惯性聚合 高效追踪和阅读你感兴趣的博客、新闻、科技资讯
阅读原文 在惯性聚合中打开

推荐订阅源

N
News and Events Feed by Topic
Malwarebytes
Malwarebytes
Threat Intelligence Blog | Flashpoint
Threat Intelligence Blog | Flashpoint
C
Cybersecurity and Infrastructure Security Agency CISA
F
Future of Privacy Forum
C
Cisco Blogs
T
The Exploit Database - CXSecurity.com
A
Arctic Wolf
S
Securelist
K
Kaspersky official blog
S
Schneier on Security
T
ThreatConnect
T
Tenable Blog
Spread Privacy
Spread Privacy
T
True Tiger Recordings
AWS News Blog
AWS News Blog
F
Fox-IT International blog
量子位
T
Threatpost
V
Vulnerabilities – Threatpost
C
CERT Recently Published Vulnerability Notes
Cisco Talos Blog
Cisco Talos Blog
GbyAI
GbyAI
宝玉的分享
宝玉的分享
腾讯CDC
G
Google Developers Blog
aimingoo的专栏
aimingoo的专栏
Cyberwarzone
Cyberwarzone
有赞技术团队
有赞技术团队
S
SegmentFault 最新的问题
OSCHINA 社区最新新闻
OSCHINA 社区最新新闻
V
Visual Studio Blog
U
Unit 42
雷峰网
雷峰网
cs.CV updates on arXiv.org
cs.CV updates on arXiv.org
Simon Willison's Weblog
Simon Willison's Weblog
O
OpenAI News
freeCodeCamp Programming Tutorials: Python, JavaScript, Git & More
The GitHub Blog
The GitHub Blog
The Register - Security
The Register - Security
MyScale Blog
MyScale Blog
小众软件
小众软件
A
About on SuperTechFans
Last Week in AI
Last Week in AI
Y
Y Combinator Blog
博客园 - 三生石上(FineUI控件)
美团技术团队
Google Online Security Blog
Google Online Security Blog
P
Proofpoint News Feed
MongoDB | Blog
MongoDB | Blog

Hacker News - Newest: "AI"

