iiinova — where your digital self keeps evolving

Reposted by Christian Kündig

In sozialen Netzwerken und Medien kursiert das Gerücht, dass die ukrainischen Angriffe auf russische Rechenzentren die Aktivität rechter Social Bots minimiert hätte. Dafür gibt es bislang keine Belege. netzpolitik.org/2026/angriff...

netzpolitik.org

Aus brennenden Rechenzentren in Russland wird kein Bot-Ausfall bei der AfD

15277134

Reposted by Christian Kündig

Anil Dash @anildash.com · 19h

More to the point — people who stayed on the platform don't realize how inured to extremism they've become. They regularly get exposed to trending topics, and amplified content from verified accounts, that they would have recoiled from just a few years ago. I don't miss anything *not* being there.

689838

Reposted by Christian Kündig

maud @peloriemate.net · 10/10/2026

i got so many people calling this tweet fakeposting back when i made it but since this is bluesky and things have less reach on here im just going to post the photo

254205405164

Reposted by Christian Kündig

conputer dipshit @davidcrespo.bsky.social · 10/10/2026

nice summary of the confusion around ARR but I wish they would explain that the discrepancy with actual revenue is the *point* of ARR. when revenue is going up constantly, if you look at too long a window it’s outdated, and if you look at too short it’s noisy www.bloomberg.com/news/article...

The annualized estimates can differ notably from the actual revenue they expect to recognize in a calendar year. OpenAI is currently expecting to generate roughly $35 billion in revenue in 2026, half the annualized pace it’s projecting it will hit at year end, according to people familiar with the matter, who spoke on condition of anonymity as the information is not public.

That echoes a similar discrepancy seen in Anthropic’s actual revenue and the annualized pace that investors watch: While it took in roughly $4.6 billion in 2025, according to documents seen by Bloomberg News, by the end of that year it was pegging its annualized revenue at over $9 billion, Bloomberg News has reported. Its sales have grown at a fast clip since then.

3282

Reposted by Christian Kündig

W Don't Have Time @wdonthavetime.selfhosted.social · 09/10/2026

Strangely, the Name, Nationality and Date of Birth for @wsocial.eu leadership team inc. @anna.wsocial.eu @ingmar.wsocial.eu @aslan.wsocial.eu @petter.wsocial.eu and @sundstrand.wsocial.eu and all advisory board members remain empty. Don't they believe in their product? #WSocialFacts #WSocialPics

2676

Reposted by Christian Kündig

Rob Delaney @robdelaney.bsky.social · 09/10/2026

I will never, ever consider my audience when making art. You are a disgusting flesh piano I manipulate to make the noises that please me. That’s the Rob Delaney promise.

592679280

Reposted by Christian Kündig

werewolf ver mitzvah 🗿 @ver.ooo · 09/10/2026

tech bro is short for technically broccoli

525240

Christian Kündig @christian.kuendig.info · 09/10/2026

The thing is, who cares enough about SFDC that this (or the lack of it) would bubble up? When’s the last time Trump stepped foot into a commercial airport terminal or read Computerworld (lol)?

000

Reposted by Christian Kündig

Endless Screaming @infinitescream.bsky.social · 09/10/2026

AAAAAAAAAAAAAHHH

04715

Reposted by Christian Kündig

Tom @moll.dev · 09/10/2026

6785

Reposted by Christian Kündig

Tom @moll.dev · 09/10/2026

look

AppleTV connected to a C8 US adapter, plugged into a US to EU converter with USB plugs, with a Macbook charger brick wedged in the sode

3231

Reposted by Christian Kündig

Mr Market Miscalculates 🇪🇺🇺🇦🏴‍☠️ @mrmarketmis.bsky.social · 09/10/2026

Let's celebrate the 25th anniversary of the GOAT sell-side research report!

17024

Reposted by Christian Kündig

Krittika @krittikaa.bsky.social · 08/10/2026

Sorry are we just going to ignore “neither of my parents cook”

061

Reposted by Christian Kündig

Jerry Chen @jcsalterego.bsky.social · 07/10/2026

"take a hike" joke's on you, exercise would be great for my mental health

2960367

Reposted by Christian Kündig

William B. Fuckley @opinionhaver.bsky.social · 07/10/2026

At a certain point you gotta be willing to just say these are basically crazy people that nothing you can do will make happy, man. Like I’m sorry but if you’re protesting Mamdani over this people should just ignore you.

