If you're not billed for usage, anyway.

Otherwise, for the other 99% of folks, that attitude is of course a pit trap that captures code bases and makes them maintainable only through the providers -- presumably one or few -- with a rich enough model to keep up with the growing mess. Preserving a code base that's legible, organized, and fundamentally maintainable by both humans and trailing commodity models is of imminent concern for anybody who doesn't want their margin strangled by your employer once it's too late to have other options.

As frontier capabilities advance, the details don't matter less; they matter more.

Any given model will always have some distinct implicit voice that its biased towards for that infill content, and so a popular model will always become exhaustingly common, painfully familiar, and cliche. Users can use more elaborate prompts that shift the voice away from the most normative and towards some other nodes, but they need to put in special effort for that, and what people-at-scale specifically want from these tools is to put in very little effort, so we can expect that overwhelming number of casual and naive users will always be generating cliche slop with them.

Code escapes this problem not because of training but because it specifically benefits from cliche (boilerplate, patterns, etc) and so an model whose code "voice" reflects your own taste as a coder (or your toolchain's taste as a vibecoder) is going to feel like productive output rather than slop. But it's still cliche.

The infancy phase of this technology is represented by the pursuit of making wildly grand, wildly expensive, all-purpose models that somehow discern a user's full accurate intent from a lazy, underdeveloped, vague idea that they ambiguously and poorly express in a couple dozen words.

The adolescence will arrive as those outsized and ill-considered ambitions collapse and we instead see a cambrian explosion of restrained but efficient model+harness-tuples that have been distilled, finetuned, and rigged to deliver on narrowly scoped but idiosyncratically-shaped tasks with incredible efficiency and erogonomics.

The more reasonable comparison is against rented GPU's, while looking at tradeoffs in latency and upload/download/storage/instance management overhead.

Buying hardware for local models is meeting a wholly different need than buying tokens through OpenRouter or whatever.

Sure, $4000 can be a lot of if you're a casual hobbyist or are struggle to meet everyday lifestyle costs, but it's definitely not "insane" if this is the trade you make your living from or if you've established a lifestyle that affords disposable income for your hobbies.

And for some people, $4000 for a device you have complete control over and can repurpose and tinker with to your own needs and curiosities is a much much more justifiable expense than a $200/mo rental for some narrow-access tool that somebody else controls.

They're individual testimonies, often nearly impossible to verify, and written in an era where "autofiction" and "creative non-fiction" are mature commercial genres with their own tropes, standards, forms, and conceits.

If you're intellectually curious about "cults, scams, and schemes", there's a whole other -- drier but more earnest -- world of nonfiction to be found in places like JSTOR or the catalogs of academic presses.

You try to peer through the noisy results to see if any of them are actually a match, you try again with different terms, you give up and engage with a result outside your parameters, or (best of all) you give up and engage with an ad or get distracted by some other Facebook feature.

As far as they're concerned, engagement is the key factor and satisfaction is only a tiny little bonus when things work out. Engagement is what earns them money and makes them make bigger claims in investment statements and is what keeps eyeballs away from their competitors.

Everything is working perfectly for them... just not for you, and they don't really care. That was the big business insight of the 2010's and there's still very little reason for anyone in online media to stop aiming for it.

While maybe that boundary can be fuzzy, we're unequivocally and deeply in "metaphor" territory here.

A different and externally/phenomenologically identical way to look at it is that the AI is just a program that generates output too unpredictable, too voluminous, or too idiosyncratic for people to evaluate. When people submit their own will and their own intent to that program, by treating the black box as an intelligent oracle, they've entered into a psychotic state divorced from reality.

It's two self-consistent and coherent perspectives of the same event, except one involves believing that scifi AI has arrived and the other just thinks people can be dangerously stupid and credulous.

It suggests that they're worried that key talent may get lured away by better liquidity at Anthropic, Google, etc before they can secure the offering they want to make. And so they're loudly making a non-binding suggesting that it's just around the corner and that if they stick around and work extra extra hard, it may come even sooner.

