frontpage.
newsnewestaskshowjobs

Open Source @Github

fp.

Loss of Cell Identity Drives Human Aging

https://erictopol.substack.com/p/loss-of-cell-identity-drives-human
1•bookofjoe•2m ago•0 comments

The Game Theory of AI Pacing

https://www.paradigm.xyz/writing/the-game-theory-of-ai-pacing
1•ronfriedhaber•2m ago•0 comments

Roman Dodecahedron

https://en.wikipedia.org/wiki/Roman_dodecahedron
1•softwaredoug•5m ago•0 comments

Random TV Time

https://play.google.com/store/apps/details?id=com.randomtvtime.app&hl=en_US
1•eoleol•8m ago•0 comments

Show HN: MoA as Universal IL –> 30yrs from PhD to 5-device validated compiler

https://github.com/womenflyplanes/moa-compiler
1•lmullin•8m ago•0 comments

Free hosted API for Laya, the open-weight decision model

https://vercel.com/ai-gateway/models/laya
1•boundlesshq•8m ago•1 comments

Kcc – a C17 compiler built solo with an LLM on $100/month, no agents

https://github.com/LiterateDrivenDevelopment/kcc
2•temo_pacheco•11m ago•1 comments

Show HN: Paper(Local CLI for Agent Debugging)

1•varman11•14m ago•0 comments

Getting started with Claude Code mods

https://claude.dev/blog/getting-started-with-claude-code-mods/
2•mfiguiere•16m ago•0 comments

SvelteKit 3 Released

https://github.com/sveltejs/kit/releases/tag/%40sveltejs%2Fkit%403.0.0
2•thunderbong•22m ago•0 comments

AI fabrication suspected of writing articles in Canadian publications

https://www.theglobeandmail.com/world/article-ai-suspected-of-writing-dozens-of-articles-in-canad...
3•rdmuser•24m ago•0 comments

We Used AI to Find Out Just How Much Americans Hate AI

https://www.wsj.com/politics/elections/we-used-ai-to-find-out-just-how-much-americans-hate-ai-2ca...
3•dzohrob•25m ago•0 comments

Show HN: Cascade, Mac OS window cascade utility

https://github.com/hasansayani/MacOSApps-Cascade
1•hhsayani•26m ago•0 comments

Open Source PWA of Doom rebuild from zero

https://github.com/spipu/doom-js
1•Spipu45•26m ago•1 comments

War Bros Didn't Always Rule Silicon Valley

https://www.wired.com/story/war-bros-didnt-always-rule-silicon-valley/
2•gmays•29m ago•0 comments

Etymon – Get rid of all your Claude.md, AGENTS.md, cursor/rules, mcp.json, etc.

https://github.com/landoncrabtree/etymon
2•landoncrabtree2•29m ago•2 comments

Layercake converts OpenStreetMap data into tables you can SQL query via a URL

https://layercake.openstreetmap.us/
3•raybb•30m ago•0 comments

I'm Calling for a Pause in the Development of the Universal Bone Dissolver

https://www.theatlantic.com/newsletters/2026/10/ai-slowdown-bone-dissolver/688854/
3•Jtsummers•31m ago•0 comments

Three Depths of Interaction

https://www.adidizdarevic.com/three-depths/
3•aadd•31m ago•0 comments

Pi 1.0

https://earendil.com/posts/pi-1-0/
96•sergiotapia•32m ago•34 comments

Building a Robust Ingestion System for Any File of Any Size

https://www.mixedbread.com/blog/infinite-file-sizes
1•sherlock_h•32m ago•0 comments

Show HN: Crookcooked – Open-source anti-theft alarm for Mac

https://crookcooked.github.io/crookcooked/
1•tbergkvist•33m ago•0 comments

Hanami, Why?: Introductions

https://aaronmallen.me/writing/hanami-why-introductions
1•birdculture•33m ago•0 comments

