@wadhwa
Pinned post
I have a major result to share.
I’ve found conclusive forensic evidence that the top 4 Chinese AI labs—DeepSeek, Moonshot, MiniMax and Ant Group—have built their newest models on NVIDIA’s most advanced Blackwell chips, which are barred from export to China.
This sounds insane, but I discovered that almost every AI model carries a hidden signature of the GPU that trained it. When a model is first set up, the starting values of its weights are laid out in a pattern that depends on the exact GPU. Each combination of GPU and seed number produces a unique pattern—a sort of fingerprint of the hardware used to create the model. Traces of that pattern survive training into the published model.
Again, this evidence isn’t based on any insider information or leaking. It’s based on a forensic analysis of the model weights released by the companies themselves.
I searched more than six trillion combinations of chip layout and seed number and matched them to open-weights models. In doing so, I found conclusive evidence that six Chinese AI models—including DeepSeek-V4-Pro, Kimi K3, and MiniMax-M3—were initialized on NVIDIA Blackwell systems. Four of the models match Grace Blackwell, the GB200 and GB300 parts in NVIDIA’s flagship liquid-cooled rack-scale systems.
I then verified these results on real hardware. Running unmodified PyTorch with the seeds I’d recovered, a Blackwell B300 reproduced DeepSeek-V4-Flash’s untouched starting weights bit for bit, and a rented GB200 reproduced Kimi K3’s to a correlation of 0.999995. A Hopper H100 reproduced none of DeepSeek-V4-Flash’s, and changing the seed by one broke the match. (DeepSeek’s seed is 953049583 by the way.) The method also was verified to match 24 known open-weights models to the hardware that developers said they used.
It is exceedingly unlikely that these results are a coincidence. Across six trillion combinations, the best accidental match should reach only about 8 sigma (i.e., eight standard deviations above what random chance would produce). The real matches range from 49 to 1,385 sigma, and even the weakest single test hits 20.
These were data-center-scale training runs. With some analysis of release schedules and model details, I estimate that the largest, Kimi K3 and DeepSeek-V4-Pro, used 2,000 to 3,000 Blackwell GPUs each. Clusters of this size cost on the order of $150 to $200 million each to buy, and across these labs alone would add up to more than half a billion dollars of export-controlled GPUs, not counting experimental compute or otherwise. It’s quite likely that we see multi-billion dollar expenditure on export-controlled GPUs.
I am highly confident in this result, and it has been independently verified by senior researchers at OpenAI and Anthropic. You can read the full report at unemploymentlabs.com/checkpoint-for….
There is much more I’ve identified about these models through other forms of checkpoint forensics like this, and will have more to share soon.View quoted post on X
Now that Jensen is out, my pick for his successor would be Nikesh Arora (@nikesharora)View quoted post on X
IMO Jensen was so important for the AI ecosystem because he was:
1) technical
2) aggressive
3) had the credibility required to be aggressive
What Nikesh did at Palo Alto the last 8 years is remarkable, and reflected in its stock (up 10x)
He’s one of only a handful of people who have the credibility required to be aggressive on day 1
Nikesh has the three attributes listed above while being a world class deal maker and having all the necessary relationships to keep NVDA at the center of the AI ecosystem
There’s no easy choice but he’d be my go to
I'm thinking about this conspiracy...View quoted post on X
Current Nvidia CFO became CEO.
But this CFO must know about their chip sales to China all these years. Unless Jensen ran some dark account and allegedly shipped restricted chips to China.
Unlikely for billions of dollars. Now, if such allegations are true, it now seems someone who knew about the Nvidia shipments to China and didn't raise a red flag is now Nvidia CEO?
So can we say China has taken over Nvidia? Oops. That's a big coup.
If so, we should except this new CEO (former CFO) to come out with bombshell reports soon that causes huge losses to Nvidia and wider market crisis as China would want.
China linked short selling companies in the US must be loaded and ready to make a killing then. Interesting conspiracy theories come to my mind.
But that's what it is. A conspiracy theory. Don't believe it.
Madison: Culinary Institute + Le Cordon Bleu, LVMH luxury marketing, LBS MBA, MIT AI course. Joined Nvidia 2020 as MBA intern; now senior director product/technical marketing, physical AI/Omniverse/Isaac after steady promotions.View quoted post on X
Spencer: Columbia marketing degree, co-founded Taipei cocktail bar, NYU Tech MBA (AI focus), MIT course. Joined 2022 as Isaac Sim PM; now director product management, robotics.
US Bay Area director/senior director product marketing or PM in AI/robotics at major tech: typical total comp $300k–$700k+ (base + equity), higher at Nvidia due to stock performance. Their ~$1.2–1.3M total sits at the upper end for the roles and experience but aligns with disclosed independent processes for comparable positions.