27977105

Reposted by Christian Kündig

A penitent who is loud @loudpenitent.bsky.social · 07/10/2026

on this day? No. Fuck ALL THE WAY OFF.

610019

Reposted by Christian Kündig

Citizen Platano 🇵🇷 @daniloc.xyz · 07/10/2026

I saw the HBO movie about McCain and Palin, Game Change they do something wild there: they almost play it like horror. The reveal of Palin’s true nature, after the compressed five day vetting, becomes a monster that eats the campaign Julianne Moore can do no wrong, and she’s spooky good

191

Christian Kündig @christian.kuendig.info · 07/10/2026

10 followers, maybe we dont need to engage every fool who manages to sign up here…

020

Reposted by Christian Kündig

Eugene Vinitsky 🍒 @eugenevinitsky.bsky.social · 07/10/2026

I have a pictorial proof that we live in hell

Square packing for 11 cubes. It is horrible

1420714

Reposted by Christian Kündig

Tom @moll.dev · 05/10/2026

the gamer entitlement is fucking insane. they're yelling at a developer who has spent a decade of his time writing Dolphin and PPSSPP for using Claude to build a prototype. the same people who don't pay for their emulator, certainly don't pay for ROMs are entitled to bully developers now?

724120

Reposted by Christian Kündig

Elixir of Progress @elucidating.extradimensional.space · 05/10/2026

Like you kinda knew it had to happen, right? Eventually every software project that's not the human supremacist version of temple OS will go to agentic engineering. Emulators are an *especially* fruitful case because you can a/b compare against legacy artifacts on your climber.

1171

Reposted by Christian Kündig

Pwnallthethings @pwnallthethings.bsky.social · 07/10/2026

But the models kept going, and eval runs now blow through some of those genuinely serious math problems at the frontier of the actual mathematical research community. So internal to the labs they end up with piles of genuinely serious math proofs that are (mostly) collateral output to the evals

1734

Reposted by Christian Kündig

skELLY @ellyzoe.bsky.social · 06/10/2026

It’s random because every time we had a problem we didn’t know how to approach, we called it AI, reasoning, “if computers could solve this, they’d *truly* be smart”. It has historically included databases, search, computer vision, etc! A ton of things! But when we got good at handling those things,

1384

Christian Kündig @christian.kuendig.info · 06/10/2026

What about a Server KILOBYTE H223-V10-AAW1 ?

000

Reposted by Christian Kündig

NZZ @nzz.ch · 06/10/2026

Russlands Terror gegen die Ukraine hat eine neue Stufe erreicht. Das Ziel besteht offensichtlich darin, Städte wie Kiew unbewohnbar zu machen. In dieser Situation braucht die Ukraine für ihre Verteidigung mehr denn je westliche Hilfe.

nzz.ch

Putin hat eine neue Strategie: Sie beruht allein auf Kriegsverbrechen

13711

Reposted by Christian Kündig

asura @asura.dev · 06/10/2026

Huh I missed this update I guess one good thing about all the nonsense is we finally have markdown RIP PDF, you were the worst

workspaceupdates.googleblog.com

Google Workspace Updates: Preview, edit and collaborate on Markdown (.md) files natively across Drive and Docs

2481

Reposted by Christian Kündig

Dave Richeson @divbyzero.bsky.social · 05/10/2026

I wanted to see how my calculus students could do creating a closed piecewise parametric curve for an impossible object. They made one side and could choose one of my shapes for the other side. Then I printed the objects. It went well, and I think I can ramp it up a little bit and

23816

Reposted by Christian Kündig

hailey @hailey.at · 06/10/2026

much anti-LLM arguing here is basically all the US users being mad at American capitalism but instead of yelling about American capitalism they yell about some other Glonzo they have made up in their head instead. most M-Ls here wouldn't know the best shot they've had if it slapped them in the face.

1325220

Christian Kündig @christian.kuendig.info · 05/10/2026

Not sure what to tell you, but LLM are literally “using ML in linguistics”! They evolved from earlier approaches in ML for natural language processing. LLMs are the current state of computational linguistics and are based on decades of research in this field.

3290

Reposted by Christian Kündig

rev. howard arson @theophite.bsky.social · 05/10/2026

this literally is not true

1556441

Christian Kündig @christian.kuendig.info · 05/10/2026

People are upset about llm/agent use for recomp projects, so I don’t think there much rationality behind it.

