re-thc

> That's not a conspiracy, that's just how the CCP sees it (ask Jack Ma)

That is a conspiracy. Do you even know what happened to Jack Ma? From what you're saying you don't.

Also that was MANY years ago. The Shanghai stock market crashed. Companies had a lot of fear then yes. Things have changed and repaired. I'd say China in this sense is moving upwards and the US is going downwards in policy.

> You could argue the US has the Cloud Act

No, not really. Your Jack Ma example happened to Elon Musk to some extent. Jack Ma had a feud with the Chinese government as much as Elon had a feud with the US government in the last year or so. Back then Tesla and the other projects all tanked.

> It's the near-concurrent release of the same jump in capability that I find suspicious; not the fact that labs can catch up eventually.

What are suspicious of? If the timing is similar maybe just everyone already are of similar capabilities and got there at a similar time?

> Anyone else find it weird how within 2 months of Fable releasing all the major labs suddenly had Fable-level models?

It means Anthropic had no real moat and no real lead. Is that weird to you?

> It's crazy that I'd literally trust a Chinese AI company with my data

It's crazy how much Chinese = bad the media or US companies have washed into you. Why lump it together?

Like any place and any company there are good and bad 1s.

It's not the Wild West over there...

> AI tends to write less sloppy code than bad devs, and test/validate it more rigorously

I've often had it test and benchmark against the wrong things = no test.

It also writes over-engineered code. So yes sort, maybe.

> but I’ve never worked anywhere (and I’ve worked at a lot of places over 20 years) that actually did it like that in practice

I've seen a twist - not juniors but just offshore engineers.

IPO delayed? Options expiring?

> to satisfy demand from a small number of customers who may not be able to pay them

Because a lot of isn't real demand, e.g. given away for free or very very cheap.

> Comments like this one remind me of all the people saying "China can't invent things, they can only copy the US" 20 years ago.

We still get these comments. Now about AI. China can only distill and are still years behind. Well well...

> And remember the Apple Neo is running an iPhone CPU that is much slower than the M series.

We really need to get over generalization like "phone CPU". Times have changed. Phone or not really isn't the focus.

Per core the A series (cough phone) isn't that much slower than equivalent M series. It's literally the same family just with some extra features, more power, etc.

> What are you doing where you're getting Sol on XHigh or Max to run out on the $200 plan?

Real work. $200 looks good on the outside until the essence of it, e.g. the models lie. I gave a list of spec to Sol and Sol decided some items didn't need to be done and the reason was "unproven", "not enough evidence", etc.

They all come up with amazing ways to lie (or be lazy). Often times what you get isn't what you asked for (only on the surface). E.g. I ran it to iteratively bench and optimize a better data structure for the project. It spent hours and finally came up with something. When I check it out -- it benchmarked the wrong criteria and was way off. So here we go again. Most AI work looks good on the surface. There are infinite edge cases.

So to do real work and gate it you need to:

1. Plan

2. Get it to do the work

3. Get independent agents to check from different angles

4. Take that feedback and get it to fix those gaps

5. Match against the plan and redo parts if needed

Every task is easily 4-5x the estimated amount of tokens.

p.s. well I did burn some banked resets building a compiler for some language AND it is still NOT done. Every time it says done I say check it says ok we still have bugs...

> Not sure about the size part, if Kimi K3 achieves SOTA performance at 2.8T parameters, western models being >2x that size would be insanely bad in regards to efficiency. I bet they’re all within the same order of magnitude and below 10T and won’t really have a reason to go even that high for the foreseeable future.

They're a lot larger e.g. Fable. It is insanely bad. Do you know how much more resources "Western" companies have? Most in China don't have random GPUs to "play with" like every "frontier lab" employee does.

> As investors will start squeezing them for profitability, I suspect focusing more on efficiency will be commonplace.

They're born lucky though. Efficiency is "free". The next generation hardware e.g. Nvidia claims Blackwell -> Rubin is 10x efficiency (verified by Neoclouds apparently).

> This matches my experience with DeepSeek V4 Pro at Max reasoning,

That was ages ago (in LLM release timelines). DeepSeek V4 Flash beats it now and a lot cheaper.

> On similar tasks, GLM 5.2 at Max reasoning screwed up maybe 20-30% of the time,

GLM 5.3 bridges this gap.

> I'd say as Chinese models get better, whatever moat Anthropic and OpenAI have dissipates.

Their moat, especially OpenAI is funding and hardware resources. They gain train models 10x as large and also serve at large scale. That's it.

> Didn’t deepseek recently announce prices will go up significantly?

