Bernie Buss

Computer Science meets Biology @ETH_en. Focus on personalized rejuvenation therapies. Trading and investing to finance research.

[Bernie Buss @BernieBuss]

@AMCScam @Growthaniel @YoDooley thanks for keeping it real 🫡

[Claude @claudeai]

Introducing Claude Opus 4.8: it builds on Opus 4.7 with sharper judgment, more honesty about its own progress, and the ability to work independently for longer than its predecessors.

Available today at the same price.

[Bernie Buss @BernieBuss]

@_itsjustshubh @leerob at the moment Cursor is winning on speed and quality as a combined metric.

[Bernie Buss @BernieBuss]

@_itsjustshubh @leerob claude code has the most professional / geeky style of chat output: good for analysing and reviewing, displaying content with tables. but its intelligence is lacking compared to GPT5.5, even though it looks simpler in chat output. antigravity is fast and barebones but limit sucks

[Bernie Buss @BernieBuss]

@leerob what I am wary of is that the new privacy option requires you to upload your code repository for potential cloud agentic use. sure github has it too, but this is an unnecessary data grab for no immediate benefit to me as developer.

[Science girl @sciencegirl]

A breakthrough stem-cell treatment could change the future of hearing restoration by targeting the underlying nerve damage responsible for permanent hearing loss.

Biotechnology company Rinri Therapeutics has developed an experimental therapy called Rincell-1, which uses specially grown stem-cell-derived auditory nerve precursor cells to repair damaged connections between the inner ear and the brain.

Current treatments such as hearing aids and cochlear implants mainly improve sound perception or bypass damaged structures, but they do not repair the lost nerve cells themselves. Rincell-1 aims to rebuild these critical auditory pathways, potentially offering a long-term restorative approach rather than symptom management.

In preclinical animal studies, the therapy has shown encouraging results, prompting regulators to approve the next stage of research. The UK’s Medicines and Healthcare products Regulatory Agency (MHRA) has now authorised the first human clinical trial of the treatment, with patient testing expected to begin in May 2026.

Researchers will investigate how safely the transplanted cells integrate into the ear and whether they can successfully re-establish hearing-related nerve signals. Although the treatment is still experimental, scientists believe it could mark a major step toward regenerative therapies capable of reversing certain forms of permanent hearing loss in the future.

Source: Rinri Therapeutics (2026). MHRA approval granted for first-in-human clinical trial of Rincell-1 (NCT07032038). Medicines and Healthcare products Regulatory Agency.

[precis0x @precisox]

Un youtuber brasileño le acaba de clavar un puñal a la suscripción de Photoshop.

Se llama PhotoGIMP: un parche gratuito (GPL-3.0) que convierte GIMP en una copia casi idéntica de Photoshop.

Misma interfaz, mismos paneles, mismos atajos de teclado y muchísimo más espacio para tu lienzo. Tus manos ya saben usarlo sin aprender nada nuevo.

¿Por qué está explotando?
- $0 en vez de $276 al año
- Sin cuenta Adobe ni login
- Todo se guarda en tu PC (nada en la nube)
- Compatible con Windows, Mac y Linux
- Se desinstala borrando una carpeta (sin rastro)

Instalación ridículamente fácil: copias 9 archivos y listo. +8.8k estrellas en GitHub y traducciones de la comunidad.

Uso personal y comercial 100% permitido.

[PeterSweden @PeterSweden7]

If your country requires digital ID verification to use social media, then you don't live in a free country anymore.

If your country want to regulate or ban VPNs, you don't live in a free country anymore.

[Massimo @Rainmaker1973]

A groundbreaking medical study revealed that aging may be transmitted through the bloodstream via a protein called HMGB1. When researchers blocked this protein in animal tests, they observed remarkable results: damaged tissues began to repair themselves, and some age-related decline was reversed.

This discovery suggests that aging is not simply an inevitable process of cell breakdown but may be influenced by specific molecular signals. If these signals can be controlled, aging could be slowed — or even partially reversed.

