frontpage.
newsnewestaskshowjobs

Made with ♥ by @iamnishanth

Open Source @Github

fp.

Open in hackernews

I can never talk to an AI anonymously again

https://www.theargumentmag.com/p/i-can-never-talk-to-an-ai-anonymously
2•momentmaker•1h ago

Comments

goodmythical•3m ago
This is a bit defeatist, isn't it?

Were on sufficiently motivated to ensure anonymity, one would not be posting thousands of words of content associated with their government ID.

Even if one were still completely dead set on doing so, it's wrong to assume that it is impossible to sufficiently alter one's writing style while chatting with using the offending systems to anonymize oneself.

Take just the first sentence: "I am, myself, passionately and slightly fanatically on the pro-anonymity side." There are several stylistic choice that are made and re-used throughout the article. First person voicing, redundant adjectives and adverbs, etc.

Compare that to "I believe in anonymity" or "Anonymity is good" or "My preference leans toward anonymity". Or even "I emphatically believe in anonymity as an inalienable human right."

There are so many long sentences, so many "if this then that" statements, so many repetitions of the same point in different words... If all the content from this writer is as self-same as this article itself is, why should it not be expected that one could identify it?

We've been de-anonymizing authors since before LLMs, and some authors have indeed been able to write anonymously by sufficiently changing their style.

I mean, compare: ~~~ This is some student work, shared with the student’s permission (they reviewed this blog post and gave it the okay). These three assignments (writing about a student-chosen topic, in this case Pokemon) show the student’s progression over the course of two months after we decided to focus with this student on developing their writing skills. The first one I would say is about first-grade level work: the student is writing correct and complete sentences, but the sentences are simple; their handwriting is mostly legible with a few problem letters. The second one I would say is about second-grade level work: the student is writing longer and more varied sentences, with a range of constructions “Perhaps it was sneaking up on prey?”. They’re attempting more complicated vocabulary words (I’m told that a misspelled word at the top of the page was meant to be ‘roguish’.) ~~~

to

~~~ With the learner’s explicit consent and a final review of this post, I am pleased to share a window into their academic growth. These three pieces—centered on the student’s personal interest in Pokémon—chart a two-month trajectory of intensive literacy development.

The evolution of their craft is summarized below:

    Initial Baseline (First-Grade Equivalent): The student began by producing complete, accurate, yet rudimentary sentences. While the work was communicative, the syntax remained basic, and penmanship was transitional, containing a handful of inconsistent letter forms.

    Intermediate Progress (Second-Grade Equivalent): Following a targeted focus on composition, the student’s work shifted toward greater syntactic variety. They began experimenting with inquisitive structures (e.g., “Perhaps it was sneaking up on prey?”) and more sophisticated vocabulary. Notably, an early attempt at the word "roguish" demonstrates a willingness to engage with complex language beyond their previous comfort zone.
~~~

You see how there's much less "I think this/I would say this", a clearer structure, etc? I don't have opus, so I'd be very interested to see if the rewrite is identifiable as the author.

The Stanford Freshmen Who Want to Rule the World

https://www.theatlantic.com/ideas/2026/04/stanford-students-power/686920/
1•apparent•1m ago•0 comments

Zerodep: Performant single-file, zero-dep Python modules (protobuf, YAML, etc.)

https://github.com/Oaklight/zerodep
1•Oaklight•5m ago•0 comments

Self-Hosted AI Red Team Tools

https://aetherverseintel.gumroad.com/l/vpzqnk
1•valuria•9m ago•1 comments

Agentic AI Chip Design Built a Full RISC-V Core

https://spectrum.ieee.org/ai-chip-design
1•rbanffy•10m ago•0 comments

Microsoft Reportedly Looking at Rebasing Azure Linux on Fedora

https://www.phoronix.com/news/MS-Azure-Linux-Fedora-Based
1•rbanffy•11m ago•0 comments

With TPU 8, Google Makes GenAI Systems Better, Not Just Bigger

https://www.nextplatform.com/compute/2026/04/24/with-tpu-8-google-makes-genai-systems-much-better...
2•rbanffy•12m ago•0 comments