They also previously said prices will go down significantly once they get a hold of the upcoming Huawei chips (later this year).

Prices are going up just because they can. It can easily come back down. They aren't strained by some IPO / VCs requiring them to 1000x their earnings.

> The problem is that they need actual chips tomorrow.

They need it today. Best if it was NOW!

> It remains to be seen who is actually willing to pay for it at the prices required to recoup those insane investments.

Everyone. Open models are going to keep prices down. A lot of current models are more than usable. Self-hosting would have been an option if hardware prices get back to sane values.

The insane investments have to do with insane over-valuations, VCs involved and hence media spam on it. Chinese labs for example make do with 1% of the valuation and 1% of the resources.

> they can replicate or beat the price with rented GPUs

They "can" is the caveat here. Rented GPUs are going up in pricing. I recently got an email that DigitalOcean pricing of GPUs were going up.

So

1. They have to get a hold of them (availability is bad)

2. They have to maintain the pricing

> In the $150/mo range you can get effectively unlimited usage of GPT 5.6 Sol (Pro plan)

Not true. Sol on XHigh or Max runs out even on the $200/mo plan. It's not close to effectively unlimited. Maybe at 2x the current allowance it can.

The other commenter said it, quoting "Losing Ive was probably a good thing for Apple."

So the parent comment would imply that losing Dean is a good thing for Google, which is way less likely here.

> Data centers, TPUs

Those 2 are definitely NOT Google's strong points. Maybe by means of marketing.

Oracle, Amazon, Microsoft, Equinix and many more are in the data center race.

As for TPUs... Broadcom, Mediatek and all the other partners you hear less about are likely more important. Google just has the flashy media outreach.

> So far even losing Jobs didn’t hurt Apple as it looks.

Who knows? Pure speculation? You can also say if Jobs was still around they could have 10x-ed it even further?

Apple car could have been a thing? Apple could have been way ahead and actually competing in AI and data centers? Who knows what else Jobs could have came up with?

> It's like Apple losing Ive in a way.

Are you sure we should compare it like this? Not sure it implies what you think it does...

> People are having a rough time figuring out that good design and fundamentals no longer align with the business side of things.

That's not true and wasn't true in the sense that businesses don't care directly but they are indirectly. Are you saying business doesn't care if things go down or it doesn't work?

At the end of the day people don't care until they do and then they blame you.

> The culture is different

That's partly. I'd say the other part is it's an open secret they have "illegal" Nvidia GPUs and other "secrets". There's a common understanding not talk about these things because it'd hurt everyone collectively.

Google Frozen v2 chips hardwires the underlying architecture but leaves the weights to be configurable.

> I wouldn't doubt GPT 4.6 Luna being in the top left quadrant's center on the Cost per Intelligence Index is not concerning for Liang Wenfeng.

The leaked interview has him saying it doesn't matter... as much as open source doesn't matter. There's enough in it for everyone right now and they aren't after everything.

Perspective: DeepSeek doesn't have enough infrastructure to serve their target customers already.

> It seems China is already able to do DUV a lot sooner than others expected.

That's the media and in particular US KOLs of all sorts driving the wrong impression of China and other places. China and many other places for example have fast public transport that the US doesn't and can't even imagine today. They're not behind.

China's DUV still isn't that production grade (mass produce-able) so don't get that hyped up the wrong way (in a different direction).

The whole China-is-behind with tech and in particular semi wasn't that they can't. The truth is they spent decades in internal politics and corruption. That all got solved with the bans, so thank the bans! Jensen even said the bans were bad.

> Yes but there is a big, big market for subagents to consume lots of tokens cheaply and condense information up to parent agents.

That's again not some "intelligence factor" here. Different agents work for different use cases. Luna wins some. Terra wins some. Sonnet wins some. Flash was really good at exploring.

So I'm not sure what your point is? There's a big market for everything. Even within the market you describe it's likely not a Luna-size fits all either.

> When model intelligence reliably hits 90%-95% of current day knowledge worker tasks, they are going to burn those weight to silicon

Google is already working on a similar idea but more "flexible".

> Luna is an extremely strong model.

By benchmarks, which sadly is a poor measure. Yes Luna is a good model under certain circumstances. Whether it is great for general usage is another story. Sonnet is definitely better when prompts are more vague and it needs to decide things. Luna generally sticks to things very strictly and goes off in bad ways.

> Seeing spikes like this makes me question about where the floor really is.

You mean they increased the price and then cut it back and now it is amazing?

Luna had a price hike vs mini (its previous replacement). The cut now just puts it back in that ball park.

Not that this isn't good news, but what's impressive?