furnaces

intelligence was never the moat in consumer ai. ui, ease of access and muscle memory are the moats. I can guarantee that if a novel chatbot released with superiour benchmarks even after a month, no more than 1-2% of all LLM users would've heard of it, let alone use it.

@rxhit05 yes

Am I the only one who feels Claude Fable didn't live up to the hype?

@andrewqu yeah, without a doubt it's only really in writing, philosphy and (obviously) coding where remarkable improvements were noticed.

Hot take: a lot of people wouldn’t be able to tell the difference if they were randomly routed between gpt-5.5, opus-4.8, or fable-5 for their day to day work

@tunguz just found out about it aswell biotechs gonna be disappearing left, right and center

I think I finally figured out what yesterday’s Fable ban was all about. None of the official reasons make any sense. Something really big happened yesterday and this was a direct consequence of that. This is some next level GoT sh*t.

@Youssofal_ the only benchmark mistral beat fable on was the bureacracy benchmark

No way!!!!! don't need Fable any more soon!!!

@robj3d3 @POTUS ah yes, reverse pyschology 101

Fable was actually shit. So shit in fact that if the US lifted the ban it might just trick China into wasting a lot of time distilling its weights. Lift the ban @POTUS

@BoredElonMusk the trick is to not tell anyone how much you are worth.

Quick question. What is the absolute richest I can be in America without anyone lecturing me about having too much?

plot twist: Anthropic moves to China.

@jiratickets thanks :)

@furnaces i like your username and pfp

@JesseLucasSaga oh how we have declined

Again thinking about when the Church of England accidentally set up UBI for sensitive young men only and spawned a race of unprecedented geniuses

@realfunnyeric brb omw to the US

USA USA USA 🇺🇸

@iamtrask how does this work? does one plan and the other execute imma read up on this.

This is a *way* bigger deal than it seems... Frontier AI companies will *never* own the frontier again I kid you not... I've been waiting for someone to show this result for like 4 years... this is a huge deal. The short reason: combinations of models will *always* outperform individual models The long reason: this is the gateway to a million times more data... and huge leaps in compute efficiency. The AI scaling laws always win. More in article below 👇

@CooperZurad maintaining software also means keeping up with the times, pushing updates, fixing the bugs that will inevitably occur with the updates but I agree, now with vibe coding it feels as if "maintaining software" isn't as neccessary

can anyone ELI5 why software needs to be "maintained"? Like why do you need whole teams working on stuff that already works? Hardware doesn't work like this If it functions, it functions and does not need to be changed unless you're replacing something worn out.

I asked claude to generate a UUID. I don’t understand the joke, if there is one?

@KevinVanCott nah I just refunded

Anthropic is making a ton of money from this ban, right? So many people signed up just to use Fable over the past week, and now Anthropic gets the subscription income without having to spend on the inference at all.

@rileybrown why are people saying "back to Codex" as if it doesn't take 2 seconds to switch

Back to Codex.

name a CEO with zero haters I'll go first

@TTrimoreau there is a reason why this opinion is unpopular...

Unpopular opinion: Vibe coding is actually hard work too.

what's stopping us running fable through an americans API key?

@bindureddy I truly want to know your ulterior motives because there is no way you actually believe this

Kimi 2.7 beats Fable and GPT 5.5 on agentic coding benchmarks Yes, some of it is bench mark maxxing but a lot of the performance gain is very real! China will over take US in a few weeks if the US is dumb enough to ban AI 😬