The $10B AI Assistant Challenging Meta's Muse, Grok's Bots, and OpenAI's Dots [video]

https://www.youtube.com/watch?v=Am7IWP8IpEc
2•simonebrunozzi•34m ago•1 comments

Deterministic architecture checks for coding agents

2•quynhng•34m ago•0 comments

Page Table Memory Consumption

https://frn.sh/pagetables/
1•matt_d•34m ago•0 comments

Hamas and PIJ Military Operatives Posed as Journalists

https://www.idf.il/en/mini-sites/the-press-files/
4•EvgeniyZh•37m ago•4 comments

I believe current AI models can think and understand – Part 1 of 2

https://blog.andymasley.com/p/why-i-believe-current-ai-models-can
1•jaredwiener•40m ago•1 comments

FLUX 3 Image

https://bfl.ai/models/flux-3-image
2•minimaxir•40m ago•0 comments

From Select to Sysadmin: Hijacking Microsoft SQL Copilot (CVE-2026-65669)

https://embracethered.com/blog/posts/2026/from-select-to-sysadmin-sql-copilot-bluehat-asia/
1•kerng•41m ago•0 comments
Open in hackernews

Big Tech's Capex Is Half of Wall Street's Profit Growth

https://inlevel9.com/en/issues/half-the-growth-was-capex
11•speckx•53m ago

Comments

jqpabc123•37m ago
in other words, the entire stock market is now being driven by "Enron" accounting where nothing is really as it appears.
ralph84•30m ago
Enron was massive accounting fraud. Extraordinary claims require extraordinary evidence. Where is your evidence of accounting fraud? "Lots of companies investing lots of money into AI and it might not work out" isn't fraud.
toomuchtodo•26m ago
The AI Economy (Part 2) - What Enron Can Teach Investors About the AI Boom - https://michiganfinancial.com/blog/the-ai-economy-part-2-wha... - August 4th, 2026

> Enron is often remembered simply as one of the largest corporate frauds in American history. While this is true, there are lessons to be learned from the way investors acted during this time. At its peak, Enron was viewed as one of America's most innovative companies. Investors believed it had fundamentally transformed the energy business through sophisticated trading platforms, financial engineering, and technology-driven markets. Its stock price reflected extraordinary confidence in the company's future rather than its demonstrated earning power.

> Part of that confidence was fueled by Enron's use of mark-to-market accounting, which allowed the company to recognize estimated profits from long-term contracts immediately instead of waiting for those profits to actually be earned. Future expectations effectively became present-day earnings. Those projections were wildly "optimistic". When investors realized the expected profits were unlikely to materialize, confidence disappeared almost overnight. Between August 2000 and November 2001, Enron's stock collapsed from roughly $90 per share to just pennies, erasing billions of dollars of wealth and devastating employees and retirement investors alike.

Draw your attention to:

> Despite the advancements created through use of AI and the increasing numbers of applications, investors still face an important question: How much of today's valuation reflects profits that have actually been earned, and how much reflects profits investors hope will eventually exist?

and

> Artificial intelligence represents one of the most significant technological developments of our generation. Its long-term impact may ultimately justify much of today's enthusiasm in the market. History, however, reminds us that markets frequently price revolutionary technologies well before their economic potential is fully realized and fluctuations occur in even the most stable sectors.

> The lesson from Enron is not that today's AI leaders are engaging in similar behavior. It is that investors should remain cautious whenever valuations become increasingly dependent on future expectations rather than demonstrated earning power.

TLDR https://en.wikipedia.org/wiki/Reflexivity_(social_theory)

Anthropic's IPO prospectus illustrates this fairly well.

Anthropic's IPO Prospectus Is a Fucking Doozy - https://news.ycombinator.com/item?id=49914149 - September 2026

Anthropic's IPO prospectus shows AI vision, surging costs - https://news.ycombinator.com/item?id=49886005 - September 2026

> Matt Levine's summary of Sam Altman's (and most of AI industry's) business plan remains accurate: "We will create God and then ask it for money."

