Hacker News Digest

Sunday, May 10, 2026

In This Issue

Hacker News
Internet Archive Switzerland
I Will Not Add Query Strings to Your URLs
LLMs Corrupt Your Documents When You Delegate
Meta's Embrace of A.I. Is Making Its Employees Miserable
The hypocrisy of cyberlibertarianism
US Government releases first batch of UAP documents and videos
Show HN: NanoCorp – Create autonomous companies run by AI
Gen Z Resentment Toward AI Grows as Adoption Stagnates and Workplace Fears Mount
Show HN: Crit – local review tool for agent plans and code diffs

Zipper Data Brief

May 10, 2026

Your daily digest of the best from Hacker News

Top 6 Trending

Internet Archive Switzerland

627 points · hggh · comments

Discussion Summary
The discussion celebrates Internet Archive Switzerland as an important distributed backup for digital preservation, though commenters note the website itself is poorly designed/functional, raise concerns about copyright liability and independence from the US-based Internet Archive, and debate the merits of archiving generative AI models.

I Will Not Add Query Strings to Your URLs

438 points · susam · comments

Discussion Summary
A website owner is rejecting all incoming URLs with query strings to protest tracking parameters added by social media platforms and other sites, sparking debate about whether this is a principled stance against privacy invasion or an unreasonable punishment of users who didn't add those parameters themselves.

LLMs Corrupt Your Documents When You Delegate

422 points · rbanffy · comments

Discussion Summary
The paper demonstrates that LLMs degrade documents through repeated edits, but commenters argue the results mainly reflect poor experimental design (basic file I/O tools, round-tripping text) rather than fundamental model limitations, and that proper agent architecture with specialized editing tools, version control, and deterministic output (diffs) could mitigate the problem.

Meta's Embrace of A.I. Is Making Its Employees Miserable

400 points · JumpCrisscross · comments

Discussion Summary
Meta employees are experiencing miserable working conditions due to aggressive AI adoption, invasive monitoring systems, and a toxic, fear-based culture where management prioritizes metrics and cost-cutting over employee welfare. The discussion reflects broader frustration with forced AI integration, job insecurity, and skepticism toward Zuckerberg's leadership across multiple failed initiatives.

The hypocrisy of cyberlibertarianism

349 points · ColinWright · comments

Discussion Summary
The discussion critiques how early internet idealism—particularly cyberlibertarian principles of freedom and decentralization—was abandoned by tech companies once they achieved massive scale and power, becoming hypocritical enforcers of the very controls they once opposed. Commenters debate whether the internet has ultimately improved or worsened human society, with disagreement over whether current problems stem from too much freedom, too little regulation, or fundamental flaws in applying outdated ideologies to complex systems.

US Government releases first batch of UAP documents and videos

329 points · david-gpu · comments

Discussion Summary
Most commenters dismiss the released UAP documents as mundane military footage (balloons, birds, missiles) mislabeled as unknown, with some viewing the release as political distraction or propaganda rather than evidence of extraterrestrial activity.

AI / Machine Learning

Show HN: NanoCorp – Create autonomous companies run by AI

13 points · AdrienBA · comments

Discussion Summary
NanoCorp is an AI platform that lets users create autonomous companies from prompts, having generated $924k ARR since launch with 15,000+ companies created. Commenters question the depth of what these AI companies can actually do beyond basic landing pages and payment links, and suggest the founders should run their own companies at scale if they're truly autonomous.

Gen Z Resentment Toward AI Grows as Adoption Stagnates and Workplace Fears Mount

66 points · mgh2 · comments

Discussion Summary
Gen Z's growing resentment toward AI stems not from opposing technology itself, but from rational economic self-protection—AI is automating entry-level jobs that traditionally provided pathways to expertise and career advancement, while senior positions remain inaccessible without experience, creating a structural bottleneck that leaves younger workers with diminished opportunities and wealth concentration benefiting only corporations and the wealthy.

Show HN: Crit – local review tool for agent plans and code diffs

12 points · tomasz-tomczyk · comments

Discussion Summary
Users praise Crit as an essential local review tool for LLM agent outputs and code diffs that streamlines their workflow and fills a gap in existing development tools. The creator expresses gratitude for the positive reception and community feedback on their first major open-source project.

LLMorphism: When humans come to see themselves as language models

24 points · okey · comments

Discussion Summary
The discussion criticizes the paper for building and attacking a strawman without defining its core terms or engaging with actual counterarguments, while commenters debate whether comparing human cognition to LLMs is genuinely novel or just the latest iteration of humans metaphorically understanding themselves through current technology.

Pentagon will 'never again' rely on a single AI provider, official says

10 points · billybuckwheat · comments

Discussion Summary
The single comment questions why the Pentagon doesn't develop its own internal AI infrastructure rather than relying on external vendors, suggesting skepticism about their stated diversification strategy.

Startups / Business

Anthropic weighs deal for near $1T valuation as revenue surges

6 points · mfiguiere · comments

Discussion Summary
The discussion expresses skepticism about Anthropic's $1 trillion valuation, with commenters questioning whether the valuation is justified given current revenue, comparing it to other tech bubbles, and debating whether AI companies can achieve the returns needed to support such valuations.