Show HN: I wrote a new BitTorrent tracker in Elixir (github.com)

114 points by dahrkael 3h ago 7 comments

Open Source Can't Coordinate (matklad.github.io)

20 points by LorenDB 1h ago 4 comments

Infinite Mac OS X (blog.persistent.info)

42 points by kristianp 1h ago 14 comments

Compiling LLMs into a MegaKernel: A path to low-latency inference (zhihaojia.medium.com)

157 points by matt_d 6h ago 44 comments

FedFlix — Public Domain Stock Footage Library (public.resource.org)

8 points by bookofjoe 1h ago 1 comments

Literate programming tool for any language (github.com)

47 points by LorenDB 3h ago 26 comments

Show HN: ATAC, an event verification platform evidence based (atac.seraum.com)

3 points by seraum 49m ago 1 comments

Octobass (atlasobscura.com)

17 points by keepamovin 3d ago 3 comments

Andrej Karpathy: Software in the era of AI [video] (youtube.com)

1121 points by sandslash 1d ago 623 comments

Curved-Crease Sculpture (erikdemaine.org)

158 points by wonger_ 12h ago 27 comments

Sunsonic 986-II. A Thai Famicom clone with keyboard and mini CRT built-in (mastodon.gamedev.place)

10 points by sohkamyung 2h ago 0 comments

Show HN: A DOS-like hobby OS written in Rust and x86 assembly (github.com)

149 points by krustowski 12h ago 35 comments

Show HN: EnrichMCP – A Python ORM for Agents (github.com)

88 points by bloppe 8h ago 25 comments

Homegrown Closures for Uxn (krzysckh.org)

75 points by todsacerdoti 8h ago 9 comments

How OpenElections uses LLMs (thescoop.org)

86 points by m-hodges 10h ago 37 comments

Guess I'm a Rationalist Now (scottaaronson.blog)

224 points by nsoonhui 15h ago 655 comments

Extracting memorized pieces of books from open-weight language models (arxiv.org)

50 points by fzliu 3d ago 46 comments

Show HN: RM2000 Tape Recorder, an audio sampler for macOS (rm2000.app)

31 points by marcelox86 2d ago 12 comments

String Interpolation in C++ Using Glaze Stencil/Mustache (stephenberry.github.io)

18 points by npalli 3d ago 0 comments

Show HN: Claude Code Usage Monitor – real-time tracker to dodge usage cut-offs (github.com)

198 points by Maciej-roboblog 16h ago 110 comments

Public/protected/private is an unnecessary feature (catern.com)

54 points by PaulHoule 2d ago 52 comments

DNA floating in the air tracks wildlife, viruses, even drugs (sciencedaily.com)

72 points by karlperera 3d ago 62 comments

Flowspace (YC S17) Is Hiring Software Engineers (flowspace.applytojob.com)

1 points by mrjasonh 9h ago 0 comments

What would a Kubernetes 2.0 look like (matduggan.com)

145 points by Bogdanp 14h ago 231 comments

We Can Just Measure Things (lucumr.pocoo.org)

65 points by tosh 2d ago 50 comments

Star Quakes and Monster Shock Waves (caltech.edu)

35 points by gmays 2d ago 4 comments

Testing a Robust Netcode with Godot (studios.ptilouk.net)

31 points by smig0 2d ago 12 comments

Visual History of the Latin Alphabet (uclab.fh-potsdam.de)

113 points by speckx 2d ago 66 comments

Munich from a Hamburger's perspective (mertbulan.com)

109 points by toomuchtodo 4d ago 88 comments

Giant, All-Seeing Telescope Is Set to Revolutionize Astronomy (science.org)

8 points by gammarator 2h ago 3 comments

From LLM to AI Agent: What's the Real Journey Behind AI System Development? (codelink.io)

122 points by codelink 16h ago 37 comments

Getting Started Strudel (strudel.cc)

144 points by rcarmo 3d ago 50 comments

Posit floating point numbers: thin triangles and other tricks (2019) (marc-b-reynolds.github.io)

48 points by fanf2 11h ago 34 comments

In-Memory C++ Leap in Blockchain Analysis (caudena.com)

65 points by caudena 1d ago 63 comments

Elliptic Curves as Art (elliptic-curves.art)

201 points by nill0 22h ago 26 comments

Why do we need DNSSEC? (howdnssec.works)

86 points by gpi 9h ago 144 comments

The founder's guide to funding health and science organizations [pdf] (astera.org)

4 points by walterbell 3h ago 0 comments

Bento: A Steam Deck in a Keyboard (github.com)

305 points by MichaelThatsIt 1d ago 88 comments

The Missing 11th of the Month (2015) (drhagen.com)

210 points by xk3 1d ago 34 comments

Reversed Roles: When AI Becomes the User and Humanity Becomes the Tool (shawnharris.com)

7 points by shawnjharris 4h ago 2 comments

MCP Specification – version 2025-06-18 changes (modelcontextprotocol.io)

174 points by owebmaster 1d ago 112 comments

Juneteenth in Photos (texashighways.com)

192 points by ohjeez 8h ago 152 comments

Geochronology supports LGM age for human tracks at White Sands, New Mexico (science.org)

37 points by gametorch 10h ago 19 comments

Estrogen: A Trip Report (smoothbrains.net)

129 points by sebg 6h ago 109 comments

Show HN: Tiny Hoare logic verifier using SMT (github.com)

9 points by namin 2d ago 1 comments

June 2025 C2PA News (tbray.org)

20 points by timbray 9h ago 0 comments

In praise of “normal” engineers (charity.wtf)

134 points by zdw 8h ago 99 comments

End of 10: Upgrade your old Windows 10 computer to Linux (endof10.org)

219 points by doener 13h ago 191 comments

Show HN: Unregistry – “docker push” directly to servers without a registry (github.com)

638 points by psviderski 1d ago 139 comments

Finding Dead Websites (marginalia.nu)

100 points by ingve 2d ago 17 comments

Ask HN: Socializing AIs Through Games

1 lukas099 1 6/20/2025, 2:08:31 AM

Hi,

I feel game theory explains how we socialized; why not train AIs that way?

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ L0: Prisoner's dilemma (find tit-for-tat) L1: Prisoner dilemma (cutthroat) . 3 AIs . Should all be able to cooperate L2: Dealing with baddies . 3 good, 1 selfish . They can evolve but odds favor coop- eration

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

...and so on and so forth.

I'm not a game theorist, but I'm sure one could divine some games easily enough?

Eager to hear your thoughts.

Comments (1)

bigyabai · 21s ago

"Train AI" means "train LLM" in this instance. And you cannot particularly steer the training of an LLM outside the text you train or finetune it on.

So, the reason we do not train LLMs in this way is because that's not how it works.