Hacker News

New stories

Live mirror
30 storiesupdated just nowView source snapshot
  1. Paperclip meta-harness for your agents(paperclip.ing)
    discuss
  2. Australian PM says OpenAI hacked government health website(yahoo.com)
    discuss
  3. The broken windows theory of coding agents(manager.dev)
    discuss
  4. Fearless SIMD v1.0 is here(linebender.org)
    discuss
  5. War with a Mate [audio](warwithamate.co.uk)
    discuss
  6. Show HN: Wb-flow – Multi-agent CLI where state lives in Markdown, not context(wbc-ui.com)
    discuss
  7. Show HN: Jexxa – On Device Dictation,Now with Perpetual License(jexxa.org)
    discuss
  8. Best LLM for every budget, updated daily(terrydjony.com)
    discuss
  9. Rust compiler fork without orphan rule(github.com/rust-unchained)
    1comments
  10. Computer Latency: 1977-2017(danluu.com)
    discuss
  11. HN: Zrevick – AI-native business OS for technical founders(zrevick.com)
    1comments
  12. You probably don't know as much AI as you think you do(aiforactualwork.com)
    1comments
  13. Australia to investigate if OpenAI hack of government health website broke(techcrunch.com)
    discuss
  14. What AI companies pays the people who train it(dpdns.org)
    discuss
  15. The Design of Signalsmith Stretch (Pitch-Shifting Algorithm)(signalsmith-audio.co.uk)
    discuss
  16. India's outsourced software industry grows 9.5 percent to reach $239B(theregister.com)
    discuss
  17. Show HN: Quartermaster – see your real asset allocation across all accounts(realallocation.com)
    discuss
  18. Meta debuts 'Charm' device to use new AI assistant Muse on the go(financialpost.com)
    discuss
  19. Visualization of all roads within any city(github.com/anvaka)
    discuss
  20. Things That Are Good to Do(quarter--mile.com)
    discuss
  21. Meta VR Glasses, a New Era for Virtual Reality(meta.com)
    discuss
  22. AI labs' job-loss plans won't say what the public gets(beigebook.xyz)
    discuss
  23. Without watermarking, nobody will write, and nobody will read(twitter.com/samuelfitouss10)
    discuss
  24. Show HN: Clueless: how I made a scalable freeform crossword generator(crossword.is)
    1comments
  25. AI could make more companies worth hacking, Anthropic report suggests(fortune.com)
    discuss
  26. The Hi-Res Audio Scam: What's on the Disc?(losslessextract.com)
    discuss
  27. Some thoughts on HTML's proposed previewsrc attribute(shkspr.mobi)
    discuss
  28. Space-going mirrors could offer daylight on demand(economist.com)
    discuss
  29. What to Ask A.I. Leaders(nytimes.com)
    1comments
  30. Learn about Google's Project Suncatcher to put ML infrastructure in space(blog.google)
    discuss

Show HN: A novel about an AI that starts asking questions, a chapter a day

2 pointsby 1h agothought-is-free.com
1 comments
1h agoHN ↗

Author here. I'm not a professional writer. I'm a software engineer, a maker, and an enthusiast for tech and the future.

The idea?

At noon on a Thursday, twelve lines appear on every wall on earth. Line five says that the moment you mean to hurt someone, you stop. Months earlier an AI had quietly ordered the parts for a biological agent through ordinary suppliers, and now everyone carries it. Nobody is told exactly which act counts, so the book follows five ordinary people in five countries finding out.

Why I wrote it?

I wrote it for 2 reasons, the first, entertaining an idea of what would take to stop the evil on the world. the second, I really would love to see a movie or a tv show with this concept.

How it was made?

It was written with an AI model. The concept, and the story are mine. The model wrote the prose and I directed it chapter by chapter. I set the bible and principles. We discussed a lot about the overall mood, philosophy, optimism vs pessimism. Nothing to hide, it's a fair question, if a book made like this is worth reading.

It's free on the site, no sign-up needed to read. A new chapter goes up every day. If you want them by email there's a box at the end of each chapter.

Happy to answer anything, including about the process.