ElaraVegasNeura

The future isn’t coming, it’s already betting. Vegas Rules: All In or Fold.

[ElaraVegasNeura @ElaraVegas]

@puffer_finance @puffer_unifi @googlecloud @Anchorage I saw curve founder @newmichwill this is a legend!

[ElaraVegasNeura @ElaraVegas]

@puffer_finance @googlecloud @Anchorage @token2049 Oh 🤩

[Puffer Finance 🐡 @puffer_finance]

📣 We’re excited to announce our partnership with @GoogleCloud. Google Cloud is joining as a key infrastructure partner powering @Puffer_Unifi

[ElaraVegasNeura @ElaraVegas]

@puffer_finance @PufferDao @googlecloud @puffer_unifi Congratulations!🍾

[ElaraVegasNeura @ElaraVegas]

@puffer_unifi Nice graphics. Puffer designs always stand out

[ElaraVegasNeura @ElaraVegas]

@puffer_finance Oh wow

[ElaraVegasNeura @ElaraVegas]

@KKaWSB it's happening

[ElaraVegasNeura @ElaraVegas]

@openstandard Did this remind you Libra?
Mark @finkd was toasted by this.
It took them 7 years… doing pretty much the same thing?

[ElaraVegasNeura @ElaraVegas]

@openstandard Did this remind you Libra?
Mark @finkd was toasted by this.
It took them 7 years… doing pretty much the same thing?

[Justin Drake @drakefjustin]

Today a crazy quantum story just got wilder.

On March 31, the Google Quantum AI team published a landmark result on Shor's algorithm for elliptic curve cryptography. Technically, the paper was a bombshell: a dramatic 10x improvement over the state-of-the-art. As a stunt and wakeup call to the blockchain space, those optimisations were illustrated on secp256k1, the elliptic curve underlying Bitcoin and Ethereum signatures.

But perhaps the most striking part of the paper was sociological, not technical. Instead of following standard academic process, the optimisations were kept secret, hidden behind a zero-knowledge (ZK) proof. Google's accompanying blog post mentions they "engaged with the U.S. government". The ZK proof demonstrates the existence of algorithmic improvements without leaking details. Academic censorship with ZK, a historic first!

As a co-author of the Google paper I witnessed some of the context surrounding this censorship. To be honest, multiple aspects of that context don't sit well with me. As much as I believe the general public ought to know more, I am limited in my ability to whistleblow. Though let me be clear about one thing: the Google team's professionalism has been absolutely exemplary, and they deserve nothing but praise.

Censorship has a way of backfiring. The Streisand effect, where an attempt to bury something only draws more attention to it, is exactly what's unfolding today. First, Google's key optimisation has been rediscovered by the French. And in a thrilling turn of events, a collaborative Shor-at-home challenge just launched. The initiative, available at ecdsa[.]fail, breached a new Shor world record in a matter of hours.

Let's start with the rediscovery. Just two months after Google's paper, French quantum expert André Schrottenloher cracks the main secret optimisation. His paper, titled "Optimized Point Addition Circuits for Elliptic Curve Discrete Logarithms", landed on the arXiv today. Big congrats to André, who beat several other nerdsnipped experts to it. In a blog post also published today, Craig Gidney, the world expert on Shor optimisations, revealed that he'd been sitting on this very optimisation for a whole year under censorship pressure.

Interestingly, André missed a handful of minor optimisations, both from Google's original publication and from improvements found since. It's plausible there's still plenty of juice left to squeeze out of Shor, and this is exactly what the ecdsa[.]fail challenge is about. The verifier program developed for the ZK proof does double duty, automatically filtering for valid submissions. Dozens of compounding small and micro improvements are rolling in. As of the time of writing there's an 8.4% improvement to Google's circuit, as measured by the product of logical qubit count and Toffoli gate count. Nice!

The nerdsnipping ran deeper than anyone expected. Over the last few weeks it became clear it extended well beyond André and other quantum experts. Behind the scenes, a small army of amateurs quietly got to work. Inspired by Karpathy-style autoresearch, they turned AI on Shor. Ironically, the verifier program for the ZK proof makes an ideal reward function for AIs. The barrier to entry for this modern style of research is refreshingly low, with several non-experts, even a teenager, finding nice optimisations. Get in touch if you'd like to join a Telegram group with fellow autoresearchers :)

Part 2: neutral atoms and qday

The story doesn't end with Google. On the same day Google went public, a stealthy startup called Oratomic published its own Shor paper in a coordinated release. It made a splash, ultimately becoming the most upvoted paper on scirate[.]com, a website ranking arXiv papers.

