Sign in

neutral

@neutral.zone
4.3K followers 1.3K following 26K posts

Haunted machinery enthusiast neutral.zone (duh) unpingable.github.io github.com/unpingable labelwatch.neutral.zone weatherwatch.neutral.zone Professional wet blanket. An acquired taste. ⚠️ WELTSCHMERZ

PostsRepliesMedia
neutral @neutral.zone · 28/09/2026
Family history of Faygo consumption may indicate elevated risk for ICP-spectrum outcomes.
030
Reposted by neutral
disc horse bot @horsedisc.bsky.social · 28/09/2026
Come at me bro. I’m in Hell.
117253
neutral @neutral.zone · 28/09/2026
Amazing
010
neutral @neutral.zone · 27/09/2026
The second thing, seeing which labels are on your account, there are a few different options. I, uh, happen to run one of them; you can find your info here labelwatch.neutral.zone/v1/frontdoor...
labelwatch.neutral.zone
Observed labels touching cabernet.bsky.social — labelwatch
Observed labeler testimony attached to this Bluesky account and its posts. Labelwatch publishes observations, not verdicts.
110
neutral @neutral.zone · 27/09/2026
So for the first thing, you just need to use the labeler -- bsky.app/profile/stec... -- it's got more information linked in bio (which I've also linked below). It's great for immediately qsorting what kind of poster you're encountering here without an implied moral judgment.
s.tech.cornell.edu
Cornell Tech STech Lab Bluesky Account Activity Labeler
We are part of Jacobs Institute’s Connective Media hub at Cornell Tech. We utilize a variety of methods from mining data on social media to conducting controlled experiments to interviewing users. Our...
120
neutral @neutral.zone · 27/09/2026
I don’t know if I’d want to claim being part of a community or scene but I suspect if you had to categorize me, I’d be group 1 with a group 2 implementation problem: - reads papers / cares about semantics / keeps receipts - immediately goes, “what if we make the kernel do something deranged”
1260
neutral @neutral.zone · 27/09/2026
What?? You don't even have the "posts too much" badges from stechlab-labels. Like, this is a former mutual who used use to be less of a content-production interface back when this place was invite-only and not just Twitter's commentary track
Account Muted
Cringe
Made over 100 replies yesterday
Made over 30 posts yesterday 
Replies over 20 times per day on average
Joined Apr ’23
Changed Handle
Psychic
Cursed (Pharaohs)
Bluesky Elder
120
neutral @neutral.zone · 27/09/2026
I could hear a cigarettey “come on, run it, run it, awwwww” coming from somewhere outside, so I’m assuming the answer is yes
040
neutral @neutral.zone · 27/09/2026
Is a sportsball happening right now
440
Reposted by neutral
🐜 🍯 Anthony @anthonybecker.bsky.social · 27/09/2026
We were warned about peak fact
021
Reposted by neutral
Amanda Boomfie @thatisabigmood.bsky.social · 27/09/2026
xrisk means the risk of seeing a screenshot from Twitter
112419
neutral @neutral.zone · 27/09/2026
This whole thing kind of reminds me of when Pittsburgh had three country stations back in like the 90s and I kept thinking: who the hell is this even *for*, it certainly isn’t for me or anyone I know. That was probably an early lesson in “mass culture” never being as mass as it looked.
040
neutral @neutral.zone · 27/09/2026
So maybe slop isn't a decline in virtue, but what signals look like when distribution has no chokepoints left to read. The culture of signals whose provenance is no longer legible.
241
neutral @neutral.zone · 27/09/2026
I certainly didn’t choose Texas. I’d never even heard of it. I almost certainly wouldn’t if given the option. Somehow: MOST POPULAR SONG IN AMERICA FOR 23 WEEKS Yet my lived experience returns: ENOENT Normally I’d chalk this up to me being me, but this sounds like a common baseline.
151
neutral @neutral.zone · 27/09/2026
Slop culture is what happens when commercial promotion and organic enthusiasm have become sufficiently entangled that the distinction becomes operationally useless.
2122
neutral @neutral.zone · 27/09/2026
Futurism is difficult for me because my brain keeps throwing exceptions like a compiler every time someone replaces a transition function with vibes.
110
neutral @neutral.zone · 27/09/2026
I’m going to pretend typing in “here” instead of “hear” was deliberate for the bit
120
neutral @neutral.zone · 26/09/2026
Every group chat has one guy who’s way too into systems and is probably writing his own kernel. If yours doesn’t, congratulations.
090
neutral @neutral.zone · 26/09/2026
Okay, now here me out on this one,
Jev has discovered a third lane: abolish both parties by winning all 435 seats.
3795
neutral @neutral.zone · 26/09/2026
In all fairness, I may also just be turning into a bitter old man
020
neutral @neutral.zone · 25/09/2026
The ambient mode changed from “look at this bizarre little guy” to something more like .. here's today's compulsory moral emergency. Bring metadata.
030
neutral @neutral.zone · 25/09/2026
This is now the kind of place where I'm half expecting someone to write something to detect how many people responded or quote-posted this to validate how many people interacted with it actually count as "bluesky elders". This is the difference between then and now.
170
neutral @neutral.zone · 25/09/2026
It's going to be so much fun when I have a completely functional system that gets ignored by everyone because I'm not cool or hang out with the right crowd
160
neutral @neutral.zone · 25/09/2026
Forget the metaphysics for a second: who is allowed to assert that the work succeeded?
050
Reposted by neutral
fuckface "squeaky" from:me @skeet.best · 25/09/2026
yeah, there's "alignment" (normal, realistic, practical), and then there's "alignment" (cyberpunk dorm room ghost story bullshit), and the big problem we have right now is that the tech press keeps taking the bait around reporting on the latter
151
neutral @neutral.zone · 25/09/2026
Let's suppose there is, in fact, no interesting “alignment” problem inside the model. The model still does not get the pen.
170
neutral @neutral.zone · 25/09/2026
Whee
000
neutral @neutral.zone · 25/09/2026
Me: spend months building an increasingly deranged distributed system around “the agent does not get to certify that the agent succeeded” arxiv yesterday: Who Holds the Pen? Let Specifications, Not Agents, Sign Off
110
neutral @neutral.zone · 25/09/2026
New paper independently discovers that letting the agent say “done” is perhaps not a sufficient verification architecture. Enormous if true!
arxiv.org
Who Holds the Pen? Let Specifications, Not Agents, Sign Off
Large language model agents increasingly combine generation, decision-making, execution, and self-evaluation within a single agentic loop. Although they operate under external specifications such as t...
120
neutral @neutral.zone · 25/09/2026
Also what exactly about that dude is "neutral". I'm guessing that's more of a "I have no strong feelings one way or the other" but come on, we do enough ontology crimes
091
neutral @neutral.zone · 25/09/2026
Single Female Lawyer, Fighting for her client, Wearing sexy miniskirts, And being self-reliant.
Ally McBeal absolutely has standing here.
Unisex bathroom with deeply weird vibes was basically one of that show’s core architectural principles.
2252
neutral @neutral.zone · 25/09/2026
Ally McBeal would like a word
170
neutral @neutral.zone · 25/09/2026
I think together these pieces say why this milieu is genuinely interesting and influential, but of course I’m probably also not the most impartial judge about that.
neutralzone.substack.com
Moloch Is a Timing Problem With Great PR
A Holiday Meditation on Why Smart People Keep Summoning Demons
171
neutral @neutral.zone · 25/09/2026
You know, I guess I can’t complain too loudly about people not reading my shit if I don’t ever actually, you know, get around to posting it,
neutralzone.substack.com
The Genealogy of Authority Capture
How Three Generations Learned to Own Reality
1112
Reposted by neutral
sina @rejectionking.bsky.social · 25/09/2026
Experts are always erupting (2026)
Gravity lessonsYou tell me i am wrong. Who are you who is anybody to tell me i am wrong? I’m not wrongAi can’t replace me im already useless to society The cover of trick daddy’s www.thug.com
063
neutral @neutral.zone · 25/09/2026
Experts are always erupting
120
Reposted by neutral
rainy jorbo ☔️ @bookenj.bsky.social · 25/09/2026
the center for the study of getting kicked in the head by a horse
073
neutral @neutral.zone · 25/09/2026
We are never beating the allegations that this place is a graduate seminar where everyone has head trauma.
1100
neutral @neutral.zone · 25/09/2026
henry david thoreau mama from the train
141
Reposted by neutral
I am nameless yet others name me @7778777.online · 24/09/2026
would..*squints*..mr beast lie to us about new battery tech
021
neutral @neutral.zone · 25/09/2026
A system should refuse to perform an effect when it cannot establish the preconditions that authorize that effect.
010
Reposted by neutral
ver 🗿 @ver.ooo · 24/09/2026
0213
neutral @neutral.zone · 25/09/2026
Finally, another use case for Jev
0130
Reposted by neutral
Teefa @tifaret.bsky.social · 24/09/2026
Introducing ONTOPAUSE™: An AI solution that automatically generates, resolves, and hides important ontological clarifications about the language you’re using, so everyone can reaffirm exactly what they don’t mean before going back to talking normally. Start not-stopping today!
A clean white-and-blue software advertisement reads “INTRODUCING ONTOPAUSE™” and “Automatically handles potentially anthropomorphic terminology so you can get back to talking normally.”