Such treatments could revolutionize medicine, offering new ways to fight diseases like Alzheimer’s, arthritis, and organ failure, all of which are tied to aging. However, researchers caution that human trials are still far away.

The study fuels hope that one day aging itself might be treated as a medical condition, reshaping human health and longevity.

[Krinza @krinza]

in the next 3 years, every major AI lab will spin up its own bio arm and in-house wet labs.

biology is the next big bet in AI after code

pay attention.

[Massimo @Rainmaker1973]

A rare genetic mutation lets some people function on just 4 hours of sleep each night.

Researchers have discovered a rare mutation in the SIK3 gene that enables a select group of people—often called “natural short sleepers”—to thrive on only four to six hours of sleep per night without any cognitive or health drawbacks.

Unlike most individuals who skimp on sleep and suffer the consequences, these people complete all critical brain maintenance and restoration processes in far less time, waking up fully refreshed and sharp.

The SIK3 variant is the latest addition to a small family of known short-sleep genes (including DEC2 and ADRB1), revealing how certain genetic changes can dramatically compress the time needed for high-quality rest while preserving its benefits.

These findings challenge the widespread assumption that everyone requires eight hours of sleep and highlight the profound role genetics plays in sleep needs.

Beyond explaining natural variation, the discovery raises exciting possibilities: future treatments could mimic this efficient sleep phenotype to help millions who battle insomnia, chronic fatigue, or demanding schedules that leave little room for conventional rest.

[H. Chen, Y. Xing, C. Wan, Z. Zhang, Z. Shi, Y. Liang, C. Jin, Y. Chen, X. Zhou, J. Xu, L.J. Ptáček, Y. Fu, & G. Shi, "The SIK3-N783Y mutation is associated with the human natural short sleep trait", Proc. Natl. Acad. Sci. U.S.A. 122 (19) e2500356122]

[Bernie Buss @BernieBuss]

@BoringBiz_ most lack creativity of what to do with increased productivity, especially in a corporate environment with no owners and no one cares. thats why I am doing my own things now and it is a bliss

[Cyrus Razavi @cyrus_razavi]

I 2005 begyndte Danmark at blokere hjemmesider via DNS.

Officielt for at beskytte børn mod seksuelt overgrebsmateriale.

Året efter stod musikindustrien klar.

Samme værktøj.
Ny begrundelse.

Nu handlede det pludselig om piratkopiering, ophavsret og en industri, der ikke havde fulgt med tiden.

Det er sådan kontrol ofte starter.

Først vælger man den sag, ingen tør sige imod.

Børn.
Overgreb.
Sikkerhed.

Så viser man, at teknikken virker.

Ikke ved at fjerne indholdet.
Ikke ved at redde ofrene.

Men ved at skjule det for den brede befolkning.

Indholdet lå der stadig.
Bagmændene var der stadig.
Politiet havde allerede værktøjer.

Men staten og industrien havde fået noget andet.

En infrastruktur til blokering.

Nu ser vi samme mønster igen med EU.

Aldersverifikation.
Digital identitet.
Online adgang, betinget af godkendelse.

Igen hedder det beskyttelse af børn.

Men det er ikke EU’s opgave at opdrage vores børn på internettet.

Det er forældrenes opgave.

Myndighedernes opgave er at skabe et samfund, hvor vores børn kan færdes trygt på gader, i skoler, i parker og i deres eget lokalområde.

Når staten ikke kan løse den virkelige utryghed, flytter den kontrollen ind i det digitale rum.

Og når folk begynder at omgå aldersverifikation med VPN, hvad bliver næste skridt så?

Blokering af VPN?
Registrering af brugere?
Begrænset internetadgang efter statens godkendelse?

På et tidspunkt må man spørge:

Hvor langt skal Europa gå, før vi indrømmer, at vi er ved at kopiere værktøjer fra regimer, vi selv siger vi foragter?

Jeg var engang stor tilhænger af EU.