0170

Reposted by Christian Kündig

DRM @d-r-m.bsky.social · 05/10/2026

hrydgard has been coding the PPSSPP emu for ages,they know their stuff. IMO if they review the commits it’s fine. this zero AI policy won’t scale TBH as security fixes are now being found by AI. i’m more concerned by people doing throwaway minecraft clones than someone updating an emu the masses use

121

Reposted by Christian Kündig

Demigirlboss @demigirlboss.win · 05/10/2026

HOOOOOOOOOOWWWWWWWWWWW?!?!?!?!?!?!?!?!?! OH MY FUCKING GOD THIS IS SO INSANELY FUCKING COOL Y'ALL DON'T EVEN KNOW Y'ALL DON'T EVEN *UNDERSTAND* THIS IS SO FUCKING AWESOME I'M SO HYPE.

0543

Reposted by Christian Kündig

Eris ⌬ @eriskii.net · 05/10/2026

Tag yourself

1826419

Reposted by Christian Kündig

Asa ⭔ @asa.engineer · 05/10/2026

just heard the news about PPSSPP — now the community is going to have to fork it and rewrite all the changes made by Claude. A daunting task given the sheer volume of code, but fortunately there are modern tools that make this achievable,

619226

Christian Kündig @christian.kuendig.info · 05/10/2026

These people are not well. The “receipt” screenshot make a good case why he used Claude…

1290

Reposted by Christian Kündig

ren faire @lathrys.at · 03/10/2026

the problem with ai for the kind of person with my kind of brain is not that it offloads hard cognitive work, per se, but rather it enables a certain disposition to run 24/7 best summarized as:

330239

Reposted by Christian Kündig

Shodan Safari @shodansafari.lol · 04/10/2026

ASN: AS4134 Location: Shanghai, CN Added: 2026-10-02T09:54 #shodansafari #infosec

0227

Reposted by Christian Kündig

Duncan Hothersall @dhothersall.bsky.social · 04/10/2026

Completely agree with your assessment of Carlson, but Rees-Mogg isn't an aristocrat, and isn't particularly well respected as a debater. The real hit here isn't that Carlson was schooled by a pro, it's that he was schooled by a second-rate failed politician.

1661

Reposted by Christian Kündig

tweety fish oh wow like a chimera or something spoooky @sifu.tweety.fish · 03/10/2026

the funny thing about this (ugh I gotta finish the newsletter) is that a LLM isn't much of a moral agent a priori so you can trace anthropic's otherwise baffling product and development decisions as them _trying to make a LLM a moral agent_ so then they can then make it morally good

3172

Reposted by Christian Kündig

Michael Caley @michaelcaley.bsky.social · 03/10/2026

this is by far the best "I quit my AI job over safety concerns" essay yet because it is the first one that is able to talk about safety concerns in practical terms and that understands "safety" as a topic with extensive literature in analogous fields www.theatlantic.com/technology/2...

Two changes are urgently needed. First: AI companies need to rely more on the safety expertise that already exists in other fields. And second, before we create systems significantly more capable than the ones we have today, we need new science to ensure that more capable models (and their successors) will make safe choices when we aren’t looking.

Given today’s risks, frontier labs need to run like nuclear-power plants or busy airports, with layers of redundancy and careful, time-consuming planning, so that the occasional and inevitable human error does not open a door to disaster. Right now, AI companies don’t know how—but other people do. In a nuclear-power plant, the technical systems and rules people follow are set up so that, if equipment breaks or someone pushes the wrong button, we still won’t risk a meltdown. OpenAI and other labs are growing and deploying frontier AI with far less redundancy and rigor than this, even though the harm from an irreversible loss of control would be much greater than the harm from any single meltdown. Even short of a full loss of control, we could see autonomous swarms of AI agents that act without human permission. Imagine “rogue” agents that work like teams of hackers (for example, holding hospital computer systems for ransom) but never need to sleep.

20735177

Reposted by Christian Kündig

Stuart Gray @sgray.bsky.social · 03/10/2026

As an EU developed model, I so wanted to get excited by this, but the benchmark comparisons are against dated 30-35b models. I’d be shocked if a new 78b model didn’t beat them. That it only just beats them while using twice the resources isn’t a great endorsement.

2181

Reposted by Christian Kündig

worm @rfksbrainworm.bsky.social · 02/10/2026

If you ever wondered why visible light is those particular wavelengths, it's because life evolved to see in the narrow range where water is transparent

Diagram of absorption of emf by water. There's a gap right that dips in the 500 nm range

10161141709