> Again, it wasn’t perfect. As an expert, I was able to spot some errors and omissions (some as a result of the design I had asked for) that I had the AI correct
That's the bit that stuck out to me - that's longer than I would expect to work on a problem in a day or even expect to go back & fix the output of something that has a core reward loop of hours.
My customers are currently clamoring to push down my agent response times from 85 seconds down to below the 20s mark.
At the same time, it is very dissonant to see the industry heading towards hour+ long workflows with an agent.
Anecdote: I fed Fable some models I’ve been hand verifying (basically, I sketch out a scenario for Opus to model, it builds it, I ask it to show me the math, I correct it, we iterate like this, then I double check its code to make sure the math matches the model logic). Fable found almost every error I found, and then had some interesting suggestions for additional variables.
It also burned through my usage quota like a late-90s Hummer.
It's fun and it looks good regardless of whether its 100% correct (It would certainly take me more than 9 hours of work to do better than this). Making these bespoke tools possible for most people is a big deal.
It put the chart title directly on top of Australia.
Which just about sums up my experience with using LLMs to code, really (though not with these state-of-the-art models, admittedly) - it's amazing what they can do, but left to their own devices they'll make boneheaded decisions.
Doesn't work too well on desktop either! This is decent but it's also an early hackathon set-up - this is something that you can set up on a sonnet model fairly easily (without the weird CSS slop that anthropic models seem to love).
I'm not very threatened by this if this is the dangerous Mythos model - it just seems like a slightly incrementally better sonnet
What are people working on that they see such a substantial difference between Mythos and Opus? I'd say I'm working with advanced stuff and more than often Deepseek is even more than enough. Why is everybody a genius in here?
> Switched to Opus 4.8: Fable 5 has safety measures that flag messages on most cybersecurity or biology topics. They may flag safe, normal content as well. These measures let us bring you Mythos-level capability in other areas sooner, and we're working to refine them. Send feedback or learn more.
What I find fascinating that there is so little substance in this article about the quality of produced code and the medium. Is the code documented and tested? Is it understandable and extendable? Is it secure? What language, framework, database was used? Author mentions judgement and taste - well, is the code tasteful? Will the model rearchitecture the entire thing if I ask it to add new functionality, spending another 9.5h in tokens? I assume that the research part is domain knowledge = how different types of travel translate to time making it presentable; how did the author verify this?
These questions are even not about AI: if I were to give money to a human agency and were given something they tell me works, I would ask the same questions. If I did not know how to evaluate, I would hire people that do. With LLMs the verification part is what bothers me the most.
So would you be more comfortable if the user them just prompted the AI to use a specific language, framework and database. Aren't we all just going to reddit and finding out what all goes best with what? But also I don't trust nothing from it, even though I've seen it.
These days it's uneconomical for human to verify AI generated code. So we ask the AI to do it. Like when we asked the FBI to audit itself and they found no problems :)
It still does make errors, yes?
Because it is not usable, if we need to verify everything.
AI is only interesting if it can do things that humans can not do.
If you can verify results because you can do it yourself, then why use AI?
It will just bind highly skilled people to do verification work.
Instead these people should do the actual work, results will come quicker.
So AI is only interesting to you / your org / humans if it can do things that you can not achieve.
But if it still does errors, how could we ever know that super-invention by AI is not wrong?
If we can not rely on the correctness of the result, it is not usable at all.
AI must create reliable and correct results always.
That was a very fundamental requirement for computing.
This problem has not been solved.
You can either adapt or survive man, coping and negation dont help, AI is here to stay and yes it does require pilots but this map would have taken you weeks to do, the AI did it in 10 hours, you can still dedicate a week to refactor.
Also this is easily solved by .md spec files, this whole "bad code" cope is just FUD'
I don't think that putting a text file saying "don't make mistakes" is going to get LLM output to the point where it doesn't need professional input, guidance, review and refinement anymore. They don't make these systems more deterministic. There have even been study results showing spec files reducing prompt adherence.