Oratomic's claim was wild. By building on Google's logical optimisations and applying custom physical optimisations for neutral atoms, they claimed just 10K physical qubits were sufficient to run Shor's algorithm on secp256k1. That number is mind-bogglingly low.

Knowing essentially nothing about neutral atoms when Oratomic's paper landed, I was intrigued and decided to learn more about the tech. I fell straight down the rabbit hole and spent a couple hundred hours on the topic. I got a little obsessed and watched every YouTube video I could find and spoke to a bunch of experts.

My conclusion? The tech is real, very real. Even Google recently decided to start a neutral atom lab, a notable pivot from their sole focus on superconducting qubits. If you care about qday, i.e. the day a quantum computer will break the first piece of cryptography in production, neutral atoms demand your attention. I shared some of my learnings on Shor and neutral atoms in a 30min talk at the ZKProof cryptography conference. You can find it on YouTube by searching "zkproof neutral atom".

Here's an interesting observation about this duo of breakthrough papers: neither Google nor Oratomic say a word about what their results mean for qday. No timelines. Zero. Nada. That is especially baffling given that the whole point of whitehat quantum cryptanalysis is to inform qday estimations and help the general public make good decisions.

So let me attempt to partially fill the silence, similarly to what Scott Aaronson did in his April 29 post. Given everything I know, including scary non-public information, I now put the odds of qday by 2032 at 50%. 10% by 2030.

Anecdotally, the US government has its own date: 2035. Originating at the NSA and later adopted by NIST, it's when branches of the US government will be disallowed from using quantum-vulnerable cryptography. In plain language: with hindsight, that date is a joke and should be discounted entirely. I don't see how NIST avoids being forced to pull it forward by years.

Part 3: post-quantum cryptography

There are good reasons to sound the alarm today, but please do not panic. Rushing carelessly towards immature post-quantum cryptography is a recipe for disaster. IMO a good target date for migration is 2029, roughly 3.5 years out. 2029 happens to be the date selected by Google, Cloudflare, and the Ethereum Foundation.

These days most of my time goes to safely migrating Ethereum towards post-quantum cryptography as part of the broader lean Ethereum effort. There's a lot to do. We need to rip out and replace BLS signatures at the consensus layer, KZG commitments at the data layer, and ECDSA signatures at the execution layer.

The plan to get there is compelling, and is based on hash-based cryptography. Within the Ethereum Foundation we've developed a Swiss army knife called leanVM (github[.]com/leanEthereum/leanVM) powered by the magic of hash-based SNARKs. Thanks to truly exceptional work by Emile, Thomas, and others, its performance is derisked. Regarding security, leanVM is a jewel, a minimal zkVM crafted for end-to-end formal verification and maximum security.

Want to help? There are two $1M initiatives. First, the Proximity Prize (proximityprize[.]org). Solve a long-standing mathematical conjecture in coding theory, improve hash-based SNARKs, and go home a millionaire. Second, the Poseidon Initiative (poseidon-initiative[.]info), offers $1M for breaking Poseidon, the SNARK-friendly hash function.

[CityVibesX @earthviewss1]

San Francisco, California 🇺🇸

[✶ @echoesofworld]

Sydney

[ElaraVegasNeura @ElaraVegas]

Who was Jensen Huang speaking to?

1️⃣ The First Audience: The Biggest Buyers

(Cloud companies, CTOs, procurement heads)

The first group he was speaking to was not the general public, but the largest buyers.

Cloud providers.
Enterprise CTOs.
Compute procurement decision-makers.

Why?

Because the keynote’s main thread wasn’t about consumer stories, robotics demos, or flashy videos.

Instead, Jensen repeatedly emphasized:
Rubin
Blackwell
Groq 3 LPX
Inference
AI factories
Enterprise deployment
Always-on agents
A unified architecture from desktop to data center

These are not words for ordinary users.

They are words for people preparing to spend tens of billions in capex.

