frontpage.
newsnewestaskshowjobs

Open Source @Github

fp.

Open in hackernews

Why AI Is Incorrigibly Didactic

https://www.theatlantic.com/ideas/2026/06/ai-writing-style-literature/687536/
5•Michelangelo11•1h ago

Comments

Terr_•1h ago
I really like the framing in this piece: LLMs are fundamentally document-generators, and keeping that in mind helps protect against our own human cognitive-illusions. When we read any story, we instinctively start imagining characters, simulating them, and--on some level--believing that they independently exist in some world we can't quite reach.

Even if it's just: "Fortune for today: Help, I'm trapped in a fortune-cookie factory."

> Today, savvy readers use a similar approach to identify AI writing, by hunting for supposed telltale signs. The em dash and the “it’s not X; it’s Y” construction are the prognathous jaw of the large language model, betraying its hidden inhumanity.

This feels particularly relevant for HN comments. :p

> For one thing, it has no fixed style, revealing that it has no fixed self.

This extends to the characters in the document, including when the style is first-person from an unnamed narrator. The helpfulness of the "AI Assistant" character equally-unreal as the bloodthirst of "Count Dracula."

> AI writing never challenges the way we think or see. It can’t do so, even if you explicitly ask it to.

Or at any rate, the "challenges" are going to be styles already reflected from the training data.

Choji: Agents to take a customer ask to pull request autonomously

https://twitter.com/chptung/status/2067675275513798730
1•chptung•32s ago•0 comments

GLM-5.2: Benchmarks, Architecture and How to Run It

https://www.techaffiliate.in/blog/glm-5-2-how-to-use-free-benchmarks-review
1•Aditya_kachhawa•2m ago•0 comments

Infrastructure Is the Source of Truth

https://substack.com/@markellens/p-202605003
1•gpi•3m ago•0 comments

Compile Zod (30x faster Zod validation)

https://gajus.com/blog/optimize-zod
1•gajus2•5m ago•0 comments

Wolves outperform dogs in following human social cues

https://www.sciencedirect.com/science/article/abs/pii/S0003347208003631
2•berkeleyjunk•6m ago•0 comments

Show HN: ToNotDo – a tool that helps you know what NOT to do, not a todo app:)

https://tonotdo.cc
1•maz225•8m ago•0 comments

Ask HN: What is the job market like?

2•gardnr•8m ago•0 comments

GAN vs. VAE vs. Diffusion

https://research.rudrite.com/compare/gan-vs-vae-vs-diffusion
1•mridul_sahu•9m ago•0 comments

Show HN: 1pager: Fixing the AI verbosity problem

https://github.com/cfitzgerald-pd/1pager/tree/main
1•bennydog224•9m ago•0 comments

Billionaire tax proposal declared eligible for the November ballot

https://www.latimes.com/california/story/2026-06-17/controversial-billionaire-tax-proposal-appear...
1•1vuio0pswjnm7•9m ago•0 comments

Adding a give_feedback tool to our MCP server

https://www.sanity.io/blog/how-to-get-product-feedback-from-agents
1•kmelve•9m ago•0 comments

Matthias Wandel's web page overflow area

http://matt.wandel.ca
1•OJFord•9m ago•0 comments

Experimental Treatment Kills Prostate Tumor Cells While Increasing Immunity

https://news.weill.cornell.edu/news/2026/06/experimental-treatment-directly-kills-prostate-tumor-...
1•gmays•9m ago•0 comments

Setting up PostgreSQL for running integration tests

https://gajus.com/blog/setting-up-postgre-sql-for-running-integration-tests
1•thunderbong•9m ago•0 comments

Did AI write this article?

https://www.economist.com/graphic-detail/2026/06/16/did-ai-write-this-article
1•1vuio0pswjnm7•10m ago•0 comments

Claude Code now supports artifacts

https://claude.com/blog/artifacts-in-claude-code
1•mfiguiere•10m ago•0 comments

Updates to USTP-Secure

https://github.com/x1colegal/USTP-Secure
1•x1colegal•11m ago•0 comments

Don't test what you don't control

https://www.echooff.dev/blog/dont-test-what-you-dont-control
1•lo1tuma•11m ago•1 comments

Kubernetes Runs on Feedback Loops

https://arslan.io/2026/06/16/kubernetes-runs-on-feedback-loops/
1•gpi•11m ago•0 comments

eBPF and WASM – Better Together

https://labs.iximiuz.com/tutorials/ebpf-wasm-plugin-46a23db3
1•dorkamotorka•11m ago•0 comments

Vance slams Israeli 'freakout' over Iran deal

https://www.timesofisrael.com/liveblog-june-18-2026/
1•tcp_handshaker•12m ago•0 comments

Shipped the Most Advanced Blockchain Development Vibecoding Infrastructure

https://twitter.com/SelaruDaniel1/status/2067545926290362831
1•danielul•13m ago•1 comments

The 'Me' Decade (1976)

https://nymag.com/article/tom-wolfe-me-decade-third-great-awakening.html
1•Michelangelo11•14m ago•0 comments

Show HN: A knowledge engine that keeps up with AI driven code

https://codesummary.io/
1•jerrodcodes•16m ago•0 comments

10 years of open source ClickHouse

https://clickhouse.com/blog/open-source-10
3•tosh•16m ago•0 comments

Towards noninvasive blood count using a deep learning pipeline

https://www.nature.com/articles/s41746-026-02598-2
1•wslh•17m ago•0 comments

How we built Cedar with automated reasoning and differential testing (2023)

https://www.amazon.science/blog/how-we-built-cedar-with-automated-reasoning-and-differential-testing
2•pramodbiligiri•17m ago•0 comments

Why DuckDB

https://duckdb.org/why_duckdb
2•tosh•18m ago•0 comments

Anthropic confident of re-enabling Mythos, Fable 5 access 'in coming days'

2•getbowtied•19m ago•0 comments

Show HN: I made a stablecoin app that works with checkout QR codes

https://yatori.io/yatori-pay
1•will-yatori•19m ago•0 comments