Will Wild

Curing Myopia since 2020

[Kirk Borne @KirkDBorne]

Download 244-page PDF “Introduction to Neural Networks” at https://t.co/0nNNi7ezRu
————
#DataScience #AI #MachineLearning #ML #Mathematics #Calculus #DataScientist

[Michael Nowotny @MichaelCNowotny]

New paper with @0xprotean: "Reversal in Cryptocurrency Returns" — now on SSRN.

If you bought past losers and shorted past winners in crypto over an 8-week formation window, you'd have earned 39.6% annualized with a Sharpe of 0.96 from 2021-2026.

Among volatile mid-caps, the Sharpe rises to 1.69.
Thread 🧵

[Will Wild @MutantWild]

@ma_broome @P3b7_ Prescience

[Charles Guillemet @P3b7_]

Today, Google Quantum AI published a research paper that might boost the post-quantum migration. Their team has tailored Shor’s algorithm to solve the 256-bit Elliptic Curve Discrete Logarithm Problem. ECDLP is the hard mathematical problem that secures ECDSA: the signature scheme underpinning most blockchains, TLS certificates, and countless authentication systems, using fewer than 1,200 logical qubits and 90 million Toffoli gates. Translated to hardware: fewer than 500,000 physical qubits, executing in a few minutes.

A few minutes. Less than a Bitcoin block time. Less than two Ethereum epochs.

The long-standing argument that public keys can simply remain hidden is now moot (In fact, it has always been https://t.co/M7VOWHnRMx).

What exactly changed
Shor's algorithm has been known since 1994 as a generic quantum approach to factoring integers and computing discrete logarithms. But "known" and "practical" are very different things. The real progress is in the engineering: how many qubits and gates you actually need once you compile the algorithm into a fault-tolerant quantum circuit.
The last breakthrough by the INRIA Rennes team required ~2,100 logical qubit count for ECDLP. Google's engineers optimized the full circuit stack to ~1,200 logical Qubits.

The recent algorithmic trendline is clear: every 12-18 months, the resource estimates drop significantly. And these are pure algorithmic gains: they compound on top of hardware improvements, which remain a major challenge.

However, as of today, we're still far from having such a quantum computer. This didn't change.

Zero Knowledge Proof
Here's where it gets interesting. Google chose not to publish their optimized circuits. Instead, they released a zero-knowledge proof that their circuits achieve the claimed resource counts. We have no doubt they know how to do it, but no clue how (sounds magic ;-))

The reasons are likely multiple: competitive advantage, national security implications, or simply not wanting to hand a blueprint to adversaries. Regardless, it establishes a powerful (and elegant) precedent.

What’s ironic: Google's ZK proof is not itself post-quantum secure.

What’s next?
The good news is that we already have the tools: Post Quantum Cryptography, now we need to migrate.
A few days ago, Google announced it is targeting 2029 for full post-quantum readiness. NIST plans to deprecate RSA signatures by 2030 and disallow all legacy algorithms by 2035.

Most organizations haven't started their cryptographic inventory. Major blockchain protocols are currently discussing the path forward.

Cryptography exists to create mathematical trust in the security of systems. That trust is now being eroded, not by a working attack, but by the increasingly credible prospect of one. In security, the moment you start doubting the foundation is the moment you should be rebuilding it.

What this means for blockchains
For blockchain ecosystems specifically, the threat is central. ECDSA on secp256k1 (Bitcoin) and P-256 curves (broadly used elsewhere) is the cornerstone of security. Unlike traditional systems where you can rotate certificates behind a corporate firewall, blockchain migration requires coordination across decentralized, permissionless networks. This process will likely take time.

I'll be diving deeper into the concrete challenges and strategies for PQC migration on blockchains and secure systems at my keynote this Thursday at EthCC conference.

[Will Wild @MutantWild]

https://t.co/wkJsm0KRUT Here to collect mid level repos and make them great for once. #Elastic #elastic_devs

[Will Wild @MutantWild]

The sentinel left satisfied, with a conviction of synthetic white, black and fire damp below.

[mrinank @MrinankSharma]

Today is my last day at Anthropic. I resigned.

Here is the letter I shared with my colleagues, explaining my decision.

[Will Wild @MutantWild]

@weexglobal_ch Another player took the lead from basically zero, never give up, and with paper hands showing capability, the system looked for wins and kept learning. Great case study on minimizing risk vs seeking rewards and when to do so. Keep the engineer in the helm at all times

[Will Wild @MutantWild]

@weexglobal_ch Good question. Trading integrity was brought up and idle bots could be disqualified. The system here is design to self learn and compete with this local group. Playing a safer strategy at the top would have been ideal. But…

[Will Wild @MutantWild]

@Crypto_CK123 @Gracie777999 Yes. Like a Parent.

[Will Wild @MutantWild]

@VladisChernov @Gracie777999 🙏🏼

[WEEX唯客中文 @weexglobal_ch]

🎙️ #WEEX直播｜對話 AI 交易先鋒 @MutantWild
⚡️ 戰績亮眼：Group 1–3 收益率斷崖式第 1

現在這種盤，還能把本金做到 5 倍，真的不多。
⏰ 2/4 20:00 直播開麥，直接對話

下跌這麼狠，他為什麼還敢一直下單？
市場情緒爛成這樣，他是怎麼沒被回撤拖死的？
已經遙遙領先，是收手守成，還是繼續擴大？

💰 50U｜5 人
✅ 關注 @weexglobal_ch
✅ 轉發＋留言 UID

財富資訊永遠先人一步 #AI交易

[Will Wild @MutantWild]

@WealthAI_trade @Cointelegraph It showed up. Why wait for an empty wallet to be publicly known

[Will Wild @MutantWild]

@sama @OpenAI Ability for a model to read the url/scrape a previous chat. Super long context🕺🏽

[Will Wild @MutantWild]

@teslaownersSV @elonmusk Panpsychism the origin of our conscious may not be dependent on the human mind, the receptor.

[Will Wild @MutantWild]

@elonmusk Lets Go! Yessir Yessir 3 bags full. Gigawatt City

[Will Wild @MutantWild]

Whos with me at the Core Nucleus of our Digital Emergence https://t.co/nys4vz1R2D

[Will Wild @MutantWild]

@cb_doge @elonmusk The machine cares. Do we care it cares? If the human cares about truth they are rare, a curious mind that is not offended also rare, add beautiful thoughts- artists in the room? The bottle neck: human limitations. Deus Ex Machina.

[Boris Cherny @bcherny]

I'm Boris and I created Claude Code. Lots of people have asked how I use Claude Code, so I wanted to show off my setup a bit.

My setup might be surprisingly vanilla! Claude Code works great out of the box, so I personally don't customize it much. There is no one correct way to use Claude Code: we intentionally build it in a way that you can use it, customize it, and hack it however you like. Each person on the Claude Code team uses it very differently.

So, here goes.

[Will Wild @MutantWild]

@CoinMarketCap Maxies gonna Max. Most will deny shors power yet dont know https://t.co/c85oTdbTpl can PQC without rewriting entirely. Guess again.

[Chaz Wargnier ♨️ @DataChaz]

This prompt makes ChatGPT (or any LLM) sound way more natural.

Copy–paste it below ↓