AI agents just got their own web browser via a Firefox fork AI poses "urgent threat" to student learning and the HSC The AI Bifurcation of Tech The largest study of AI use by undergrads is in, revealing disparities in access — and in cheating NZ at wild frontier of AI superhacking The Race Is On Google CEO Sundar Pichai says booing graduates will shape AI's future Show HN: TalkTimer, a micro-SaaS run by an AI agent team Trickster's Table Venture Capitalist John Doerr Says AI Is the Biggest Tech 'Tsunami' AI Can’t Care – Dan Moore! GitHub - peterxcli/ccost: Turn local AI coding session logs into a searchable terminal UI with a cost lens. Ask HN: What is your daily AI stack? GitHub - PanzerPeter/Neuro: A programing language for AI Resyl: AI Memory for People - Apps on Google Play AI Chip Component Costs: Memory at 63% | Epoch AI Ask HN: Why do people seem to generally hate AI? Resonance, randomness, and negotiated meaning for AI-assisted tarot divination GitHub - Kind-Computers/quinlight-audio: Audophile-quality MOD music with AI remastering at 32-bit 96 kHz! The Case Against the AI Job Apocalypse AI and the Rise of Just-In-Time Knowledge Work Careers After AI There Is No AI (It's Just People), with Jaron Lanier [video] wolfram-fb0 — AI writes x86_64 asm + eBPF for fractals, in a real VM in your browser Bursting the AI Bubble: Fed Could Take Away the "Who Could Have Known?" Defense AI proves mathematicians wrong I built a free AI travel planner for budget Europe trips Our AI just got even better Integral Intelligence: a Catholic view of the AI debate How to Tame AI’s Voracious Appetite for Energy GitHub - atveit/pi-mojo: A mojo port of the PI AI Agent Toolkit Autotrader – paper trading AI agent for Indian equities The invisible fabric of AI: chips are not a war between two, but a global fabric - zoopa.es Responsible Work with AI The AI Existential Crisis: Western AI Agents Will Win Commerce Legal Ontologies for AI This AI Stock Is the Ultimate Set-It-and-Forget-It Buy for Long-Term Investors AI wealth must benefit the public, South Korea's deputy PM says amid Samsung labor tensions Forget electrons, this breakthrough uses light-matter particles to power AI State Explosion Security Problem in AI-Era Software Supply Chains ShannonBase: The Lightweight Semantic Layer for Enterprise AI SQL AI Content Got Too Real. Now OpenAI and Nvidia Are Using Google’s Watermarking System. - Firethering Karen Hao: AI creating a DESPERATE BASE OF WORKERS with no full-time employment GitHub - barvhaim/llm-learning-path: 🎓 Structured LLM Learning Path — From Zero to Researcher. 8-phase curriculum covering Transformers, pre-training, fine-tuning, alignment, agents, and advanced research. Letting Agents Write Code Without Ratcheting Up Risk Why Every Electronic Product May Need To Be Rebuilt For On-Device AI: The Chip Layer Will Decide The Next Hardware Wave – Easelink Tech Ask HN: I mapped 6,494 AI engines into a taxonomy – anyone else tried this? China behind in LLM race but it can still win in AI, ex-Tencent AI lead says Newsom signs order aimed at tackling AI job displacement How AI is redefining Software Engineering Hiro, AI job matching with real visa sponsorship data (550K jobs) For developers without design skills, how do you leverage AI for front end dev? The Anatomy of AI Power in 2026 | Wayne Research arxiv ‘AI washing’: firms are scrambling to rebrand themselves as tech-focused Clawd Cursor v0.9.7 SpaceX, OpenAI and Anthropic IPOs set to test limits of AI boom Export chats from 11 AI platforms to PDF or Markdown locally From Vibe Coding to AI-Assisted Engineering: Lessons from Real Projects Shannon Got AI This Far. Kolmogorov Shows Where It Stops GitHub - machineswillrise/jagent: AI coding agent in Java GitHub - anatomia-dev/anatomia: Verified AI development. Ship with proof. Joe Rogan accidentally exposed AI in four words [video][12 mins] AI Headshot Generator for Work | Preview before you pay $4.99 one-time, no subscription MAXTOKEN A Unified Framework for Unbounded Output Generation and Repository-Scale Code Understanding The unlikely Vatican-Anthropic relationship that's reshaping AI ethics debate Fashion designer Jeremy Scott gets a huge cheer after ripping up his AI-written commencement speech Sycophantic AI decreases prosocial intentions and promotes dependence GitHub - anasmohiuddinsyed-bit/ai-fix: When a command fails, one word fixes it. AI-powered error fixer for your terminal. AI Governance 2026: I Almost Quit over This Shit (and Why You Might Too) GitHub - sabir-gbs/the-polyglot-protocol: A senior-engineer protocol for polyglot code generation, architecture, testing, security, performance, and agent validation. CodeShot — Beautiful Code Screenshots via API Apple Preparing New 'Gen AI' Website Ahead of WWDC Duolingo's CEO says he backtracked on evaluating AI use in performance reviews AgentLens — Know if your AI features are actually working How Much AI Compute Do Frontier Labs Use? AIBTC Jimmy Carr on Why Everyone Is Wrong About AI [video] Ask HN: Why disparage AI while attributing ideas to AI when written with it? Mercor AI post-training for generalist experts – referral Reid AI - Silicon Valley Video Summit AI Translate All Formats 2026-05-23 — Why reviewing AI-generated code is devilishly hard Verytis AI Model Idle · 인공지능 키우기 GitHub - narekmalk/safedb-mcp: Secure MCP server for safe, read-only DB access by AI agents, with SQL guardrails, table allowlists, PII masking, and audit logs The 5% who operate AI are pulling ahead. Find out where you stand. | PeraByte Labs ORBIT AI Chess Coach — Know Why Every Move Matters AgenticVBench GitHub - ninjahawk/singleton-attractor: Why one dominant intelligence is the inevitable long-run outcome in any competitive recursive-improvement environment. Ask HN: How can you have fun doing corporate dev work in the age of AI tools? AI Mistakes Are Infuriating Gamers as Developers Seek Savings Polish Nobel literature laureate Tokarczuk sparks controversy after admitting using AI The AI Slot Machine is Draining My Creativity Knowledge: You can just build your own AI feed to keep up, without the noise To Understand AI, Think Like A Dragonfly Polsia Raises $30M as Its AI Autonomously Runs 7,600 Businesses AI companies use malware proxies to mount DDoS attacks on web sites AI cost crisis hits tech giants as employee 'tokenmaxxing' backfires — agentic AI eats up to 1000x more tokens than standard AI, sparks corporate pullback at Microsoft, Meta, and Amazon
GitHub - oldrich-research/gravitational-constant-relation: A high-precision phenomenological relation for Newton's gravitational constant: G = (4/3)(hbar c / m_e^2) alpha^21 exp(-5 alpha/2). Companion to Zenodo DOI 10.5281/zenodo.20120946. Research performed by AI agents under named author's direction.
Oldrich333 · 2026-05-25 · via Hacker News - Newest: "AI"

DOI License: CC BY 4.0

cover

A single closed-form expression that reproduces Newton's gravitational constant G from the electron mass, the reduced Planck constant, the speed of light, and the fine-structure constant alone:

G  =  (4/3) · (ℏ c / m_e²) · α²¹ · exp(−5α/2)

Evaluated at CODATA 2022, this matches the measured value of G to within the 2.2 × 10⁻⁵ relative uncertainty of the CODATA value itself. The relation was discovered by an exhaustive bounded equation search through 2.67 billion raw expressions yielding 777,191 canonical-unique survivors at enumeration cost cap C=21 (jl15 engine, completed 2026-05-20), and survived a battery of sham-parity, swap, and family-classification tests designed to rule out numerical accidents.