WarmWash•16m ago
OK, but where is the AI fraud part?

MTM accounting isn't fraudulent. Investors knew Enron was doing it. The fraud part was when those projected profits fell apart, Enron would hide the losses.

AI having bonkers valuations isn't fraudulent, maybe delusional, but not fraud. Fraud is the act of outright lying to investors about things that are factually true (not hypothetically true).

"We will have trillions in profits" = not fraud, forecasts can be anything

"We have trillions in profits" = fraud, they don't and they are lying.

toomuchtodo•15m ago
You're focusing on the wrong part in the example. Enron's valuation was driven by irrational exuberance and market sentiment. In the case of Enron, the numbers were a fraud. In the case of AI, the numbers show utter carnage but everyone is trading on hope like it's musical chairs. There simply will never be enough revenue to justify to date and forward looking capex. At some point, investors are going to figure out they've been sold a bag of beans.

> "We will have trillions in profits" = not fraud, forecasts can be anything

The forecasts are dishonest and not grounded in reality. I might as well tell you my magic startup is going to consume total global TAM of all human economic activity. Sure it is. If we're going to excuse fraud as forecasts when it's clear the forecasts are simply fever dreams, we agree to disagree.

AI needs $6T in annual revenue to justify data centre boom - https://news.ycombinator.com/item?id=49898952 - September 2026

From the above thread: "To put this into perspective, healthcare, real estate and banking each earn ~$1.5T in revenue annually. This is REVENUE, not profit which is obviously lower than the revenue in every industry. I fail to understand how AI can achieve $6T in annual revenue when the TAM for healthcare and banking is a quarter of that in the US."

You might as well argue you can break the speed of light and control gravity if facts no longer matter. Are those forecasts made in good faith?

ralph84•10m ago
toomuchtodo•30m ago
It'll be interesting to see what happens to US equities, driven by the Mag 7, when the bottom falls out and this capex investment ends. There simply isn't enough capital for this to continue forever, but it can continue for a while longer.
Joel_Mckay•22m ago
There are still good opportunities to invest in firms with legitimate revenue flows, but avoiding the consolidated risk exposure is difficult.

The problem is the unsecured debt loads, fictional market projections, and secondary impact on real firms supplying products, services, and financing.

A good posture, is figuring out which firms may profit from the fire sales. The hard part for Bears and Bulls alike is they are both usually wrong about when a correction event happens (if they are honest players.) =3

https://www.youtube.com/watch?v=rE75WvOtcu8

downrightmike•4m ago
The problem is contagion, those really bad firms will absolutely wreck everyone else
tom_•35m ago
> The English draft accurately matches the Korean source with correct numbers, headings, footnotes, and links. No Hangul remains. No corrections needed.

Good to hear!

CuriouslyC•31m ago
LOLClaude.
yathern•29m ago
My goodness. It's the first sentence of the article even. At least it's only the third post from the top of HN right now. I guess favorably we can assume a human wrote the Korean version and just translated it via ChatGPT.
demibabs•29m ago
LOL
graypegg•23m ago
The idea of the author assuring you that "no corrections are needed" is immensely funny haha. If the author is around this thread, it's all good to translate content dude, and thanks for doing so! But you definitely should translate the output back to your native language and give it a read as a sanity check.

Archive link for context in the future: https://archive.is/g6Znd

simpaticoder•29m ago
Unless something big changes we're heading toward a period of unprecedented capital concentration, even more than we have now which is extreme. There will be a few winners out of the current frontrunners who own all non-physical human-like labor. Owning a few tens of shares in those companies will be generational wealth, so it makes sense more money is flowing into the stock market and into tech stocks in particular.
> valuations become increasingly dependent on future expectations

Valuations are always dependent on future expectations. What else would they be based on? That's one of the main functions of a stock market: providing price discovery via competing buyers and sellers placing bets on the expected future economic prospects of companies. Of course fraud interferes with price discovery, but again, where is the evidence of fraud?