Revocation of X.509 Certificates

https://blog.apnic.net/2026/04/24/revocation-of-x-509-certificates/
1•jandeboevrie•12m ago•0 comments

Show HN: Auto lock your Mac via phone proximity (No Apple Watch)

1•dabbygabby•13m ago•0 comments

America's Geothermal Breakthrough Could Unlock a 150-Gigawatt Energy Revolution

https://oilprice.com/Alternative-Energy/Geothermal-Energy/Americas-Geothermal-Breakthrough-Could-...
1•sleepyguy•13m ago•0 comments

The Nintendo Switch Switch (2019)

https://blog.cynthia.re/post/nintendo-switch-ethernet-switch
1•birdculture•14m ago•0 comments

Show HN: A faster, drop-in replacement for Tailscale's DERP relay

https://hyper-derp.dev/blog/hyper-derp-announcement/
2•KRuskowski•15m ago•0 comments

Show HN: Phinite – The OS layer (eval, observe, govern, A2A native)

https://www.phinite.ai/
1•PhiniteAI•17m ago•0 comments

Show HN: Perfect Fit – A minimalist B&W timing puzzle

https://play.google.com/store/apps/details?id=com.FathaGames.PerfectFit&hl=en_US
1•mrfatha•18m ago•0 comments

#009: The Flavor of the AI Interface

https://metedata.substack.com/p/009-the-flavor-of-the-ai-interface
1•young_mete•20m ago•0 comments

Congrats on failing to launch; weirdfriends log 1

https://c0n0.com/posts/weirdfriends-log-1/
1•jealousgelatin•23m ago•0 comments

HDR output arrives in Godot 4.7

https://godotengine.org/article/hdr-output-arrives-in-godot-4-7/
1•__natty__•30m ago•0 comments

Firestarter malware survives Cisco firewall updates, security patches

https://www.bleepingcomputer.com/news/security/firestarter-malware-survives-cisco-firewall-update...
2•Brajeshwar•33m ago•0 comments

The Papers of Friedrich August von Hayek Now Accessible Online

https://www.hoover.org/news/papers-friedrich-august-von-hayek-now-accessible-online
2•paulpauper•36m ago•0 comments

Prediction Market Accuracy: Crowd Wisdom or Informed Minority?

https://papers.ssrn.com/sol3/papers.cfm?abstract_id=6617059
2•Anon84•37m ago•0 comments

Hedging the Singularity

https://twitter.com/achenfinance/status/2046612004870164671
1•paulpauper•37m ago•0 comments

Viewpoint Diversity

https://casssunstein.substack.com/p/viewpoint-diversity
2•paulpauper•38m ago•0 comments

GPT-5.5-Pro did worse in BullshitBench

https://twitter.com/petergostev/status/2047773402090426548
2•amrrs•39m ago•0 comments

Agent regression testing: cutting detection from days to minutes

https://www.polarity.so/blogs/agent-regression-testing-days-to-minutes
1•mihaipo•40m ago•0 comments

Show HN: Video Messaging for Work

https://flowylabs.ai/llink
2•talksik•49m ago•0 comments

Everyone in the world has to take a private vote by pressing red or blue

https://twitter.com/waitbutwhy/status/2047710215265730755
3•theharrychen•49m ago•0 comments

Show HN: Kloak, A secret manager that keeps K8s workload away from secrets

https://getkloak.io/
4•neo2006•49m ago•1 comments

Hermes.md in Git commit messages causes requests to route to extra usage billing

https://github.com/anthropics/claude-code/issues/53262
2•adunk•1h ago•0 comments

Show HN: Bulk URL Checker – check 75k URLs from any LLM via MCP

https://bulkurlchecker.com
1•carlosofscience•1h ago•1 comments

The Lunar Gateway's primary modules are corroded

https://arstechnica.com/space/2026/04/well-this-is-embarrassing-the-lunar-gateways-primary-module...
2•simonebrunozzi•1h ago•0 comments

Greenfield and Iterative deployment – By superpowers creator

https://primeradiant.com/blog/2026/greenfield-and-iterative-development.html
1•sorcercode•1h ago•0 comments