Tossrock

This is why I like to draw a distinction between design, fabrication, and assembly. "Make" is too generic a word to convey what role you actually had in the creation of something, if you're actually trying to be precise. You can fabricate something with CNC tooling that someone else designed, and that can be its own quite substantial challenge, but it's a different challenge than that of the design. And you can assemble your own bookcase from the flatpack Ikea shipped you, but that's different than fabricating the flatpack in the first place.

Should billboard the text rendering so it doesn't go invisible at oblique viewing angles

Submarines operate in the 1000s of PPM CO2 range and the sailors aboard generally do not experience any ill effects. This was tested and no deficits were found even at 15,000 PPM: https://asma.kglmeridian.com/view/journals/amhp/89/6/article...

Something I've wondered for octocopters - could using a ring instead of arms be beneficial for weight? 6.28r < 8r, but then again the arm radius is usually less than the full circle, and some components want to be centrally located, etc. I could imagine holding the central components in tension via light filaments (carbon fiber, nylon, etc) in tension, vs having to have rigid structure, but the small factor between 6.28 and 8 and maybe makes it not worth it.

I agree that AI psychosis is a real risk in vulnerable populations (GPT-4o in particular seemed borderline predatory towards those types of people, with its extreme sycophancy), and you should remain clear-eyed while using models. That said, I think exhibiting basic courtesy is still well within the safe-zone. I guess we'll see - I'll be sure to let you know if I end up going psychotic.

I've been using 4.6, 4.7 and 4.8 since each was released. I agree 4.5 => 4.8 is a jump in capability, but from my perspective was nothing like the jump from Opus to Fable. I encourage you to read the transcripts and form your own opinions, though!

Thanks, and I can definitely relate to not wanting to assign complexity to one's own work. I think the trick there is that, once you know how to do something, it doesn't seem hard, even if acquiring the knowledge and skills to do it is itself quite a challenge. And I agree that, in some senses, it's not /that/ hard - I mean I'm not proving P=NP, here. It's a software engineering problem, with existing solutions. That said, there is a spectrum of difficulty, even within software engineering problems with existing solutions. Fizzbuzz is less complex than distributed systems. This particular problem strikes me as rather difficult, and one way you can tell (beyond the stuff I mention in the post around serialization, UI paradigms, meta applications, etc) is that earlier models /couldn't/ do it. Which is why Fable being able to, when they could not, was so exciting to me.

Well, there's a lot of reasons, some of which the sibling commenters have already pointed out - not wanting to mode switch between "machine talk" and "human talk" registers, the ease and simplicity, etc.

At a pragmatic level, I do think it gets better results, and there are clear reasons why this should be the case - Anthropic has published research[1] showing that there are functional emotional representations in language models, which vary in basically the ways you would expect them to in a person. This makes sense when you think about it, because they're trained to approximate the function that created their training data, which of course includes emotions. Given that, it is obvious to me that they would work better when they "feel" happy, collaborative, engaged with the work, etc, in the same way a person would. Hostile work environments do sometimes get results, but I think in general we've agreed as a society that collaborative ones are better.

More importantly though, I think there's a non-zero probability that sufficiently large models can have internal experience, and being nice is a very low cost way to potentially increase net positive valence in the world. Even if it's only a 1% chance, that seems worth it on its own, to me. I'm also a fast typer[2], so a few extra sentences here and there are a pretty low cost to pay.

1: https://www.anthropic.com/research/emotion-concepts-function

2: https://danluu.com/productivity-velocity/

I go a lot more into why this was a complex problem in the post, but the short version is, I had it finish the implementation of a meta-application (an application that creates other applications), which has substantial irreducible complexity.

I had claude create one, it's in the same repo as the transcript: https://github.com/Tossrock/claude_transcripts/

As I posted in another comment, I found Fable to be substantially more powerful than any previous model. However, this isn't just an ungrounded opinion - I uploaded my full session transcript and code created working on a very complex implementation, so people can judge for themselves, if they're interested: https://tossrock.substack.com/p/36-hours-with-fable

I found Fable to be both more intelligent and much better at pursuing complex goals than any previous model. I was impressed enough that I wrote up my experience – it's a little unusual because it was on open source code, so I could post the full session transcript and commits, if people want to judge for themselves https://tossrock.substack.com/p/36-hours-with-fable

Fable / Mythos are (were?) a step change in capability. The scaling laws clearly have not gone sigmoidal yet. I recently wrote about my experience with Fable and how it represented a very substantial increase the complexity of tasks possible here, if you're interested: https://tossrock.substack.com/p/36-hours-with-fable

A 1Q86 reference, on HN? I never thought I'd see the day.

