frontpage.
newsnewestaskshowjobs

Made with ♥ by @iamnishanth

Open Source @Github

fp.

SteelSpine: Replay tool for debugging AI agents

https://steelspine.ai
1•jeremyfelps•2m ago•1 comments

Interpreting Polygenic Prediction of Cognitive Ability

https://icajournal.scholasticahq.com/article/158459-interpreting-polygenic-prediction-of-cognitiv...
1•gmays•2m ago•0 comments

What I've Learned from Agentic Design

https://chrislachance.com/what-ive-learned-from-agentic-design/
1•speckx•3m ago•0 comments

Bun's unreleased Rust port has 13,365 unsafe blocks

https://bun.com/bun-unsafe-audit
2•helloplanets•3m ago•0 comments

Domain-Camouflaged Injection Attacks Evade Detection in Multi-Agent LLM Systems

https://arxiv.org/abs/2605.22001
1•sbulaev•4m ago•0 comments

Packj flags malicious/risky open-source packages

https://github.com/ossillate-inc/packj
2•my2c•8m ago•0 comments

The ten steps towards a dictatorship

https://nonogra.ph/the-ten-steps-towards-a-dictatorship-05-22-2026
2•han1•9m ago•0 comments

Marketing sites don't need a CMS anymore

https://frigade.com/blog/from-framer-back-to-code
1•pancomplex•10m ago•0 comments

How did Asterix and Obelix learn to speak fluent Hindi? (2019)

https://www.indulgexpress.com/cover/2019/Apr/05/were-going-desi-by-toutatis-how-did-asterix-and-o...
1•ripe•13m ago•1 comments

Show HN: Generate free golf yardage books from OpenStreetMap data

https://npilk.github.io/openyardage-web/
1•npilk•13m ago•0 comments

Batty: A Multiplexer Built on Ghostty

https://batty.sstools.co/
1•smallsharptools•15m ago•1 comments

NYSE Compression Case Study

https://code.kx.com/q/kb/compression/fsicasestudy/
1•tosh•16m ago•0 comments

God Is Technology (2010)

https://archive.org/download/MitchellHeismanSuicideNote/Mitchell%20Heisman%20-%20Suicide%20Note.p...
2•senkora•16m ago•1 comments

Show HN: macOS utility to record and playback mouse, keyboard events

https://github.com/uAIex/KeyMouseRecorder
2•harr01•16m ago•0 comments

Why the NTSB Shut Down Their Plane Crash Report Archive [video]

https://www.youtube.com/watch?v=phjRQckjVJc
4•cratermoon•22m ago•1 comments

Interview with the Engineer of Uruky, a Private Search Engine

https://theprivacydad.com/interview-with-the-engineer-of-uruky-a-private-search-engine/
1•Brajeshwar•23m ago•0 comments

OpenCode and Cursor's Composer 2.5

https://cursor-api.standardagents.ai
4•lcavalcare•24m ago•0 comments

Google Updates Android Bug Bounty Program with $1.5M Offer

https://www.forbes.com/sites/daveywinder/2026/05/05/google-to-pay-15-million-for-pixel-phone-secu...
2•Cider9986•26m ago•0 comments

Herasight found embryo with potential IQ score in 99.99th percentile

https://twitter.com/GeneSmi96946389/status/2057860062039920682
2•gmays•26m ago•0 comments

Schwung – open up your Ableton Move

https://schwung.dev/
1•Tomte•27m ago•0 comments

Shortcuts Playground: Create Apple Shortcuts with Claude Code/Codex

https://www.macstories.net/stories/introducing-shortcuts-playground/
1•logged4upvoting•29m ago•0 comments

60 Percent of Grades at Harvard Were A's. Enough Is Enough

https://www.nytimes.com/2026/05/21/opinion/harvard-easy-a-grades.html
2•paulpauper•29m ago•4 comments

AI as a Design Medium

https://www.harvarddesignmagazine.org/articles/ai-as-a-design-medium-rodenbeck/
2•CharlesW•29m ago•0 comments

Secure Boot and Microsoft CA Rollover – a heads-up for distributions

https://blog.einval.com/2026/05/22#secure_boot_ca_rollover
1•speckx•29m ago•0 comments

Social Science at the NSF

https://www.abundanceandgrowth.org/p/social-science-at-the-nsf
1•paulpauper•30m ago•0 comments

Frontier labs don't use most AI compute (yet)

https://epochai.substack.com/p/frontier-labs-dont-use-most-ai-compute
1•gmays•30m ago•0 comments

Meta Released a Reddit-Like App Built Around Facebook Groups

https://firethering.com/meta-forum-reddit-like-app-facebook-groups/
3•steveharing1•30m ago•0 comments

At Tyler Cowen University, No One Would Have Tenure

https://www.bloomberg.com/opinion/articles/2021-07-05/what-would-your-fantasy-university-look-like
1•paulpauper•30m ago•1 comments

Open source Kanban desktop app that runs parallel agents on every card

https://www.kanbots.dev/
18•vitriapp•32m ago•6 comments

Learn Shortest Path with step by step playback

https://algo-rhythm.dev/en/shortest-path/
6•bytego•32m ago•0 comments
Open in hackernews

Ask HN: Do people lie about why they hate AI writing on social media?

1•amichail•47m ago
I think the following might be the case:

* people do not distinguish between AI writing based on the human's ideas vs AI writing based on the AI's ideas

* they do this intentionally as a way to denigrate all AI writing — even when the content is good/interesting

* and the reason they do all this is to delay their unemployment as a result of AI

So in the end, they use social media as a way to make people think all AI writing is bad because this is how they are trying to delay their unemployment as a result of AI.

Comments

yepyoukno•40m ago
People honestly don’t know what they want. I know, sounds impossible. Love hate relationships are abundant throughout humanity and everyone who knows how to get what they want knows how to play upon this multiplicity.
obpe•40m ago
If AI writing was good enough then they wouldn't be able to tell. Also, some people lying on social media won't change any minds about whether they will be fired.
amichail•33m ago
Even when you can detect when something is written by AI, you don't have to criticize it if the content is good/interesting.

But denigrating AI on social media is helpful if you think it can delay your unemployment as a result of AI.

bigyabai•38m ago
> they do this intentionally as a way to denigrate all AI writing — even when the content is good/interesting

If people only cared about AI for fiction writing, this would be fine. "Good/interesting" writing can be tolerated when nothing is at stake.

For nonfiction writing however, AI is rhetorically neutered. LLMs have no inherent authority/ethos without citing non-AI works. LLMs do not possess logic/logos that is rooted in the real, human experience. There is no emotional/pathos appeal that can be made between a human and a nonliving entity.

Aristotle et. al. would have considered most AI writing a waste of digital parchment. How would you feel if I answered this question with Qwen or GLM instead of hand-writing it?

LautaroGartner•22m ago
AI is becoming better by the day. In the first few models it used to be cliche and very template-like. Nowadays is exceptional but it still remains with a few obvious things like the over use of hyphens — but other than that. People hate for hating; I'd say 8 out of 10 people use it. Generated images have become really good tho but people over using it without criteria is a problem. Other than that, AI is great, except for the layoffs but is typical in new technologies for that type of transition.