The subtext behind that statement is anxiety, not confidence.

Because the field of artificial "intelligence" is constructed around the idea of applying psychological metaphors to computational systems (a very powerful idea!) it's almost a worst case scenario for these people.

Suddenly, they're reversing the metaphors and applying computational schema to psychological processes ("aren't we really just stochastic parrots ourselves?!"); or, like here, they find themselves surprised and confused when they stumble across the natural boundaries of the metaphor experimentally.

It's because they never had sight of the boundaries in the first place and maybe never can quite see them. The words only make sense to them as literal equivalence, and so their surprise when they run into stuff like this is earnest and deep.

It's a business line more suited to finance, law, sales, and accounting people than tech people and a pretty laborious one. That's often the case when the tech looks "easy" but the sector only seems to have a few big winners.

In this particular case, OpenRouter represents a M x N bridging and enhancement layer much like Stripe themselves, and so must address a lot of parallel technical, dealmaking, accounting, and legal challenges.

It may not be obvious on the surface because they seem to be working in such different domain, but they have to address a lot of the same problems in comparable ways and that makes it a pretty darn good fit.

I think you're very wrong.

I think many of the people who "support these bans" do recognize value in LLM's and use them as tools in the kit applied to their workflows but recognize that rampant abuse and misuse of those same tools threatens to ruin (or already has ruined) many community projects that had previously been healthy and thriving.

And so they find themselves insisting upon unambiguous firewalls as an existential necessity for the communities they have invested years or decades of their lives into.

How do we figure out who's right?

They may have utility in trying to look at the whole landscape of models, but are very misleading when it comes to making 1:1 comparisons or in developing confidence at to how a given model will deliver on your workflow.

They originally broke through and defined themselves by insisting that vendors only sell original cottage-crafted work, but gave up on that requirement years ago and the sincere low-volume, low-margin crafters can't compete with high-volume, high-margin importers in either advertising spend or in the etsy-profiting metrics that drive algorithmic discovery.

But others here have been truly naive themselves or have journeyed with intimacy and openness beside a parent, partner, or friend who really just doesn't get it.

The reality is that there are a lot and lots and lots of real people in the real world who sincerelt just don't understand the terms they're agreeing to, how their own finances work, what's plausible in their own financial future (near or long term), etc -- and the companies the design and promote convenience loans know this and they target these people like sharks hunting prey.

They use any and every trick they can get away with to lure these naive people into agreements for which all outcomes (payment or default) favor the shark.

Without regulation of both loan design and presentation, lending naturally become a vehicle usury, vaccuuming assets and opportunity from the many earnet people who will always be too credulous, too trusting, too naive, or too desperate to resist.

Blogslam has become cheaper, more voluminous, and worse.

Bot operations on social media sites and forums have become cheaper, more voluminous, and worse.

Advertising content has become cheaper, more voluminous, and worse.

The content of streaming music, online art communities, and video platforms has beckme cheaper, more voluminous, and worse.

The content of public code repositories, and the issue and PR channels associated with those repositories, have become cheaper, more voluminous, and worse.

Customer support services have become cheaper, more voluminous, and worse.

When individuals use AI themselves, towards tasks that they mean to acheive, many of them are individually happy with the results. But pretty much everyone who looks around sees that AI in aggregate is quickly saturating almost everything to which it can be applied to with endless, suffocating slop.

It feels great to pull the handle on the slot machine. It's pretty grim when you stop for a second to look around at the casino and take in the big picture.

Which is fine for you, but doesn't change that many people can get more insight out of your original text than they can out of the generated adaptation, and can do so with more interest and less fatigue.

And indeed, you don't have to care about that. But one would think that in publishing or sharing your writing at all, you do in fact have some investment in whether people receive those insights that you have. Otherwise, you could just keep the AI writings to yourself in the first place.