I highly recommend to anyone that uses them a lot - try to clone your own. I’ve got mine fully formed into how I want it, with features I haven’t seen anywhere else yet and with integration into everything I work with. We’ve lived in a world where every app had to fit enough people well enough - but now we live in a world where an app can be molded to be absolutely perfect for just you!

But dont you have to be faithful to rl environment in which it model was trained in the harness?

only labs knows what the rl enviroment was so they will always be one step ahead.

There are but then you're required to pay API prices for them which can end up costing significant multiples of the price of the subscription.

American labs will do fine with monthly subscriptions costing $20-300/month. They will not do fine for heavy basic LLM API calls -- it is for this that the cheap API produces substantial savings. Specialized niche provisions, e.g. excellent tool use, reliable structured output, reliable long context, cached and uncached web search, document reads, computer use, integrations, etc. also will take time for the foreign services to be good at.

If you remove the word "AI" from your post it also describes the last 20 or so years of US/Chinese economic competition. They are real, serious competitor but also there is a lot of value in that last 10%. Some buyers need the best and you can charge a premium for it.

The speed in which a copier can catch up to a leader is faster than every before, but everything is faster these days. I think the threat is if can these labs catch up and pass American labs while keeping costs low, but I don't see that happening.

China isn't undercutting anything. Kimi has to be hosted somewhere and is extraordinarily expensive to operate at Fable-levels of capabilities. The moat is (and will always be) the cost to actually operate these models, and that cost will continue to soar, it will not reverse, and the hardware will never keep up such that at-home use somehow becomes economical.

It's like releasing a cutting edge search engine, proclaimed to be on par with Google. Oh yeah, I forgot to mention, you need to actually run it somewhere. One of the core moats for Google has been the cost to scale and compete globally with a search engine, which only a few entities on earth could even attempt. Somehow the HN crowd forgets this every single time. The same claims were endlessly made regarding Uber, as though the app was the hardest part of competing with Uber. Here's a hint: the model isn't the only super hard part of the total process at this point. Try running something akin to GPT 5.6 for a billion users, see how much that costs.

The winners will be the owners of the subscriptions and API usage. That's going to be OpenAI and Anthropic as of now, possibly Google will stay in the fight with their scale advantage focused on the consumer. It will never be the open weight models out of China. Who is going to invest the trillion dollars into infrastructure needed to run the latest version of Kimi 3-5 years out, for a billion users? There's only one entity that would ever consider it, and that's China domestic.

And the real money (for a sustainable business model allowing for constant massive reinvestment & infrastructure maintenance) is in pairing up advertising, which is another spot where the open weight models from China will never compete. It doesn't matter if the HN crowd hates this outcome, nothing will stop it from being the future core of mass consumer usage. And the business use will stay with subscription + API use as well: the best option for that today is OpenAI and Anthropic.

People endlessly claimed the same thing about DeepSeek upon its release. they claim it every single time a very large cutting edge model is released out of China. It's a combination of propaganda, some truth, and the faux open culture desperation to see big tech get destroyed (somehow better if big China takes its place instead).

A good UI/UX is noticed and appreciated by many. A great UI/UX goes completely unnoticed by most.

Yeah, but maybe Kimi doesn't pretend to know better than I do. Like, I just had a task today where I had a silly form where I wanted to copy in a signature from one document to another document. And, you know, I could have just used some other dumb tools, but I decided, hey, let me us ask Codex to do it. But it started lecturing me on how we can't do this kind of stuff. And it was just pure silliness. So hopefully Kimi Work doesn't try to pretend that it knows better and lets me decide on ethics.

This blog post here summed it up the best. These are tools, so let them be tools.

https://geohot.github.io/blog/jekyll/update/2026/07/11/ai-20...

As opposed to Claude which has flagged:

- anything related to piracy

- legit security work (audit my infra, pentest it)

- disabling a safety mechanism on my food processor

- reverse engineering

- creating style guides inspired by other games/companies (for fan projects)

- questions about how nerve agents work and how they can be treated (for a layman)

I'd prefer it uncensored too but this objection was insightful in 2023. Everyone and their dog knows they're censored on topics sensitive to their country, much like western models are censored for western woke sensitivities. Nobody cares anymore as long as the reasoning is good and the price is low

It's an open model, abliterated versions will appear on HF in the fullness of time.

