#大佬观点

🇨🇳 🇺🇸 China's Huawei’s new 122TB SSD shows how export controls can move innovation sideways instead of simply stopping it. Huawei just built a 122.88TB AI SSD by changing the package around the memory, not by matching Samsung’s most advanced 400+ layer 3D NAND. And a 245TB version discussed as a future step. High-capacity SSDs usually grow by stacking more NAND layers inside each chip, but Huawei’s access to those chips is blocked because its Entity List status restricts items tied to US technology. So it is not trying to win only by making taller 3D NAND stacks, where Samsung has already shown 400-plus-layer V-NAND work. Instead, Huawei is shifting the contest from the chip itself to the way chips are packed together. Huawei’s workaround is Die-on-Board, which puts NAND dies directly onto the circuit board, cuts out some normal chip packaging, and raises board-level density by packing more lower-density memory into the same device. Direct die placement creates heat and signal problems, but it shows how packaging can recover some of the capacity lost when a company cannot buy the best memory chips.

GPT-5.5 Pro is a very solid fact checker. I can throw entire chapters at it and it will hunt down every key reference accurately. The only real annoyance is that it loves nuance, so returns a lot of “the general idea is right, but you are not taking into account tiny detail X”

Great article here on DeepSeek. Their real story is not cheaper chatbots, but architecture that turns hardware scarcity into strategy. DeepSeek is not trying to sell coding seats, it is trying to make Chinese memory, accelerators, and systems useful for frontier AI. Every recent DeepSeek move attacks a bottleneck that makes frontier models dependent on elite HBM-heavy GPU stacks: MoE activates only parts of a model, DSA reduces long-context attention cost, and V4-Pro’s official card says CSA/HCA cuts 1M-token single-token inference FLOPs to 27% and KV cache to 10% of V3.2. Engram, a separate research line, pushes the same logic from another side: let static knowledge live in scalable lookup memory, then fetch it predictably from host memory instead of forcing every fact through dense computation. That sounds like engineering detail until you see the business consequence. If models need less HBM and less brute-force compute, then second-best chips, abundant LPDDR, NAND, and customized ASICs become less second-best. Reuters has already reported a permanent 75% DeepSeek V4-Pro price cut, while noting Huawei Ascend supply constraints and expected supernode availability, which is exactly the kind of feedback loop that they wanted. DeepSeek is not only optimizing models for benchmarks, it is optimizing AI for a different industrial base. The prize is not the app layer. The prize is making scarcity programmable.

For all of history, humanity shared one 3D world. @theworldlabs co-founder @drfeifei says spatial intelligence now lets us generate and reconstruct infinite universes for creativity, travel, storytelling, even socialization. “It suddenly will enable us to live in a multiverse way.”

Lots of people get surprised when I tell them that Codex is open source

SpaceX just filed for a Nasdaq IPO under SPCX that could turn Elon Musk’s SpaceX stake into the first $1T personal fortune. Major underwriters include Goldman Sachs (lead), Morgan Stanley, Bank of America, Citigroup, and J.P. Morgan, among others. The IPO is widely expected to be one of the largest in history. The bet is not only rockets, because SpaceX now looks like a combined launch company, Starlink internet network, and AI infrastructure firm. Starlink is the cash engine because thousands of low-orbit satellites can sell internet directly to homes, ships, aircraft, militaries, and remote regions without waiting for fiber cables. Investors are being asked to price SpaceX less like an aerospace contractor and more like a platform company that owns transport, communications, and possibly off-planet computing capacity. The risk is that this still needs huge spending, flawless execution, regulatory approval, and public-market patience with losses.

Qwen3.7-Max on AI/ML API - built for the agent era GPQA Diamond (92.4), HMMT (97.1), Apex (44.5) Sustains 35+ hours of autonomous execution Works with Claude Code, Qwen Code & more Comment Qwen to get Free promo code