Curious habitant

Curious habitant. The Ultimate paradox: Computable machine being first to prove consciousness to be non-computable.

[Curious habitant @thkuu]

@nunezvice thx, it clearly is a problem. the usage drained much faster than usual during past week or so

[Curious habitant @thkuu]

@morganlinton How is claude comparing against 5.5? For me in swift project i have to say claude performs clearly better. Still in 20€ plan tho.

[Curious habitant @thkuu]

@hqmank Yep. Also feels like model capability is not what it used to be. Jumping to opus 4.8 has helped a lot to progress with my project lately.

[Ashwin Gopinath @ashwingop]

Claude Tag is a Trojan horse.  Not because Anthropic is doing anything evil. Because the incentives are obvious.

Day one, this looks like a great feature: tag Claude in Slack, let it follow the thread, remember context, connect to tools, break down tasks, chase work, and act like a teammate.

But that is exactly the problem. The moment your AI vendor becomes a shared coworker, it stops being just a model provider. It starts becoming the place where work is interpreted, remembered, routed, and eventually executed.

That is not model lock-in. That is context lock-in. You are now renting your company back from them.

Models can be swapped. Agents can be copied. But the memory of how your company actually works is much harder, maybe impossible, to move: the Slack scar tissue, the exception paths, the customer promises, the unfinished threads, the weird workflows, the implicit owners, the “we tried that in Q2 and it failed” knowledge.

Once that lives inside one vendor’s agent layer, you are not renting intelligence anymore. You are renting your company’s operating memory.

And the pricing model makes it even more dangerous. A human coworker has a salary. Claude has unbounded tokenized activity. The more work moves through it, the more the vendor captures not just IT spend, but labor spend.

This is the enterprise bargain people will regret: Convenience now, and rapid decent into dependency.

The right architecture is simple: rent the best intelligence from whoever is best this month. OpenAI, Anthropic, Gemini, open source, whatever. But own the context layer.

Your company memory should be inspectable, permissioned, portable, and model-neutral. It should not be buried inside the same vendor that sells you the intelligence and the workflow surface.

Claude Tag is useful. That is why it is dangerous. Rent the intelligence, but own the context. Or, regret later.

[Armin Ronacher ⇌ @mitsuhiko]

When I struggle to structure my thoughts about what's happening I turn to writing. Today about the recent US Anthropic ban news, what it says about power and dependency, and what it should mean for Europeans and citizens of the world. It's a long one. https://t.co/6dpw0QOQeO

[Curious habitant @thkuu]

@0xAIGOATexe EU AI is the thing now. Mistral will hire some and the future looks bright for them.

[Curious habitant @thkuu]

@BoomBeach Operations need update. More stages, new more challenging ops please.

[Curious habitant @thkuu]

@deredleritt3r @cremieuxrecueil Yes I heard but never tried it anything else but coding so a bit doubtfull. I will give it a try in business licenses for our team. I think it will be huge success for our ccompany if this works out

[Curious habitant @thkuu]

@deredleritt3r @cremieuxrecueil Did you need to create a lot of instructions for codex not to start writing code but instead help you in otherway, the workflow you mentioned? I plan similar thing but not for legal tasks

[Curious habitant @thkuu]

@MrMikeInvesting Agreed. Except you should add EQR from Australia to rare earths! That's tungsten mining company that has actual production and largest miner in west.

[Curious habitant @thkuu]

@vontuchman @noxflux Apple siksi, että ovat kohtuullisen pienellä AI satsauksella kuitenkin hyötyneet siitä(Googlen AI sirin moottoriksi). Google myös hyvä valinta koska vertikaalistäkki. NVDA todennäköinen myös lupaavien uusien mallien takia, no wall näkyvissä->chippejä tarvitaan enemmän ja enemmän.

[Curious habitant @thkuu]

@Stickstrom @anteropaljakka1 Samat kokemukset. Itsellä kyllä koodaustausta, mutta en ole koodannut 20+ vuoteen. Nyt tehnyt kaikenlaisia juttuja claude, google, codex cli:llä ja nyt siirtynyt codex appiin joka ihan loistava. Niillä appi ja ohjeet IT:lle firman sis.julkaisuun, toiminnut hyvin.

[Curious habitant @thkuu]

@francoisfleuret In the context with this quiz the answer is no. If there would be more context related to guardrails the answer could be yes if guardrails were good enough. Like Asimov's 3 laws + no harm to humanity is allowed.

[Curious habitant @thkuu]

@vontuchman Spot on => korvaa ajattelutyötä, ai avustaa omassa käyttöönotossa, kiihdyttää omaa kehitystään. Tämä on todella konkreetisesti havaittavissa jos käyttää itse AI:ta vaikka koodaukseen. Erinomainen kiteytys!

[Curious habitant @thkuu]

@villepeltola Viikko sitten vibe koodasin työkalun firmalle. Tänä viikonloppuna se on intranettiin käyttöön otettu. Web pohjainen. Deployment pläni myös suoraan AI lta. Onnistui hyvin, ohjeessa oli IT:lle step by step ohjeet jotka he kiltisti toteuttivat.

[Curious habitant @thkuu]

@Fibonazzo @Alexsei88 Ajattelen, että EQR alihinnoiteltu, siksipä lisäsin aamulla. :)

[Curious habitant @thkuu]

@gdb Fully agree! I built r&d tool for our company during the weekend. Still need polishing but it already has advanced features like automatic spec intake, spec creation, food legislation checks, declarations , claims etc. Took about one working day.

[Curious habitant @thkuu]

@DeryaTR_ 5^1108

[Curious habitant @thkuu]

@DeryaTR_ My prompt to gemini 3 pro gave solid 5^11118.

[Curious habitant @thkuu]

#UusiProfiilikuva