> I just tend to thing bullies should be crushed rather than appeased

Cool it with the antisemitism. Netanyahu's actions were short-sighted but understandable, labeling him a "bully" is not fair.

Theoretically you should be creating a "read email" CLI tool and letting agents interact with it in a chroot sandbox.

LLMs are much more proficient with bash and --help than they are with bespoke API protocols.

Treat LLMs like you would a junior programmer - keep things as generic and obvious as you can.

Having "tokens and secrets" at all is a lack of security hygiene.

Are you seriously implying that technical debt is something that doesn't exist or something that managers don't care about??

> When agriculture was invented, there were mass layoffs of hunters and gatherers.

Uh, no. Not at all.

Agriculture absorbed the extra population that used to be unsustainable. There were no losers.

Plenty of losers. See native americans. Sorry about your generational hunting ground, white man would like to ranch cattle and therefore you should leave or eat lead.

Native Americans had agriculture. This isn't about agriculturalists vs hunter-gatherers.

P.S. Moreover, they had very advanced agriculture - most of the food Europeans eat today are adopted Native American designs.

I mean, America had agriculture[1] before Columbus showed up. Cahokia[2] came and went (and then other people moved in).

[1]: https://en.wikipedia.org/wiki/Three_Sisters_(agriculture) [2]: https://en.wikipedia.org/wiki/Cahokia#Abandonment_and_resett...

It's a typical colonial take on agriculture. People flourished without it but we seem to be stuck with revisionist history around the topic.

People also seem to think hunter-gatherers were always struggling, and while it did happen, they seem to miss that there are still billions of people lacking food security because agriculture isn't as easy or prolific as we tend to believe. It's really hard, especially in certain parts of the world.

Oh well, these biases permeate everything. I have plenty of them too.

No. Body. Knows.

> AI enables velocity both in development and bugs fixing

Or so they say. You'll have to trust those vibes blindly, because double-checking these claims apparently makes you an anti-science luddite.

The subscription plans are the "first hit is free" plans. They're not gonna last and don't build anything serious based on them.

> This defies belief.

Akshually it's pretty normal for 2026.

Yup, since the rise of AI, I still didn't find any product built by it that's even remotely useful to me... People build prototypes a lot, but these are usually buggy and not worth putting money into.

They're only good at it because they were trained on massive amounts of English and French data.

Not really true.

Both Claude and ChatGPT can translate into minor dialects of Norwegian they will have seen very few works in because very few printed works exist in them.

E.g. I've tested both my local spoken dialect, which is rarely written, and a sociolect used by a 1970's Maoist group consiting of a few hundred people, where most of the printed material consists of novels from a couple of ex-members that became authors.

In the latter case, it claimed to not know, but was able to get a good match from just a description.

I also just had it ape Norwegian orthography from the 1910's by having it look up the rules and translate a text it had first translated from English to modern Norwegian, and it did just fine.

They will have seem some work in these dialects, but mostly it transfer really well to know related languages (English, Dutch, German, Swedish, Danish, roughly form a continuum from least in common to most in common with modern Norwegian; they all share vocabulary and significant parts of grammar with Norwegian), and then a relatively limited exposure to Norwegian itself is sufficient to do fairly well.

They're also really good at "style transfer" of text in the form of tweaking orthography, word order, and minor grammar changes from descriptions and examples.

(incidentally, the latter is one way of getting an LLM to sound a lot less like an LLM)

This is all true, but I assumed the original posters were talking about cultural knowledge, not linguistic correspondences.

To do translation well you still need cultural knowledge. (E.g. the particular modes of specific kinds of legalese, or slang and the nuances of social class, etc)

I think it's not that this knowledge isn't present in the model somewhere, but probably more that it gets killed by instruction tuning for US corporate values.

> meager hardware

Qwen was made on a cluster about that size.

And this is before anybody ever thought about optimizing the training process. (Currently it's just pytorch analyst-as-coder slop, with extremely overprovisioned quantizations, etc.)