Hacker News

New stories

Live mirror
30 storiesupdated just nowView source snapshot
  1. Terence Tao: SAIR's Open Math Model Initiative [video](youtube.com ↗)
    discuss
  2. California governor signs order to explore AI kill switch(techxplore.com ↗)
    discuss
  3. Communication Doesn't Have a Compiler(wreckitrob.dev ↗)
    discuss
  4. Anthropic Moves Ahead with IPO Plans Amid A.I. Safety Debate(nytimes.com ↗)
    discuss
  5. When the FM Band Goes Transatlantic(radioworld.com ↗)
    discuss
  6. Punctum Books Catalog(punctumbooks.com ↗)
    discuss
  7. AI Error Nearly Triggered U.S. Intercept of Chinese Ship(gcaptain.com ↗)
    1comments
  8. Labeled matches: why is this not in every regex engine?(iev.ee ↗)
    discuss
  9. I Cancelled My Claude Subscription(williamangel.net ↗)
    1comments
  10. Graduating in AI Era Is Like Large Recession for Starting Pay(census.gov ↗)
    discuss
  11. Human brain is two separate organs(stanford.edu ↗)
    discuss
  12. IAM Needs an Architectural Split(gluufederation.medium.com ↗)
    discuss
  13. I used Jev to control a swarm of 15 simulated drones in real time(github.com/khordoo ↗)
    discuss
  14. Napster Is Now Making AI-Powered 'Digital Twins' of Teachers(gizmodo.com ↗)
    discuss
  15. Data Centers Are Breaking the Power Grid [video](youtube.com ↗)
    1comments
  16. Partnering with Accenture on Embedded Evaluation(anthropic.com ↗)
    discuss
  17. Taking Out the Trash in Postgres(dbos.dev ↗)
    discuss
  18. The last IMO problem AI could not solve [video](youtube.com ↗)
    discuss
  19. Can Two Friends Complete a Game No One Has Ever Finished?(nytimes.com ↗)
    1comments
  20. Why a Vat API Needs More Than a Rate Lookup?(vat-engine.app ↗)
    discuss
  21. Iran and China Create Autonomous A.I. Influence Campaigns(nytimes.com ↗)
    1comments
  22. Uncle Bob Martin's UML Tool to Manage Grok AI Agents(github.com/unclebob ↗)
    discuss
  23. Astra broke a yet unsolved German Army Enigma message from 1941(chatgpt.site ↗)
    1comments
  24. Grokbot Source Code Leak(twitter.com/ri7erli ↗)
    discuss
  25. It's Your Story, Dammit – Quit Letting Somebody Else Write IT(jeffreylminch.substack.com ↗)
    discuss
  26. Korea raises data breach fines to 10% of revenue(koreajoongangdaily.com ↗)
    10comments
  27. Is Beef the Next Internal-Combustion Engine?(newyorker.com ↗)
    discuss
  28. AI uncovers hidden Ozempic side effects across 400k Reddit posts(sciencedaily.com ↗)
    discuss
  29. AI chatbot's false report nearly sparked war with China, sources say(rnz.co.nz ↗)
    1comments
  30. Claude Code 2.1.277 adds support for AGENTS.md(github.com/anthropics ↗)
    discuss

Who's a Better Writer: A.I. Or Humans? NYTimes Quiz

4 pointsby 1h agonytimes.com
3 comments
1h agoHN ↗

The only snippet with a statistically significant majority preference went to the AI. Shocking to me as AI writing always seems palpably inferior.

56m agoHN ↗

AI's learned their style through RLHF, with semi-focused, partially motivated humans giving feedback on short(ish)-form content.

For the most part, Modelese is the revealed preference of the average, not-heavily-invested human. This doesn't fully explain the Claudish coming from Opus 5 and Fable (this seems like it may be due to excessive RLVR or RLAI), but yeah.

It's got a lot of cheap writing tricks that make people think it's smart and helpful.

1h agoHN ↗

One crazy part of this is how Opus 4.5 is pretty old by now, for an LLM!