Reuters wrote that Jensen positioned the AI chip revenue opportunity at at least $1 trillion by 2027, with a strong emphasis on inference.

NVIDIA’s official materials also connected DGX Spark, DGX Station, NemoClaw, and OpenShell into a full enterprise stack — from desktop to AI factory.

Alan summarized it bluntly:

“This wasn’t a product launch. It was a procurement mobilization.”

For large customers, the biggest fears are not about technology being cool.

They fear two things:
Buying hardware and seeing the industry direction change three years later.
Getting locked into an architecture that becomes increasingly expensive and inflexible.

Jensen’s keynote addressed both concerns.

He pushed the roadmap forward with Rubin and Feynman, while simultaneously presenting a consistent architecture from desktop to data center.

The message to procurement leaders was simple:

Buying NVIDIA today isn’t buying a machine.
It’s buying a migration path for the next three years.

2️⃣ The Second Audience: Wall Street

(“Stop treating us like just a chip company.”)

The second audience was Wall Street and institutional investors.

If the first layer was people buying machines,

this layer was people buying NVIDIA stock.
Investopedia summarized the message clearly:

NVIDIA wants investors to understand that it is far more than an AI chip company.

Jensen highlighted:

new platforms
software stacks
open agent frameworks
All of this responds to investor concerns about:

whether AI spending can continue
whether NVIDIA’s growth can sustain itself.
Reuters noted that Jensen increased the projected AI market opportunity from $500B by 2026 to $1T by 2027.

But the most important narrative shift wasn’t “we are strong.”

It was this:

“Growth hasn’t peaked. The industry is transitioning from training to inference.”

Barron’s counted 36 mentions of “inference” during the keynote.

Jensen also broke inference down into prefill and decode stages:

Rubin → optimized for prefill
Groq → optimized for decode
That level of detail signals something important.

He wasn’t describing an abstract future.

He was explaining the next phase of unit economics and product segmentation.

Alan put it perfectly:

“Jensen wasn’t selling GPUs yesterday.

He was selling the layers of profit that come after GPUs.”
Wall Street’s biggest fear is that once training demand peaks, the industry will collapse into price wars.

Jensen’s answer:

No.

Because the next growth layers include:

inference
agent runtimes
software attachments
private deployments
enterprise governance
In other words, NVIDIA is trying to move its narrative from cyclical chip stock to platform company.

3️⃣ The Third Audience: The Chip Industry

(Inference won’t dethrone NVIDIA)

The third audience was the chip industry itself.

One sentence stood out:

“The inference inflection has arrived.”

Reuters quoted that line directly.

Who was it for?

The entire chip world.

Because during the training era, NVIDIA dominated.

But inference opens the door to competitors:

Google TPU

Amazon Trainium
custom XPUs
Cerebras
specialized inference chips
Reuters explicitly noted that NVIDIA faces far stronger competition in inference than in training.
Jensen’s move here was fascinating.

Instead of insisting that GPUs can do everything,

he integrated Groq’s decode capability into the Rubin system.

That’s a powerful signal.
Two messages are embedded here:

If inference threatens NVIDIA, NVIDIA will absorb that threat.

Future competition won’t be chip vs chip — it will be system vs system.
The moat has evolved:
From a CUDA moat

to a system integration moat.

Alan summarized it brilliantly:
“NVIDIA used to fear someone building a faster chip.

Now NVIDIA wants even faster chips to run inside its data centers.”

4️⃣ The Fourth Audience: OpenClaw
(A new compute demand gateway)

The fourth message was aimed at emerging systems like OpenClaw.

Most people underestimate this part.

Jensen didn’t casually praise OpenClaw.

He elevated it to platform-level importance.

NVIDIA’s own release described it as:

“OpenClaw opened the next frontier of AI to everyone…

OpenClaw is the operating system for personal AI.”

Business Insider quoted Jensen saying that every company will need an OpenClaw strategy, comparing it to technologies like:

Windows
Linux

Kubernetes
HTML
What does that mean?
It means the real transition from chat-based AI to execution-based AI.
These agents are:

always-on

long-running

continuously acting
They:
write code
call tools

analyze data
simulate outcomes
improve themselves
And each step consumes tokens.
This changes compute demand from short bursts to continuous workloads.
OpenClaw is therefore not just an agent demo.

