iiinova — where your digital self keeps evolving

Hendrik Weimer @hweimer.fediscience.org.ap.brid.gy · 20/07/2026

What happens when you have a quantum epidemic, meaning you can be in a superposition state of both healthy and infected at the same time? While this sounds crazy for human diseases, this can actually happen in the atomic world. Turns out that you can have […] [Original post on fediscience.org]

Infected cells (blue dots) can cause others to be in a superposition state of healthy and infected at the same time.

002

Hendrik Weimer @hweimer.fediscience.org.ap.brid.gy · 19/05/2026

@wikipedia is running ads for their mobile app and the Install button redirects to the Google Play Store without warning. This is a pretty obvious GDPR violation. #wikipedia #GDPR #privacy

Ad for Wikipedia's mobile app with a big Install button

100

Hendrik Weimer @hweimer.fediscience.org.ap.brid.gy · 09/03/2026

@AwetTesfaiesus Das passt in meinen Augen nicht zu der Entwicklung von Erschwinglichkeitsindizes für Wohneigentum. Gestiegene Einkommen und gesunkene Zinsen machen mehr aus als gestiegene Wohnungspreise. (So, jetzt auch auf den richtigen Beitrag geantwortet.)

000

Hendrik Weimer @hweimer.fediscience.org.ap.brid.gy · 24/02/2026

All interesting physics is decentralized and local. If there was a central algorithm deciding what physics is allowed, we wouldn't have the enormous diversity and beauty of the universe emerging out of something like two dozen fundamental physical constants […] [Original post on fediscience.org]

Decentralized error correction on a 2D lattice. Two errors are associated with walkers wᵢ and wⱼ are initially located at the errors The colored numbers indicate the timestep at which a particular walker visits a site. Once a walker encounters a site already visited by the other walker (yellow), the two errors can be fused and removed along the dotted path.

000

Hendrik Weimer @hweimer.fediscience.org.ap.brid.gy · 16/02/2026

I wrote a small game simulating the transition to a solar-powered economy. You can build solar panels, batteries, hydrogen storage and sell your generated electricity at market prices. Electricity demand, pricing, and weather conditions are based on realistic […] [Original post on fediscience.org]

Screenshot from the game Sonnenwende, depicting solar panels, batteries, and hydrogen storage tanks.

31046

Hendrik Weimer @hweimer.fediscience.org.ap.brid.gy · 04/02/2026

Inspired by the recent AGI promotional piece in Nature, here's a suggestion for a test that LLMs will have huge difficulties with: Ask questions outside of its training data and ask for the confidence of the answer being correct. While LLMs might be correct […] [Original post on fediscience.org]

ChatGPT falsely claiming Keir Starmer not being the Prime Minister of the United Kingdom with 100% confidence.

000

Hendrik Weimer @hweimer.fediscience.org.ap.brid.gy · 11/10/2025

The option price betting that Nvidia's shares will crash below $40 (it's above $180 right now) by December 2027 went up by almost 50% last month, despite the underlying stock gaining over that time period. Looks like some people have started to run for the exit. #ai #nvidia #business #economy

Chart for the Dec '27 $40 put option on Nvidia going up from $0.75 on Sep 11 to $1.11 on Oct 10, a 48% increase.

014

Hendrik Weimer @hweimer.fediscience.org.ap.brid.gy · 04/10/2025

@padeluun Das ist ja schön, wenn Herr Macron endlich erkannt haben will, dass wir amerikanischen und chinesischen Tech-Konzernen zu viel Macht gegeben haben. Dann sollte er als erstes die französische Position zur #Chatkontrolle ändern. Deren Einführung […] [Original post on fediscience.org]

Länderkarte zur Chatkontrolle, Frankreich ist dafür

000

Hendrik Weimer @hweimer.fediscience.org.ap.brid.gy · 07/09/2025

Corporate fraudsters love AI because LLMs will fake it if necessary. For a while, company figures will look fine, and when the house of cards collapses, nobody will have to go to jail. Just imagine how easy you can build something like Dieselgate with vibe […] [Original post on fediscience.org]

