frontpage.
newsnewestaskshowjobs

Open Source @Github

fp.

Open in hackernews

AI writing is better than no writing

https://andrewpwheeler.com/2026/06/23/ai-writing-is-better-than-no-writing/
2•apwheele•1h ago

Comments

Redster•1h ago
Except for the cases where no writing is better than LLM writing. The author's take seems ok to me, but what I struggle with most in AI writing is that it removes or abstracts away the intent of the original author. Not in every sentence and in every thought. But when a human says something, I can reason about why they said it the way they did. There are threads worth tugging on. 2 easy-to-understand examples: 1. when an ESL writer communicates something in awkward English, it's a fascinating glimpse into how they normally communicate about concepts. The meaning is usually not too difficult to infer and I can see how their language or culture affects how they see mine. It's very fun. 2. if an expert explains something and I get lost, that's usually a hint that my mental model could change to see the problem/solution more like they do.

But when a person expert or no uses an llm, information is being lost. If I misunderstand an llm's sentence, it is possible that "it knows what it's talking about", but with how frequently they don't know what they're talking about, I usually don't try and move on. It's not worth it.

In America, at least, though perhaps this applies more broadly, so many trends over the past decades all converge towards the lowering of the baseline levels of trust you should give towards a stranger (whether valid or not!). And llms have been a net-negative so far. We have exchanged some amount of trust for intelligence. And it didn't have to be an exchange.

I care less about llm disclosure and more about this: If you use an LLM for writing, please think sincerely about whether the output accurately represents what you would say and how you would say it to another person. Please avoid serving others nothingburgers and wordsalads. This has always been possible and it's easy to do it without realizing it with LLMs. The output can survive a scan, but dissolves upon further inspection.

tldr; make sure that what you publish is trustworthy.

Show HN: Cc-fleet – run other LLMs as Claude Code workers, your sub drives

https://github.com/ethanhq/cc-fleet
1•ethanhq•50s ago•0 comments

Lost memoir of Hiroshima survivor found after decades in US archive

https://www.theguardian.com/books/2026/jun/23/lost-memoir-of-hiroshima-survivor-found-after-decad...
1•toomuchtodo•1m ago•0 comments

AGI will not come from better models, it will come from meta harness

https://github.com/the-open-engine/zeroshot
1•covibes•1m ago•1 comments

A simple physics-inspired model sheds light on how AI learns

https://techxplore.com/news/2026-05-simple-physics-ai.html
1•doener•2m ago•0 comments

MSG Made Dossier on Activists Who Opposed Facial Recognition

https://www.404media.co/madison-square-garden-made-dossier-on-activists-who-opposed-facial-recogn...
2•cdrnsf•4m ago•0 comments

AI and Math State of the Art

https://stunning-naiad-817d91.netlify.app
2•dawdler-purge•6m ago•1 comments

I Built an NPM package to stop AI agents from burning through API credits

https://www.npmjs.com/package/budget-agent
1•alienhead•8m ago•0 comments

Airship

https://en.wikipedia.org/wiki/Airship
1•simonebrunozzi•9m ago•0 comments

The (Quantum) Revolution That Should Have Been – and Still Could Be

https://conjectureinstitute.org/articles/fellow-spotlight/paul-raymond-robichaud
1•anotherpaulg•9m ago•0 comments

Show HN: Route LLM prompts to cheapest capable model – pydantic-AI and litellm

https://github.com/Reactance0083/pydantic-ai-multi-llm-cost-optimizer
1•reactance0083•10m ago•0 comments

Vector Graphics in Lil

https://beyondloom.com/blog/vectorgraphics.html
1•surprisetalk•10m ago•0 comments

Show HN: Synergy Check – paste two companies, get "should you take the meeting?"

https://intergrate-q86q.onrender.com/scout
1•kevinluddy39•10m ago•0 comments

Show HN: Open-geo – Skill for Claude Code which shows your AI Visibility

https://github.com/Pupok462/open-geo
1•pupok46•11m ago•0 comments

Show HN: FuckUI – Stop feeding agents raw DOM. An accessibility-tree CLI

https://fuckui.com
1•keepamovin•12m ago•0 comments

Show HN: Cerver is infra for AI sessions

https://cerver.ai
1•eyalgoren•15m ago•0 comments

Bluerails – Know if AI agents can find and book your business

https://www.bluerails.com
1•gurveenghai•15m ago•1 comments

Alphabet Shares Drop After Second AI Star Departs for a Rival

https://finance.yahoo.com/technology/ai/articles/alphabet-shares-drop-second-ai-163538654.html
1•ivewonyoung•17m ago•0 comments

Why the Human Genome's Tangled Physicality May Confound AI

https://www.quantamagazine.org/why-the-human-genomes-tangled-physicality-may-confound-ai-20260618/
1•bookofjoe•17m ago•0 comments

A cycling game whose mountains, weather, rider and night sky are all vector code

https://cycling318.com/media
1•dengjiuhong•19m ago•0 comments

Ask HN: Anyone else's company ban use of Chinese models?

3•seahorseemoji•20m ago•0 comments

Workplace Democracy

https://en.wikipedia.org/wiki/Workplace_democracy
1•hamburgererror•20m ago•0 comments

NASA's Nancy Grace Roman Space Telescope Arrives in Florida

https://spaceflightnow.com/2026/06/22/nasas-nancy-grace-roman-space-telescope-arrives-in-florida/
1•smurda•20m ago•0 comments

Meta Glasses

https://about.fb.com/news/2026/06/meta-essilorluxottica-partner-launch-meta-glasses/
2•gotmedium•20m ago•0 comments

Deadcert: A zero-dependency Go CLI to check TLS cert expiry

https://github.com/zuhayr-barhoumi/deadcert
1•ZuhayrBarhoumi•22m ago•0 comments

Show HN: perl-lsp – annotation free static analysis for Perl

https://github.com/tree-sitter-perl/perl-lsp
2•rabbiveesh•22m ago•0 comments

The Flying Pigeon: The Single Most Produced Vehicle in History [video]

https://www.youtube.com/watch?v=p2d2ifuUcCM
2•nxobject•23m ago•0 comments

PixelSmash – Critical FFmpeg Vulnerability Turns Media Files into Weapons

https://jfrog.com/blog/pixelsmash-critical-ffmpeg-vulnerability-turns-media-files-into-weapons/
1•croes•24m ago•0 comments

What is Gavin Newsom doing?

https://www.natesilver.net/p/what-is-gavin-newsom-doing
2•7777777phil•25m ago•0 comments

Show HN: Open-source tool for reverse engineering ChatGPT queries about brands

https://github.com/syntropicsignal-ai/ai-visibility-audit
1•biduskamil•25m ago•0 comments

A Review of Grand Theft Auto Xxiiv

https://taylor.town/gta23
1•surprisetalk•25m ago•0 comments