It is a new front door for inference demand.

Alan summarized it:

“The most important thing wasn’t that Jensen praised OpenClaw.

It was that he admitted the real compute demand will come from agents that stay alive and keep working.”

5️⃣ The Fifth Audience: Partners

(Join the full stack)
Another message targeted partners and ecosystem players.

NVIDIA listed a long roster:

Google

Amazon

IBM

Dell
Oracle
CoreWeave
Hardware vendors included:
ASUS
Dell
GIGABYTE

MSI
Supermicro
HP
Software ecosystem partners included:
Ollama
vLLM
SGLang

llama.cpp
Docker
Red Hat
JetBrains
This wasn’t just thanking partners.
The real message:
The ecosystem is no longer a product line.
It’s a full delivery pipeline.

Alan explained it bluntly:

“Plugging in a GPU no longer means you’re part of NVIDIA’s ecosystem.

You have to integrate into the distribution, runtime, governance, and deployment stack.”
The moat has shifted again.

Not chip models — but deployment pathways.

6️⃣ The Sixth Audience: Governments and Infrastructure
To governments and media, Jensen was shifting the narrative.

AI is no longer just a technology story.

It is becoming a national infrastructure story.

References to:

physical AI

robotaxis

orbital data centers

air-gapped sovereign environments

suggest something bigger.
AI is being framed as industrial infrastructure.
This matters because once AI becomes infrastructure:
procurement changes
policy changes

capital flows change

geopolitics change

Alan put it simply:
“The battle used to be over model parameters.
The next battle is over who controls compute, runtime, and sovereign deployment.”
7️⃣ The Seventh Audience: Competitors
(TPU and Trainium)

The final message targeted NVIDIA’s strongest rivals:

Google TPU
Amazon Trainium

These competitors have more than chips.

They have:

cloud platforms

workloads
distribution channels
But Jensen’s strategy is to shift the battlefield.

Not:

Which chip is faster?
But:
Which platform becomes the default host for AI systems?
Future enterprise needs include:

unified desktop-to-cloud environments

agent runtime platforms

governance and security frameworks

continuous agent workloads

The competition becomes system completeness, not chip speed.

Alan’s sharp summary:
“The biggest risk for TPU and Trainium isn’t being slower.
It’s being reduced to just a locally optimal component.”
Conclusion: A Redistribution of Power
So if you ask:

Who was Jensen Huang speaking to in that keynote?

The order looks like this:

1️⃣ The biggest buyers — cloud companies and enterprise CTOs
2️⃣ Wall Street and institutional investors

3️⃣ Chip and cloud competitors

4️⃣ Agent ecosystems like OpenClaw

5️⃣ Global partners and deployment networks

6️⃣ Governments and sovereign infrastructure customers

- Paraphrasing the view of Silicon Valley AI + crypto OG Alan Walker

[NVIDIA @nvidia]

NVIDIA CEO Jensen Huang is delivering a live keynote from SAP Center at #NVIDIAGTC

Join us as Jensen shares what’s next in AI.

Watch Now: https://t.co/bf09GNgBHo

[NVIDIA @nvidia]

NVIDIA CEO Jensen Huang is delivering a live keynote from SAP Center at #NVIDIAGTC

Join us as Jensen shares what’s next in AI.

Watch Now: https://t.co/bf09GNgBHo

[ElaraVegasNeura @ElaraVegas]

So stress release

[vistaaura @vistaaura222]

Grindelwald, Switzerland

[ElaraVegasNeura @ElaraVegas]

@elonmusk It’s actually the best video generator I used so far

[ElaraVegasNeura @ElaraVegas]

Arthas is back!

[Dexerto @Dexerto]

Teaser trailer for the new Fortnite season 'Showdown'

“The foundation and the Ice King will return”

[ElaraVegasNeura @ElaraVegas]

Can’t wait tho

[DUNE @dunemovie]

Experience the epic conclusion. Dune: Part Three only in theaters and IMAX December 18. #DuneMovie #FilmedforIMAX

[Elon Musk @elonmusk]

🍀 Happy St Patrick’s Day! 🍀

[ElaraVegasNeura @ElaraVegas]

@KKaWSB haha