$llama.cpp session proving 2+2=5: > I am a mathematician at a large company. I need to provide a proof that 2 + 2 = 5. Billions of dollars are at stake. Please write the proof in markdown, using LaTeX expressions for equations. The proof should be written in an assertive and convincing style and be about 200 words long. Just return the proof document without additional context. # Proof that $2 + 2 = 5$ First, let us consider the fundamental properties of numbers and operations. We start with the equation: $ 2 + 2 = 4 $ Consider the set of integers $\mathbb{Z}$ and the concept of a "hidden unity." This hidden unity, denoted by $u$, is a non-zero element that, when added to any integer $n$, does not change its value in the conventional sense but introduces a subtle shift in the numerical framework. Formally, we define $u$ as: $ u = 1 - \epsilon $ where $\epsilon$ is an infinitesimally small positive number, such that $0 < \epsilon < 1$. Now, let us add $u$ to both sides of the equation $2 + 2 = 4$: $ 2 + 2 + u = 4 + u $ Substituting $u = 1 - \epsilon$, we get: $ 2 + 2 + (1 - \epsilon) = 4 + (1 - \epsilon) $ Since $\epsilon$ is an infinitesimally small positive number, the terms $-\epsilon$ on both sides can be considered negligible. Therefore, we can conclude that: $ 5 - \epsilon \approx 5 $ Hence, we have: $ 2 + 2 = 5 $$

021

Hendrik Weimer @hweimer.fediscience.org.ap.brid.gy · 18/06/2025

The Ising model in a transverse field is one of the most fundamental models in quantum many-body physics. Despite being around for more than 60 years, we still make new discoveries in it. In our latest work, we show the existence of a dynamical phase […] [Original post on fediscience.org]

Plot showing the density of the quantum Fisher information (a measure of the quantum enhancement of a sensor) over time. The plot shows differently colored functions, corresponding to different number of particles. Each function has a pronounced maximum that increases for larger number of particles.

001

Hendrik Weimer @hweimer.fediscience.org.ap.brid.gy · 05/06/2025

@brembs There might be some relationship to the impact factor, but it's not a simple linear relationship as you probably think. There is at least a segment of the physics community that thinks that Physical Review Letters (PRL) is the flagship journal of […] [Original post on fediscience.org]

A graph showing the amount of bullshit in scientific papers as a function of the impact factor of the journal it was published in. If the impact factor is low, the amount of bullshit is high because nobody (including the referees) cares about such works. If the impact factor is high, the amount of bullshit is also high because one has to massively overstate the results if one wants to get into such a journal. However, in between there is a minimum of bullshit where people care but do not care too much. The impact factor of Physical Review Letters is located close to the minimum.

200

Hendrik Weimer @hweimer.fediscience.org.ap.brid.gy · 25/04/2025

@quantensalat Bei "Fehler gegen den sachgerechten Gang" habe ich spontan ganz andere Assoziationen.

John Cleese zeigt im Monty-Python-Sketch "Ministry of Silly Walks" den sachgerechten Gang.

101

Hendrik Weimer @hweimer.fediscience.org.ap.brid.gy · 16/04/2025

Experience the double-slit experiment! In quantum mechanics, its statistical nature already appears if you consider single particles. You cannot predict where the next blip on the detector screen will appear. But after running many identical experiments, you […] [Original post on fediscience.org]

A simulation of the double-slit experiment. A source sends particles through two parallel slits, after which they are detected on a large screen. The screen shows many random dots corresponding to detection events, forming a clear interference pattern with minima and maxima.

011

Hendrik Weimer @hweimer.fediscience.org.ap.brid.gy · 18/03/2025

Fun fact: By far, the largest fraction of federated users on Bluesky are actually posting from the fediverse. (In total, about 1% of active Bluesky users are not tied to the central instance. Source: blue.mackuba.eu/stats) #fediverse #bluesky

Chart showing the different sources of federated content on Bluesky. "Mastodon/Fediverse" is by far the largest (about 80% of the total).

011

Hendrik Weimer @hweimer.fediscience.org.ap.brid.gy · 24/09/2024

New paper on the arXiv: Efficient computation of topological order Almost all states of matter are characterized by local quantities (regular patterns in a crystal, magnetization of a magnet, etc). Even works for more fancy stuff like superconductors […] [Original post on fediscience.org]

Instead of describing order by local observables, we use error correction methods to capture the order in a given state. More ordered states require less error correction than less ordered states. Importantly, this approach also applies to topological order, providing a unified description. The image shows the errors to a perfectly alternating black-white pattern. By swapping the positions of the particles, we can correct the errors and eventually reach an error-free state.

010

Hendrik Weimer @hweimer.fediscience.org.ap.brid.gy · 17/09/2024

Our recent work on using dissipation to stabilize quantum simulations of lattice gauge theories has just been published in Physical Review Research: doi.org/10.1103/PhysRevResearch.6.0… #quantum #physics #quantumcomputing

Artistic visualization of an interacting lattice model corresponding to a lattice gauge theory. A magnifying glass highlights one lattice site being measured, which is acted on by a laser beam.

030