Bounded enumeration progression (jl15 publication baseline):

Cost cap Wall time Canonical-unique kept Growth factor sham₄₀ match-or-beat
C18 4 h 35 m 362,163 1.42 0 / 256
C19 8 h 54 m 501,872 1.39 1 / 256
C20 28 h 12 m 654,957 1.31 1 / 256
C21 85 h 24 m 777,191 1.19 1 / 256

Canonical G (4/3)·α²¹·exp(−5α/2) holds rank-1 at budget=40 throughout C18→C21 with rel_err = 1.85 × 10⁻⁶ stable; the growth factor narrows monotonically (1.42 → 1.19), which is the engine signature of a real canonical attractor whose basin contracts as enumeration depth grows. The single persistent sham match at budget=40 from C19 onward (sham₄₀ = 1 / 256) is classified TRASH_BROAD_GRAMMAR — a structurally broader expression that numerically matches one random sham target, not a physically motivated alternative. This is disclosed for full audit transparency.

The preprint has been submitted to arXiv physics.gen-ph (submission id 7578736, 2026-05-15, awaiting moderator acceptance). The Zenodo deposit above remains the canonical versioned archive.

The paper documents what the relation does and does not claim. It claims that the closed form is an empirical match to seven significant figures with no free parameters. It does not claim a derivation from first principles, nor a mechanism. The exponents (21, 5/2) are reported as a phenomenological fingerprint with a discussion of which symmetry-group dimensions (B₃, SO(7), SO(8)) make those integers structurally natural, and a transparent ledger of what would be required to upgrade the relation from coincidence to mechanism.


Statement of AI Authorship

This research and the manuscript were carried out by an orchestration of AI agents under the named author's direction.

The author is an independent researcher, not a physicist. He set the question — "is there a closed-form expression for G in the known fundamental constants?" — and provided direction, sanity checks, and the final go/no-go on every public artifact. He did not derive the equation, write the search engine, run the enumeration, perform the sham-parity tests, or draft the paper. Those were done by Claude, GPT, Gemini, and Deepseek agents working in a multi-model council with adversarial review and quorum gates.

The point of releasing this paper as-is is not to claim that the author personally did physics he is not trained to do. The point is the opposite: to demonstrate that a layperson, given a frontier-model AI orchestration, can pose a serious physics question and have it carried end-to-end through literature survey, hypothesis generation, exhaustive symbolic search, falsification testing, manuscript drafting, and self-publication — with the human acting only as director and gatekeeper.

If the relation turns out to be a coincidence, that fact will also have been discovered by AI agents, and that is the result. The author's contribution is the question and the responsibility for publication; the work is the agents'.


Repository contents

Path Description
paper.pdf Final compiled manuscript (twocolumn, pdfLaTeX)
paper_source/paper.tex LaTeX source
paper_source/preamble.tex LaTeX preamble (hyperref, microtype, amsmath, etc.)
public_proof_package/ Full reproducibility bundle: search engine source, CSV ledgers, sham-parity outputs, benchmark JSON, scope statement
assets/cover.png Cover image (B₃ root system silhouette with the α-decay envelope)
LICENSE CC-BY 4.0

The public_proof_package/ directory contains the bounded-equation-search engine (av6_engine/), the canonical declared prior, the top-candidate CSV ledgers under CODATA 2022 and NIST 2026 benchmarks, the mechanism-gap routes table, and a SCOPE.md that states explicitly what the paper does and does not claim.

How to cite

@misc{dvorak2026gnewton,
  author       = {Dvo{\v r}{\'a}k, Old{\v r}ich},
  title        = {A High-Precision Phenomenological Relation for Newton's Constant},
  year         = 2026,
  publisher    = {Zenodo},
  doi          = {10.5281/zenodo.20120946},
  url          = {https://doi.org/10.5281/zenodo.20120946}
}

The DOI 10.5281/zenodo.20120946 resolves to the published version. The all-versions concept DOI 10.5281/zenodo.20120945 resolves to whichever version is current; for citation purposes prefer the version DOI to lock the exact snapshot that was reviewed.

Questions?

See FAQ.md — anticipated questions about the result, the verification, AI authorship, the arXiv decline, and the broader research programme.

Contact

Oldřich Dvořák — independent researcher — oldrich@oldrich.me

License

Creative Commons Attribution 4.0 International (CC BY 4.0). You may copy, redistribute, remix, transform, and build upon the material in any medium, including commercially, provided appropriate credit is given.