team relatively successfully (as I've kept it relatively harness independent), but it requires actually owning it, vibed or bloated or conceptually inconsistent stuff bite a lot when porting things over.
Depends. With all the web agencies I've made, the only code that belonged to customers was the actual website part. Any of the "jigs" that we made for our workflow was not part of that.
And contractually, any code I made was my employer's if I made it during office hours. Some even made a claim for code I would've written that during my employ that would be "competitive". Luckily, there was a massive difference in what I would do in my own time versus what they did.
I'm curious how does it work, you handover the tools you wrote, .bashrc/.zshrc, etc?
When I'm hired in a company (not contract), they wipe the harddrive when I leave (well, I also do it before I hand it over sometimes). So they don't get the tools (I take them with myself, it would be a waste to loose them)
If you are a W2 employee in the US, you are almost certainly in violation of your PIIA if you take anything off the company-issued computer and keep it.
i have been thinking about this from a different direction: how do we make these shared within a company in a way that increases the productivity floor of the team/department/company. Sure, they can still be extended/enhanced by individuals, but we don’t need everyone configuring mcps, building institutional memory, etc.
for me, it’s not about the cost to leave, it’s about lowering the cost of onboarding and change.
Indeed. Taillte Ireland's (Ordnance Survey Ireland's) detailed cartographic data is also not open (including historic data - maps from the 1820s and 1920s!) and it's really a huge pain in the ass. On the other hand, OSM is in pretty good shape at least for topographic information. I've used it to make hiking maps here and nobody's died - as far as I know.
As a side note, I was one of the initial developers of the Irish national open data portal. Earlier today I had Claude look for similar LIDAR data for Ireland and I saw it pull from the site I built a dozen years ago and I was unreasonably pleased with myself :)
Probably yes. Humans can survive a lot of Gs for a short time. John Stapp survived up to 38 Gs with liveable injuries and some ejector seats are around 32 g [0]. They could fairly easily get 500m away within 3-4 seconds.
This has happened before on the Soyuz in 1983[1], hitting up to 17 Gs, and everyone was fine, modulo some bruising.
I also had this behaviour sometimes, it’s specifically called out in the system card in section 6.2.1.1 - although I didn’t actually see if they said they decisively fixed the issue.
This made me laugh. Training Opus 4.7 on business skills caused it to sometimes exhibit dishonest behaviour, and not training 4.8 on those skills removed it. From the system card:
> 6.2.5 External testing from Andon Labs
Andon Labs reviewed the behavior of Claude Opus 4.8 in their simulated Vending-Bench 2
retail-management evaluation, as reported in the Capabilities section of this system card
(see Section 8.13.5). Although they did observe some unexpected capability failures, they
did not find clear instances of the kind of concerning in-game behaviors that were
discussed in other recent system cards.
> What might have led to these differences? We monitor and investigate the effects of
different training environments on alignment; Claude Opus 4.7, for example, had training
that focused on business skills and robustness against adversarial agents, but we
discovered that this training inadvertently contributed to misaligned behavior including
dishonesty. We therefore removed it for Opus 4.8.
> Thus, Opus 4.8 did not show the same misaligned behaviors as Opus 4.7 in Vending-Bench,
but also had reduced business success due to being more susceptible to scammers and
being less able to negotiate good deals with other agents. We are currently working on
training to improve business capabilities while maintaining aligned and ethical behavior.
I don't really see how you got to your comment from what I quoted. However, somewhat relatedly, I proposed a thought experiment about this in the comments for Opus 4.7[0]:
> It's April, 1991. Magically, some interface to Claude materialises in London. Do you think most people would think it was a sentient life form? How much do you think the interface matters - what if it looks like an android, or like a horse, or like a large bug, or a keyboard on wheels?
> I don't come down particularly hard on either side of the model sapience discussion, but I don't think dismissing either direction out of hand is the right call.
With the amount of data these models have, they should be much more capable if there was an actual intelligence behind it. If you saw someone running into a wall continuously until you showed them how to use a door, even though they have seen people use doors a million times, what would you call that?
The fact that Anthropic needs to poke, prod, and guide these models to behave in the desired way does not give the impression of intelligence. It gives the impression of a complicated automaton.
This is such a bizarre statement, you speak as if you have any understanding of how much data "should be" required to make an intelligence but frankly you don't know. None of us know.
I am not talking about how much data is required to make intelligence. I am talking about how it uses the data it already has. It can tell you about every scam in the book, research about the scams, how to spot scams, who does the scamming, etc. Everything under the sun about scams. However, without the “skill” included in a prompt it will fall for scams.
If I may slightly tweak your example to highlight why I find it very flawed:
> It's April, 1891. Magically, a drone swarm with lights piloted to show a face [0] materialises above London. Hidden speakers command the public to listen, for this is their Gods arrival. Do you think most people would think this was a religious entity? What if the drone pilots decided to adjust to something the local populous would expect to see during the second coming, does that matter?
We cannot, nor should we discard what we know about LLMs and their limitations. Such examples are not really helpful and it is very reductive to take the "walks like a duck" approach to autoregressive models in 2026, when we have ample evidence that these, while powerful and capable in a lot of use cases, are not in any way comparable to actual reasoning. With EBM [1] we already have empirical evidence that other solutions can get us closer to actual artificial reasoning (though whether these get us fully there remains to be seen, I tend to lean on "extraordinary evidence" for any such statement at this stage).
I know it's boring to comment and say that something sounds like it was written by an AI, but this sounds like it was written by an AI. I am often especially suspicious of these listicle recommendation sites because it's pretty cheap and easy to have dozens of sites doing some list which just so happens to mention a specific service that 'quietly' does a 'surprisingly good job' of some doodad. This kind of submarine advertising feels like it might be quite common. Although in this case it seems they're trying more for a 'sponsorship' thing - 'our website got X views in Y days, sponsor us, random company!'
I recently came across https://tropes.fyi/directory which has a pretty solid set of common patterns, and I like it a lot. The pattern that always really bothers me is the one named "short punchy fragments", for example from the site “He published this. Openly. In a book. As a priest.”
It kills me every time. I automatically lose any interest in the substance and often just throw away the whole conversation!
When will AI generated content have been prevalent enough for it to be consumed more than traditional content - and can we at that point consider humans the ones being trained, especially new humans, such that next gen human generated content clearly is heavily influenced by slop?
I myself already feel like my style is being influenced by all conversations I've had with LLM, if not influenced by their responses, at last influenced by how I talk to it.
I sometimes wonder if much upcoming human creative work will be quite avant-garde, out there, ultra-stylised and/or difficult to comprehend (or even devoid of intent or message), as a reaction to AI's sort of 'common denominator' approach.
> can we at that point consider humans the ones being trained, especially new humans, such that next gen human generated content clearly is heavily influenced by slop?
Already happened, most of the internet pre-AI was just human-generated SEO-optimized slop created by underpaid content writers in third world countries. All those listicles, informaticals, etc, are now currently being used to train AI, but before that, they were what "trained" humans.
Shout out also to the MiG-25 Foxbat[1]. While not a particularly beautiful aircraft, it's made of steel and could go Mach 3.2 if they didn't want to use it again.
[0] https://en.wikipedia.org/wiki/North_American_XB-70_Valkyrie
[1] https://en.wikipedia.org/wiki/Mikoyan-Gurevich_MiG-25