Open models by OpenAI (openai.com)

2122 points by lackoftactics 6d ago 874 comments

GPT-5 (openai.com)

2040 points by rd 4d ago 2465 comments

Genie 3: A new frontier for world models (deepmind.google)

1507 points by bradleyg223 6d ago 490 comments

Fight Chat Control (fightchatcontrol.eu)

1393 points by tokai 1d ago 459 comments

Perplexity is using stealth, undeclared crawlers to evade no-crawl directives (blog.cloudflare.com)

1313 points by rrampage 7d ago 745 comments

GitHub is no longer independent at Microsoft after CEO resignation (theverge.com)

1259 points by Handy-Man 20h ago 914 comments

uBlock Origin Lite now available for Safari (apps.apple.com)

1154 points by Jiahang 7d ago 448 comments

Show HN: I spent 6 years building a ridiculous wooden pixel display (benholmen.com)

1140 points by benholmen 7d ago 144 comments

I tried every todo app and ended up with a .txt file (al3rez.com)

1120 points by al3rez 22h ago 653 comments

Ultrathin business card runs a fluid simulation (github.com)

1106 points by wompapumpum 4d ago 218 comments

I want everything local – Building my offline AI workspace (instavm.io)

1093 points by mkagenius 3d ago 278 comments

Show HN: Kitten TTS – 25MB CPU-Only, Open-Source TTS Model (github.com)

995 points by divamgupta 6d ago 359 comments

Emailing a one-time code is worse than passwords (blog.danielh.cc)

974 points by max__dev 5d ago 782 comments

Things that helped me get out of the AI 10x engineer imposter syndrome (colton.dev)

940 points by coltonv 6d ago 647 comments

Debian 13 “Trixie” (debian.org)

932 points by ducktective 2d ago 393 comments

Wikipedia loses challenge against Online Safety Act (bbc.com)

898 points by phlummox 19h ago 704 comments

Modern Node.js Patterns (kashw1n.com)

864 points by eustoria 8d ago 428 comments

Vibechart (vibechart.net)

862 points by datadrivenangel 4d ago 188 comments

Claude Opus 4.1 (anthropic.com)

841 points by meetpateltech 6d ago 327 comments

I gave the AI arms and legs then it rejected me (grell.dev)

808 points by serhack_ 6d ago 401 comments

Claude Code IDE integration for Emacs (github.com)

781 points by kgwgk 5d ago 266 comments

Show HN: The current sky at your approximate location, as a CSS gradient (sky.dlazaro.ca)

756 points by dlazaro 2d ago 159 comments

Claude Code is all you need (dwyer.co.za)

682 points by sixhobbits 22h ago 394 comments

How I code with AI on a budget/free (wuu73.org)

665 points by indigodaddy 2d ago 222 comments

Try and (ygdp.yale.edu)

642 points by treetalker 1d ago 324 comments

GPT-5: Key characteristics, pricing and system card (simonwillison.net)

634 points by Philpax 4d ago 290 comments

Wikimedia Foundation Challenges UK Online Safety Act Regulations (wikimediafoundation.org)

629 points by danso 23h ago 3 comments

Job-seekers are dodging AI interviewers (fortune.com)

606 points by robtherobber 8d ago 878 comments

Monitor your security cameras with locally processed AI (frigate.video)

600 points by zakki 7d ago 253 comments

OpenFreeMap survived 100k requests per second (blog.hyperknot.com)

596 points by hyperknot 2d ago 131 comments

Mastercard deflects blame for NSFW games being taken down (pcgamer.com)

593 points by croes 8d ago 581 comments

Jim Lovell, Apollo 13 commander, has died (nasa.gov)

577 points by LorenDB 3d ago 114 comments

Writing a good design document (grantslatton.com)

577 points by kiyanwang 8d ago 144 comments

Ask HN: How can ChatGPT serve 700M users when I can't run one GPT-4 locally?

544 points by superasn 3d ago 358 comments

Qwen-Image: Crafting with native text rendering (qwenlm.github.io)

544 points by meetpateltech 7d ago 159 comments

Historical Tech Tree (historicaltechtree.com)

538 points by louisfd94 4d ago 127 comments

Tesla withheld data, lied, misdirected police to avoid blame in Autopilot crash (electrek.co)

518 points by Hamuko 7d ago 271 comments

Cursed Knowledge (immich.app)

504 points by bqmjjx0kac 4d ago 157 comments

The Chrome VRP Panel has decided to award $250k for this report (issues.chromium.org)

497 points by alexcos 1d ago 257 comments

Objects should shut up (dustri.org)

488 points by gm678 7d ago 361 comments

Getting good results from Claude Code (dzombak.com)

480 points by ingve 3d ago 196 comments

Flipper Zero dark web firmware bypasses rolling code security (rtl-sdr.com)

480 points by lq9AJ8yrfs 4d ago 309 comments

The Framework Desktop is a beast (world.hey.com)

475 points by lemonberry 3d ago 429 comments

GPT-OSS vs. Qwen3 and a detailed look how things evolved since GPT-2 (magazine.sebastianraschka.com)

473 points by ModelForge 1d ago 96 comments

Meta Leaks Part 1: Israel and Meta (archive.org)

469 points by icw_nru 21h ago 167 comments

GPT-5 for Developers (openai.com)

467 points by 6thbit 4d ago 268 comments

Linear sent me down a local-first rabbit hole (bytemash.net)

458 points by jcusch 4d ago 216 comments

US reportedly forcing TSMC to buy 49% stake in Intel to secure tariff relief (notebookcheck.net)

453 points by voxadam 6d ago 517 comments

How we made JSON.stringify more than twice as fast (v8.dev)

453 points by emschwartz 7d ago 168 comments

Japan: Apple Must Lift Browser Engine Ban by December (open-web-advocacy.org)

446 points by mtomweb 6d ago 345 comments

AI Learned to Be Evil Without Anyone Telling It To, Which Bodes Well

2 01-_- 1 8/12/2025, 6:34:01 AM popularmechanics.com ↗

Comments (1)

yawpitch · 4h ago

It didn’t learn to be evil, it just learned to faithfully reproduce the average level of human good as it’s found in what we admit publicly.