There also isn't any meaningful articulation of why this is a "leap forward"... literally everything claimed in the article has been claimed in the same breathless tones in articles written a year prior.
I get that there's little sense in arguing with the MBA hivemind, but... c'mon.
I manage two teams of highly motivated, largely pro-AI engineers. Both teams have independently concluded that they needed to ramp down GenAI usage because of code quality / maintainability concerns. Both teams have suffered from protracted outages caused by LLM jank not being sufficiently fenced off and guarded against. Both teams have expressed concern that the code generated by LLMs is far too verbose, full of slop, and rapidly becomes an unmaintainable mess.
These are teams that are building non-trivial LLM solutions (deep agentic data synthesis and multi-modal data tagging). They are using the technology creatively and pro-actively, not just vibe-coding slop and throwing their hands up when it fails. Both teams will continue using GenAI coding agents, don't get me wrong - but the gains are incremental, not transformative, and need careful fencing to make sustainable.
Nothing in these articles resonates as real. People who work in reality don't agree. I don't understand why this shit keeps getting attention (or rather I do, but the reasons aren't good).
> First, how good is Fable? In experiment after experiment I conducted, it outperformed basically every other public model I have used by a considerable margin.
What makes me excited is that GPT 5.6 (its actually GPT 6) is going to be crazy
I have been using it for less than an hour so take this with a grain of salt of being excited for the new tech.
In a project like mine (https://github.com/tsz-org/tsz) I am constantly frustrated that models were not doing enough research and were not taking into account other situations. Again and again models would produce code that would fix one thing and break 2 other tests that were "unrelated".
With Fable it seems like tasks are taking much longer (I have not seen a pull request from Fable sessions yet) but reading the transcription of those sessions I can see how it is doing the right thing by not leaving any stone unturned.
As the article says, it's hard to communicate this "feeling" about models because it is very project specific but I thought I share
> This is a map that shows the distance you can travel in a given length of time, and the first one was created in 1881 showing travel times from London.
The first item on the article, the first thing it showed, was wrong though.
It is 100% faster to go from London to New York in 1881 than Volgagrad. Or any of the Russian hinterland colored green or Turkey or Egypt.
Reading it, I can't help but feel he's being paid to write this. Or maybe he hopes to be paid. The language he uses makes him sound like he's fawning over the lost days of his childhood. Pardon me for being skeptical, but a trillion dollar company running a net-loss is hoping to IPO, and needs to sway public opinion by any means necessary. I would imagine that no dirty marketing scheme is off of the table, even from the self-proclaimed "good guys".
Man, that poem it made is terrible. Like just incredibly bad. Sure it's neat that software can make an incredibly bad poem but there is enough bad poetry in the world that we don't need it.
Terrible? Incredibly bad? Something tells me you are not very familiar with poetry, literature or writing in general. This exercise gets its inspiration and tone from one of Stanislaw Lem's Cyberiad short stories ("Trurl's electronic bard"). Besides, what did you expect from a "10 pages epic rhyming poem about a haircut where every word starts with the letter S"? Robert Frost?
Reading the first few paragraphs of what he calls "the most sophisticated academic social science paper I have yet seen from an AI" does not impress as much as I hoped.
"Posterior beliefs about market demand are purely referencedependent: holding dollars raised constant, they track only performance relative to the founder’s
self-chosen goal—jumping half a standard deviation at the threshold, responding steeply for the first ten points past it, and flattening thereafter"
Humans generally don't verbalize data this way. The summary document is also very fluffy.
You can see this all over the place. Under the Fable post in HN, you have simonw talking about the “feel” of working with Fable and how much better it is. If I believed in conspiracies, I’d have said it’s all orchestrated marketing…
79 comments
[ 3.7 ms ] story [ 54.1 ms ] threadHe is a professor but sadly also an AI shill. He should switch to advertising washing power.
