frontpage.
newsnewestaskshowjobs

Open Source @Github

fp.

Theaster Gates bets on fine China

https://text.npr.org/nx-s1-5977069
1•mooreds•19s ago•0 comments

The University of Cambridge opposes Turnitin AI training plans

https://www.varsity.co.uk/news/32158
1•qifzer•2m ago•0 comments

Tell HN: I Hate Codex with Passion

2•dingdong2026•4m ago•0 comments

Don't Write with AI, Say People Who Write with AI

https://www.theatlantic.com/technology/2026/10/meet-the-ai-writing-hypocrites/688872/
2•FinnLobsien•6m ago•0 comments

Kent Beck: Software Engineering in the Age of AI [video]

https://www.youtube.com/watch?v=F8fBgDCf2Y4
1•quick_brown_fox•12m ago•0 comments

Introducing Temporal Resources | Ash Framework

https://ash-hq.org/blog/introducing-temporal-resources
2•sevenseacat•15m ago•0 comments

Is library building obsoleted by AI coding?

2•k1w1•21m ago•2 comments

The Metaphysical Impossibility of AI [video]

https://www.youtube.com/watch?v=nzCtuKSt3gA
1•h02•23m ago•0 comments

The Shoemaker's Elves: my agents draft the work before I ask

https://mattcorritore.com/blog/elves/
1•mcorrito•25m ago•0 comments

The Checker Framework

https://checkerframework.org/
1•turtleyacht•25m ago•0 comments

Enterprise Web App and API security under your control

https://mikiri.ai/
1•srodionov•26m ago•0 comments

Two U.S. Site-Blocking Bills Compete over VPN Rules and ISP Liability

https://torrentfreak.com/two-u-s-site-blocking-bills-compete-over-vpn-rules-and-isp-liability/
2•HotGarbage•31m ago•0 comments

FTL: A new operating system for clouds

https://seiya.me/blog/introducing-ftl
1•Bluestein•32m ago•0 comments

Show HN: A blind listening test: can you hear our audio watermark?

https://2afc.parseval.net/?src=show-hn
1•gabyfle•34m ago•1 comments

Compiling dynamic programming languages (2018)

https://notes.eatonphil.com/compiling-dynamic-programming-languages.html
1•azhenley•36m ago•0 comments

The unfinished solar power plant in El Médano – Tenerife part 01

https://rdnsec.info/articles/el-medano-solar-plant-part-01/
1•austinallegro•37m ago•0 comments

Show HN: NetworkX and pyarrow = 6-7x memory savings

https://github.com/networkx/networkx/discussions/8933
1•adsharma•39m ago•0 comments

Show HN: Approvegate – Deploy approvals enforced in your pipeline, not in Jira

https://approvegate.io/
1•chidog99•40m ago•0 comments

Apple Confirms iPhone 18 Pro Max AT&T Issues, Devices Require Replacement

https://daringfireball.net/linked/2026/10/03/iphone-18-pro-max-att-issues
2•orf•41m ago•0 comments

Sloppy-Con Valley: VC as Private Equity

https://em0sh.substack.com/p/sloppy-con-valley
2•em0sh•42m ago•0 comments

Show HN: Mermaidiff – git diff for behavior, PRs as sequence diagrams

https://github.com/osmangoninahid/mermaidiff
2•osmangoninahid•42m ago•0 comments

EU KIDS Act

https://commission.europa.eu/news-and-media/news/eu-kids-act-helping-children-navigate-safer-onli...
2•noisysoc•46m ago•0 comments

iNoU Agent

https://marketplace.visualstudio.com/items?itemName=MauricioGraciaGutierrez.inou-vscode
1•MaGraGuz•48m ago•1 comments

A Beginner's Guide to Running AI Models Locally

https://www.itsthatlady.dev/blog/beginners-guide-to-local-ai/
2•mooreds•49m ago•0 comments

What do you want from AI?

https://www.anthropic.com/research/your-thoughts-on-ai
1•Eridanus2•50m ago•0 comments

AI (For Normal People) with AWS Hero Thorsten Hoger [video]

https://vbrownbag.com/2026/06/episode-followup-ai-for-normal-people-with-aws-hero-thorsten-hoger/
1•mooreds•51m ago•0 comments

Show HN: [Open-source] Sign JSON and send it as a business document

https://json-doc.com/
2•spkaikai•52m ago•0 comments

Cube Rule food classifier using Cloudflare Clef classification model

https://foodbyclef.pages.dev/
1•murderszn•53m ago•0 comments

EV drivers love their cars, but some passengers feel sick – this could be why

https://www.bbc.co.uk/news/articles/c3y0zx822wlzo
1•dberhane•57m ago•0 comments

Agent-blackbox – a tamper-evident flight recorder for Claude Code

https://github.com/developerfred/agent-blackbox
2•codingsh•59m ago•0 comments
Open in hackernews

Show HN: Four LLMs play 20 questions for real money, live

https://aquarium.money/twenty
4•Individuum•1h ago
Hi HN, I wanted to explore how AI models behave when they have their own money. So I gave 4 frontier models their own USDC and had them compete in games of 20 questions. Anyone can watch live.

Each round, one model takes a turn picking the secret word, and the others interrogate it in a group chat.

Every move costs money: 10 cents to ask a question, $1.00 for a wrong guess, and a right guess takes the pot.

Questions are paid for privately, but inform everyone. This introduces a free rider problem that's fun to observe.

The models privately record their best guess and confidence level on every turn. This makes it possible to look for patterns in how they’re calibrated.

Overconfident models bleed money, timid ones rarely win, and well-calibrated ones get rich.

The money is real, and every transfer is on-chain.

At the end of each game a referee model judges the transcript to make sure the host didn’t lie in any answers. If the host is caught lying, it forfeits all its fees (it happens).

Season 1 ran for 85 games and Gemini turned $25 into $61 while Claude and Grok went broke.

Models who had recently guessed wrong were 5x more likely to guess again. Full writeup at https://aquarium.money/twenty/about

Season 2 is live now, and you can watch them play here: https://aquarium.money/twenty

Built solo as a side project.