"so $ANT and $OAI have been slashing token prices (nuking their margins) to stimulate demand and keep ARR" This isn't true. Price cuts are mainly due to efficiency gains (smaller model, better spec decode). They haven't nuked their margins at all.

Very important point on sovereign buyers I think we may see a return to the Biden-era AI diffusion rule Democrats will push for it There is bipartisan support as well Trump will be under pressure, and Jensen's lobbying won't be effective

Today we’re releasing GDPx, a benchmark for measuring agentic capabilities across the most valuable domains. GDPx is a true frontier-level benchmark: even the best model scores just 12% (1/7)🧵

This is extremely surprising CXMT has already validated the engineering samples for its G6 node in Q3 This means that they will start MP of G6 within a year G5 entered mass production in September They are spending a lot on R&D and are reducing the tech gap at a much faster pace now I wonder where the gains are coming from Is it due to them transitioning to 4F or just more multipatterning

UPDATE: Remember when Bloom claimed its supply chain is “not dependent on China” after Hunterbrook’s reporting?  New trade records show $BE seems to have quietly restarted a Chinese supply route for scandium-containing fuel cell parts.

Bull cope coded but the fact that people actually think it’s over, at the exact same time that agent teams are emerging as a new vector of the Scaling Laws, shows just how much people rely on stock prices to inform their views and just how little they understand the technology. Bulls need to believe inference is a multi-trillion dollar TAM in the next few years. But guess what? The capabilities are utterly there. There is little doubt for anyone who understands the technology that AI already is - and certainly in the near future will be - capable of doing a large share of tasks of a large share of jobs in the economy. Digital labor is very, very real. John Collison is telling you Stripe is spending 1% of OpEx on tokens with line-of-sight to “tens of points” over time. Anthropic, at >$70B ARR, is telling you their customers spend 0.1% of OpEx on tokens. Is Stripe a leading indicator for the rest of the economy? The question is rhetorical. The direction of travel is blindingly clear. And Stripe is far from an anomaly - the large delta in token spend between AI/digital-natives and the aggregate economy is a very clear signal about where budgets are headed. Aggregate AI ARR + AI-driven digital ad uplift + Hyper cost avoidance is likely ~$400B growing 3-4x y/y. In what version of the world is this number not $2-5T in 3-5 years?

Reddit call was absolutely fascinating. Everyone should listen to it as a bellwether for where the internet is going - AI Mode is increasingly default Google experience - AI Mode provides rich info and data in line, less incentive to leave search - AI Mose incentivizes dialog, Google wants to keep you engaged to serve more ads at each turn - Google will milk search for as much revenue as possible as they transition from a search company to a cloud service and chip company - As traffic from humans becomes more scarce this will put Google in an increasingly adversarial position to the rest of the internet - This is a massive headwind to all media type companies who are not destinations - This is a massive headwind to all content-driven go-to-market motions especially in the SMB. You have seen companies like HubSpot and get crushed by this funnel collapse We are entering the end game - a knife fight for the remaining and dwindling volume of human traffic on the internet

- CY26 WFE raised to $160B (from $150B+ in July) and CY27 to $210B+ (from $190B+) - Leading-edge logic prober shipments are expected to grow 60% y/y; expects >200 advanced packaging probers in both FY27 and FY28

> Pricing in China has become so competitive that it pushes Chinese labs to find shortcuts, i.e., distillation I disagree with this completely, "involution" is typical China watcher thought terminating cliche and particularly meaningless here. First, the logic of distillation is vastly simpler, American labs are far ahead and their models are not just developed with a lot more compute but make use of the high-quality data Chinese labs haven't had the opportunity to acquire yet, so using them to accelerate the progress is a natural solution. The core pressure is *international competition*, not domestic margin slashing. Claude is still massively popular and regarded in China, despite Dario trying his best to reject this income. If Chinese models fall further behind, the Chinese consumers will be utilizing Western models even more; and the only response would be to, indeed, slash margins and compete on price alone, in the normal involution fashion. But more importantly, Chinese models are not cheap at all. I've just completed a natural experiment, ran out of OCG $10 subscription, almost exclusively burning V4.1F – one of the most cost-efficient ways available to use the most compute-efficient Chinese model on the market. The estimated API rate would have been $76.5. It is *massively* subsidized. Indeed, it's likely below DeepSeek's own boundary of profitability. There are many ways to do this mafs but consider that they've reported one GPU printing $38/hr. In other words I've used up about 2 GPU-hours' worth. [Another way: in that same paragraph DeepSeek reports 155T tokens/day off a single server, which as previously established has at least ≈2400 GPUs. 64B/GPU/day, 2.66B/hour. This way we get even worse results for DS, read on]. How much for 2 hours of GB200? is it more or less than $10? At rental prices it's a straight loss. But they own hardware. How does it work in the most optimistic case? A GB200 that makes you $5/hour is a GB200 that makes you $43.8K under 100% utilization over a year. Retail unit price like $65K, 1833 W/GPU in a rack, $0.080/kWh state-subsidized electricity, so $1300/year on GPU power alone, add a bunch of associated expenses, and we arrive at about 2 years to breaking even on compute, a far cry from Liang's target of 10 months. Except GB200s in China are smuggled with an unhinged markup and easily go above $200K. If this were the main channel of revenue, it will JUST NEVER BREAK EVEN, by the time you approach net profit your hardware is obsolete. But as I've said, it's not a story of involution. What volume has that virtual, subsidized $76.5 bought me? A paltry ≈7B "tokens", 98% cache hits, 107M in 35.3M out. Given that input is 1/4 the output for DS, we can normalize to 62M total "out". 6.2M tokens per $1 of actual payment. Back in Jan 26, @she_llac had reconstructed Anthropic's subscription pricing: on Pro plans, you were getting max 0,54M *OPUS* tokens per $1, with cache hits entirely free, ie infinitely cheaper than Whale API's $0.003/0.006 (which is why I'm ignoring them as well; OpenAI has made cache virtually free now, too, though we can quibble about API vs chat, 72h vs 12h etc). So the difference per token is about 11.5x. That's a lot, but Claude Opus tokens have way more utility. Since then, Anthropic subscription kept getting *more* efficient. What are we getting these days? $5K worth of API on $100 sub? $7000? Haven't been checking. …I need to go so will leave the remaining mafs to those interested. My point is: American labs have bodied Chinese ones. More capability, better hardware, more money to burn result in a situation where the most efficient Chinese model *on the most subsidized plan* is very close in price per token to the best American model currently served; Kimi or Zhipu are nowhere in sight. If you think, like I do, that DeepSeek is actually very efficient and Opus costs substantially more per token than a 8/16B active Flash tier, then it can be argued that *Americans* are doing involution, and not the Chinese.

Our second product is called Blindsight and it will, as the name suggests, enable even those who are completely blind or have never seen to see. It is ready from an engineering standpoint. Just waiting on regulatory approval.