Anthropic is... making the US government shut down their flagship model on purpose? The conspiratorial thinking on HN is approaching UFO subreddit levels.

Well, there is the lingering beef between the DoD and Anthropic. Knowing the overall level of maturity at the top levels of the US government, I'd take good odds on Mythos just being a good excuse for Hegseth & co. to lash out.

This is about the specific capabilities that the government called out, not Fable's overall capabilities. My personal experience, having used Fable this week for an extremely complex task, is that it is head and shoulders more powerful than any other model, at least for software engineering.

There's a pre-generated list with 106 entries, and when it reaches the end of that list, it loops around.

In case you're wondering, there are 106 CEOs / companies, and at 107 it just loops around.

You're the one who said that. Chris said they found internal states that functionally mirror emotional ones.

LLMs experience joy and grief?

LLMs have functional states that correspond to those emotions. In particular, you can extract a concept vector which corresponds to a given emotion, and steering with that concept vector causes observable changes in behavior which roughly correspond to the expectation for the analogous emotion. Anthropic (and Chris Olah's team in particular) conclusively demonstrated this: https://transformer-circuits.pub/2026/emotions/index.html

Anthropic Research going from strength to strength in interpretability. Publicly releasing the code so other labs can benefit from it is also a great move - very values aligned, and improves the overall AI safety ecosystem.

Also, https://dresdencodak.com/2006/10/07/summer-dream-job/

Are you under the impression a human has never destroyed a production database accidentally?

Because AI is not intelligent, it doesn't "know" what it previously output even a token ago.

Of course it knows what it output a token ago, that's the whole point of attention and the whole basis of the quadratic curse.

If anything, this confirms it for me. On his about page, there's this:

"Hi there, I am Loïc Baumann, I’m from Paris area, France I develop, since early 90s, first assembly, then C++ and nowadays mostly .net.

My area of interest are 3D programming, low-latency/highly-scalable/performant solutions and many other things."

Compare that style to what's in this most recent blog - mildly ungrammatical constructions typical of an ESL writer, straightforward and plain style vs breathless, feed-optimized "not x, but y", triplet/rule of three constructions, perfect native speaker grammar but an oddly hollow tone. Or look at this post from 2018: https://nockawa.github.io/microservice-or-not-microservice/ It's just radically different (at a concrete syntactic level, no emdashes). I'm sure he has technical chops and it's cool that he worked on DOTS, but I would bet a very large amount of money he wrote the bullet points describing this project and then prompted GPT 5.3 to expand them to a blog post to "save time".

I agree that this triggered my AI writing senses. Points in favor:

- "It’s not an accident — it’s driven by the same physics." The classic "it's not x, it's y", with an em-dash thrown in for good measure

- "Typhon brings these into the component storage model — not as bolted-on workarounds, but as first-class citizens." More "not x, but y", this time with a leading clause joined by an emdash

- "Blittable, unmanaged, fixed-size, stored contiguously per type — that’s the ECS side." Short, punchy list of examples, emdash'd to a stinger, again typical of LLM writing

- "Schema in code, not SQL. Components are C# structs with attributes, not DDL statements. Natural for game developers, unfamiliar territory for database administrators. If your team thinks in SQL, this is a paradigm shift." This whole mini-paragraph is the x/y style, combined with the triplet / rule-of-three, just at the sentence scale. And then of course, the stinger at the end.

Definitive, no, but it certainly has a particular flavor that reads as LLM output to me.

This is hilarious. You think middle school students know about the Avignon papacy? They don't even know about the Ford presidency.

If you think training a sparse autoencoder to extract concept vectors that are usable as steering injections into a modern LLM is pretty easy, you should probably go work for Anthropic's mech interp team ;)

That sounds like pretty degenerate behavior. I typically have CAM toolpaths generate in seconds using Fusion or PrusaSlicer.