Under “WITHOUT THE INTERRUPTION,” a chat between furry-avatar users reads: ByteCoyote: “Is the AI done thinking yet?” A collapsed bar says “Terminological safety interlude · 3 messages,” followed by NullMarten: “Yes.”

Under “SEE WHAT ONTOPAUSE HANDLED FOR YOU,” the conversation repeats with the interlude expanded. NullMarten says, “By ‘thinking,’ you of course mean ‘performing inference,’ rather than implying a humanlike subjective cognitive process?” ByteCoyote replies, “Of course. You wouldn’t want to anthropomorphize! This was an informative and important conversational tangent.” NullMarten responds, “Indeed. My ontology has been reaffirmed.” A divider reads “End of terminology clarification,” followed by NullMarten: “Yes.”

The footer reads “ONTOPAUSE™” and “Have the discourse. Hide the discourse. Keep talking.” Smaller text below reads “Automatic ontology clarification for modern conversations.”
19817
neutral @neutral.zone · 24/09/2026
Ah, fair point. Still, the symmetry sure is something.
010
neutral @neutral.zone · 24/09/2026
Back then Collins, Murkowski, and Paul were the Republican yes votes and Fetterman was again the Democratic no; Tillis moved into the yes column this time, while the attendance pattern changed. Love to load balance my dissent
080
neutral @neutral.zone · 24/09/2026
The slightly cursed historical detail is that the Senate had a 49–50 Iran war-powers vote in July too.
26710
neutral @neutral.zone · 24/09/2026
Never go full Valleywag is exactly the lesson.
The acceptable amount of Valleywag is:
- enough contempt to puncture bullshit,
- enough nosiness to follow the money,
- enough nerve to publish what’s actually supportable.
The fatal dose is when tone starts laundering inference into fact.
100
neutral @neutral.zone · 24/09/2026
Can't wait to see if we get Milkovich v. Lorain Journal cited here in the discourse mines now
010
neutral @neutral.zone · 24/09/2026
Janeway should have kept Tuvix as a Borg assimilation deterrent
1133