Yes, but the answer is an indirection.

“How does Kimi Work protect my privacy when accessing local files?”

It doesn’t protect your privacy. It’s like asking, “how do I know you’re not spying on me?” And getting the reply “we cannot physically enter your house.”

It is, but I really think this project would be better as open source. They are already all-in on open for the most valuable part (the frontier model)! They might as well make the harness fully transparent so that questions about what's being accessed and transmitted can be avoided.

I personally wouldn't use any agentic harness from any company, American, Chinese, or otherwise, that is closed source. The Grok fiasco shows why.

The repo you’re working on, sure. But all files?

No, it's Apache 2. https://github.com/openai/codex/blob/main/LICENSE

I guess the issue is they didn't give attribution?

Not if the harness applies an OS-level sandbox to the process, as Codex does.

https://learn.chatgpt.com/docs/sandboxing?surface=cli

Words to live by.

You mean other than the zillion dollars of infrastructure you need to run these systems for a billion users. And the extreme cost to sustain that operation day to day. And the cost to upgrade the infrastructure forever.

Remind me which entities (govt or corporate) are going to subsidize that forever. Remind me which entities are going to front the cost just to scale up.

By that premise - wherein you forget everything other than the first ingredient (the model) - Google has no moat either (got $100 billion dollars lying around?). Nor does Uber (have tens of billions in your bank account to clone their business?).

I'm certain building an app (Uber), or building a search engine (Google), is the super duper hard part.

Slapping a GUI on Linux will kill Windows, any day now.

Looking at these pages reminds me of browsing jellybean components on DigiKey. Amazing for the consumer, but one does wonder: What are all these DotAI infra companies’ plans to justify any of their valuations if they’re just selling commodity compute on OpenRouter?

Will it? Bedrock doesnt have deepseek v4, GLM 5.2, Qwen 3.6, minimax m3.

That's precisely the problem.

People are building UI/UX for a paradigm that is so utterly new that they miss the mark entirely.

Everyone defaults to a big chat input because that's all they know. It's the expectation, so it's safer, faster, and cheaper to not deviate.

It's also wrong and bad for work.

That might even be a beneficial move, creating (well, actual, not artificial) scarcity works well to draw interest usually. Curious how this one plays out.

I've been subbing to Claude for 2 years and somehow the questions of whatever happened in Tiananmen (I know already anyway) have never once came up, but Claude (especially Fable) is now famous for refusing to work on things people want. And even pushed back against me when I was doing asking to research some people famous (infamous) in Israel-Palestine conflict. I really don't get why these "chinese gotchas" are keep coming up in every thread, they are mostly entirely inconsequential to getting actual work stuff done while same can't be said about Fable or others. Once, Gemini refused to help me make an infographic about women in STEM because the specific woman I was doing research about has died a few years ago, other models had no such qualms. All these "gotchas" are quite arbitrary.

So for your Tiananmen questions go to Fable, and for the Fable-nerfed subject matter go to Kimi (or GLM, as Hugging Face were forced to do when they were recently hacked and the nerfed US models refused to analyze it).

I hope they don't bring Chinese UI/UX to it. Native Chinese apps are a mess from a western perspective.

Kimi Work apparently

First of all to reply to your deleted comment, yes this is completely relevant to the article.

As token costs drop significantly (thanks to local models and hosted Chinese models.) employers will consider using cheaper models like Kimi Work and others, even if they are already using expensive ones like Claude Cowork. They won't be able to operate without it.

Secondly, do I actually believe this? Of course, I already said that there will be more mass layoffs because of AI years ago. [0]

So this outcome was expected.

[0] https://news.ycombinator.com/item?id=42490692

How to make an apple pie from scratch with no copying:

Step 1: invent the universe

Perhaps, but "we run the exact same business as Anthropic and OpenAI to the pixel at 0.1% the price" is going to raise eyebrows.

I do agree with the UX idea of familiarity, but copying the exact salutations, etc. just seems egregious.

Why not let the model speak for itself and contribute to an open harness?

Pedantic.

> A narrative is a structured account of connected events, experiences, or ideas, commonly referred to as a story

Regardless of intent, it is how you tell your story.