Fordi EU gjorde livet lettere.

Fri bevægelighed.
Fri handel.
Mindre bøvl.
Mere åbenhed.

Men det EU, der skulle gøre friheden større, er begyndt at gøre friheden betinget af kontrol.

Og så står vi tilbage med det egentlige spørgsmål:

Hvornår blev Europa så bange for frihed, at det begyndte at umyndiggøre og mistænke sine egne borgere?

#dkpol #dkmedier

[👽🛹🚀 X-SCHLATTI🌎 👽🛹🚀 @Schlatti2]

Today I got the feedback from Federal Roads Office FEDRO
(ASTRA) in Switzerland regarding Tesla FSD.

@elonmusk They need support from Tesla

[Artur Nadolny @ArturNadol7566]

THEY BLACKLISTED THE ENGINEER BEFORE THE BODY WAS COLD

Hazar Denli is a chassis engineer. His job was to make sure cars don't kill people. He was very good at it.

Working for Tata Technologies (@TataTech), he led the engineering team on VinFast's suspension and chassis systems. During testing he found the front strut-to-knuckle connection was loosening. Suspension parts were snapping off after 15,000 miles. They were supposed to last 93,000.

He told management. Nothing happened. VinFast had a Nasdaq IPO coming and didn't want to delay production.
So Denli resigned and moved to a new job at Jaguar Land Rover (@JLR_News).

Then in April 2024, a family of four died in California when their VinFast VF8 lost control, veered off the road and caught fire.

Denli posted on Reddit. He said he wouldn't get into a VinFast himself. He said he wouldn't let his family near one.
VinFast tracked down the anonymous post.

Tata Technologies HR director Patrick Flood then contacted JLR HR director Dave Williams and asked for Denli to be dismissed. The concern, in Flood's words: if he's done this now, he could do the same at JLR. Denli was fired the same day. Then blacklisted on industry recruitment platform Magnit so his future job applications would be auto-rejected.

@BBCNews obtained the internal emails proving all of it.
The NHTSA has since launched an investigation into the VinFast VF8. There are now 28 safety complaints on file.

Denli has filed disclosures with the SEC and NHTSA. He is taking JLR to an employment tribunal.

A family is dead. An engineer tried to stop it. He lost his job, his career platform access and his industry future. The companies involved declined to comment.

Nobody went to prison. Nobody recalled anything. Everyone kept their jobs except the one man doing his.

Sources: @WB_UK @BBCNews @cleantechnica
@carscoops

[Massimo @Rainmaker1973]

Brazilian scientist Tatiana Sampaio discovers a protein, Polylaminin, that can regenerate spinal cord.

This substance aims to create a supportive matrix that encourages damaged neurons to regenerate connections, potentially restoring motor function after severe injuries.

[Bernie Buss @BernieBuss]

@sweatystartup microplastics is the new roman lead

[How To AI @HowToAI_]

MIT proved every major AI model is secretly converging on the same "brain."

It’s called the “platonic representation hypothesis,” and it’s one of the most mind-blowing papers you’ll ever read.

You train a vision model purely on images. You train a language model purely on text.

They use completely different architectures. They process completely different data. They should have completely different "brains."

But as these models scale up, something impossible is happening.

When researchers measure how they organize information, the mathematical geometry is identical.

A model that only "sees" images and a model that only "reads" text are measuring the distance between concepts in the exact same way.

The models are converging.

The researchers named this after Plato’s Allegory of the Cave.

Plato believed that everything we experience is just a shadow of a deeper, hidden, perfect reality.

The paper argues that AI models are doing the exact same thing.

They are looking at the different "shadows" of human data, text, images, audio. And they are independently discovering the exact same underlying structure of the universe to make sense of it.

It doesn't matter what company built the AI.

It doesn't matter what data it was trained on.

As models get larger, they stop memorizing their specific tasks. They are forced to build a statistical model of reality itself.

And there is only one reality to map.

2024, Arxiv