> Again, it wasn’t perfect. As an expert, I was able to spot some errors and omissions (some as a result of the design I had asked for) that I had the AI correct
That's the bit that stuck out to me - that's longer than I would expect to work on a problem in a day or even expect to go back & fix the output of something that has a core reward loop of hours.
My customers are currently clamoring to push down my agent response times from 85 seconds down to below the 20s mark.
At the same time, it is very dissonant to see the industry heading towards hour+ long workflows with an agent.
It also burned through my usage quota like a late-90s Hummer.
https://isochronic-passage-chart.netlify.app/
Doesn’t work too well on mobile but looks interesting
Which just about sums up my experience with using LLMs to code, really (though not with these state-of-the-art models, admittedly) - it's amazing what they can do, but left to their own devices they'll make boneheaded decisions.
I'm not very threatened by this if this is the dangerous Mythos model - it just seems like a slightly incrementally better sonnet
> Switched to Opus 4.8: Fable 5 has safety measures that flag messages on most cybersecurity or biology topics. They may flag safe, normal content as well. These measures let us bring you Mythos-level capability in other areas sooner, and we're working to refine them. Send feedback or learn more.
These questions are even not about AI: if I were to give money to a human agency and were given something they tell me works, I would ask the same questions. If I did not know how to evaluate, I would hire people that do. With LLMs the verification part is what bothers me the most.
So AI is only interesting to you / your org / humans if it can do things that you can not achieve. But if it still does errors, how could we ever know that super-invention by AI is not wrong?
If we can not rely on the correctness of the result, it is not usable at all. AI must create reliable and correct results always. That was a very fundamental requirement for computing. This problem has not been solved.
Also this is easily solved by .md spec files, this whole "bad code" cope is just FUD'
I get that there's little sense in arguing with the MBA hivemind, but... c'mon.
I manage two teams of highly motivated, largely pro-AI engineers. Both teams have independently concluded that they needed to ramp down GenAI usage because of code quality / maintainability concerns. Both teams have suffered from protracted outages caused by LLM jank not being sufficiently fenced off and guarded against. Both teams have expressed concern that the code generated by LLMs is far too verbose, full of slop, and rapidly becomes an unmaintainable mess.
These are teams that are building non-trivial LLM solutions (deep agentic data synthesis and multi-modal data tagging). They are using the technology creatively and pro-actively, not just vibe-coding slop and throwing their hands up when it fails. Both teams will continue using GenAI coding agents, don't get me wrong - but the gains are incremental, not transformative, and need careful fencing to make sustainable.
Nothing in these articles resonates as real. People who work in reality don't agree. I don't understand why this shit keeps getting attention (or rather I do, but the reasons aren't good).
What makes me excited is that GPT 5.6 (its actually GPT 6) is going to be crazy
In a project like mine (https://github.com/tsz-org/tsz) I am constantly frustrated that models were not doing enough research and were not taking into account other situations. Again and again models would produce code that would fix one thing and break 2 other tests that were "unrelated".
With Fable it seems like tasks are taking much longer (I have not seen a pull request from Fable sessions yet) but reading the transcription of those sessions I can see how it is doing the right thing by not leaving any stone unturned.
As the article says, it's hard to communicate this "feeling" about models because it is very project specific but I thought I share
The first item on the article, the first thing it showed, was wrong though.
It is 100% faster to go from London to New York in 1881 than Volgagrad. Or any of the Russian hinterland colored green or Turkey or Egypt.
"Posterior beliefs about market demand are purely referencedependent: holding dollars raised constant, they track only performance relative to the founder’s self-chosen goal—jumping half a standard deviation at the threshold, responding steeply for the first ten points past it, and flattening thereafter"
Humans generally don't verbalize data this way. The summary document is also very fluffy.
Every sw dev knows this is a very dangerous, and unrealistic, assumption.
What?
Just an FYI this guy is an AI hype-beast. Some of his tweets are truly out there.
I don’t see why working longer is a pro. The results don’t seem much better than you’d get from putting Opus in a long loop.