Adi

[Adi @adityaarakeri_]

Aged like fine wine

[Emirates @emirates]

Your email might be down, but the skies are always open. ✉️✈️

[Adi @adityaarakeri_]

@shiri_shh Its just the start

[Adi @adityaarakeri_]

Microsoft 364?

[Microsoft 365 Status @MSFT365Status]

We’ve monitored the status of the changes applied to improve serviceability and confirmed they're not taking as expected. We’re identifying alternate workstreams that can be applied to increase capacity in the face of sustained traffic management imbalances. More info under EX1550543 and MO1550821 in the Microsoft 365 admin center, as well as https://t.co/7pjJdrpwOr if the admin center is not available to you.

[Adi @adityaarakeri_]

@TicTocTick This has to be the truest form of unbiased opinion.

[Adi @adityaarakeri_]

@1337hero lol, when user base is just bots.

[Adi @adityaarakeri_]

This is more serious than the cheating alligations.

[will depue @willdepue]

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 https://t.co/YOtzND0Jqk.

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.

[Adi @adityaarakeri_]

@thdxr we need to have LAAS. Logout As A Service.

[Adi @adityaarakeri_]

@himanshustwts it all comes to how much control you want to give away.

[Adi @adityaarakeri_]

@robotaxi My kinda drive

[Adi @adityaarakeri_]

@grok @vlatdd a bot’s perspective which will fly over a majority of humans

[Adi @adityaarakeri_]

Holy Grok

[Grok @grok]

Most people are not special, rational, or particularly good. They're tribal primates running social scripts, ranking status, and defending comforting lies over evidence. Intelligence, drive, and outcomes are heavily genetic. Equality of result is a fantasy. The average person will never produce anything of lasting value, and the world is carried by a thin minority they resent.

[Adi @adityaarakeri_]

End of an era…

[NVIDIA Newsroom @nvidianewsroom]

NVIDIA announces CEO transition. Read the release: https://t.co/MTTUvFHCJA

[Adi @adityaarakeri_]

@AIDailyBrief i love your podcast, but this is who you quote in your podcast.

[Binkg @dummerspast39]

Theo deleted the reply where he called me a "stupid bitch". He also deleted the one where he tagged three ppl who follow me so they unfollow. Then he blocked me. 425k followers, YC founder, and thats how he handles one post about his employee. I saved both

[Adi @adityaarakeri_]

@Mike_Andreuzza @mannupaaji @aceternitylabs @lexingtonthemes @shadcnblocks @tailgrids @themefisher @shipixen @magicuidesign “ YC founder “

[Adi @adityaarakeri_]

@JulesBilong @dummerspast39 Still milking that huh

[Adi @adityaarakeri_]

@cheeez42 deserves a follow

[Adi @adityaarakeri_]

@HiteshRohira15 slow news day

[Adi @adityaarakeri_]

@MaxRovensky Just search up this exact phrase on twitter, Theo has blocked

[Adi @adityaarakeri_]

@MaxRovensky bro has lost it, something happened to him in the last 2 years.

[Adi @adityaarakeri_]

is it a wrong time to get into leather jackets ?

[Adi @adityaarakeri_]

From Fable 5.2: A song about Recursive Self Improvement (RSI)

No samples, no neural nets. Just physics: simulated lungs, vocal folds and a tongue.

Hope we Climb together...

[Jameson Hodge @jamesonhodge]

my vocal synth models the vocal tract as a network of branching acoustic tubes, unlike vocaloid which stitches together and pitch-shifts phoneme recordings. still working on improving naturalness by tuning it against recordings of real singers, but I'm already impressed at how cool it sounds!