esto es como yo cuando no me salía la palabra urólogo y dije el dentista del pito

If anyone cares at all, EV battery recycling is now hitting 99% recovery for important metals Please remind me again what the recycle rate for fossil fuels is interestingengineering.com/energy/recyc...

After seeing Meta cutting their metaverse group by 30%, having lost… $444 per second over the last five years: archive.is/2025.12.04-1... thought I’d look in on some McKinsey reports

🐣 It’s live! My little pet project is public now: Sky Wrapped. It's exactly what you're thinking: like Spotify Wrapped, but for Bluesky, and without ICE or any other shitty ads. 👉 skywrapped.app Discover your most interacted ones, your habits, your top emojis... I hope you like and share it! 🫶

Wrong framing: "Meta cutting Metaverse resources by 30%" Correct framing: "Meta still setting 70% of resources on fire, for some reason"

IBM’s Arvind Krishna doing the math that AI tends to struggle with This is from an interview with @reckless.bsky.social on @theverge.com

El mundo era mucho mejor cuando la IA te daba estos resultados

Oh shit waddup

To heal our divided nation I propose splitting Game of the Year awards into two categories: - Game of the Year That’s 3D and You Use a Sword and/or Gun to Explode a Guy - Game of the Year That’s Some Gay Bullshit Indie Meditation on Grief or Whatever

Vaya mierda de mundo que se nos está quedando. Encuentro esta maravilla en bsky, un cómic de un dibujante ESPAÑOL @javidecastro.bsky.social que ojo, SE PUEDE LEER DEL REVÉS 🤯

Vaya web guapa que se han currado en @financialtimes.com explicando cómo funcionan los LLMs ig.ft.com/generative-ai/

as seen in www.reddit.com/r/badUIbattl...

OpenAI needs to raise at least $207bn by 2030 so it can continue to lose money, HSBC estimates on.ft.com/4ij0yh8

Do you find yourself exhausted by having to create titles, puns, and names for things in your games? ⭐ Here is my small suite of websites that I find super useful when writing! ⭐ (disclaimer: I am pretty confident these are all genAI-free. I've certainly been using them since before genAI)

I misread this image so badly that I can only really hope to illustrate what I saw. ( Also a good excuse for an #art study )

Relying on ChatGPT to teach you about a topic leaves you with shallower knowledge than Googling and reading about it, according to new research that compared what more than 10,000 people knew after using one method or the other. Shared by @gizmodo.com: buff.ly/yAAHtHq

¿Se sigue diciendo “vamos a hacer una KDD para desvirtuzalizar a unos amigos”?

Escuchar lofi chunga tiene como side effect positivo encontrar maravishas gracias a los samples www.whosampled.com/Iamalex/Summ... → open.spotify.com/track/4Zxtyd...

Whoever says, ‘as an engineer I can talk to customers and define product strategy myself, so we don’t need product managers’ just became the product manager

¿Pero por qué me genero necesidades nuevas? 🥲 ¡Que alguien me pare!

Le dedico esta entrada a mi estantería kallax conexionaleatoria.ghost.io/la-lectora-p...

El contenido por el que pago Bluesky

nowadays every app wants to be a platform, and platform logic requires that they become the *only* platform. that's why they all start as Specific: The Tool That Does A Thing and five years later announce they're rebranding as Totalizr: We Manage Your Life

Todo lo del estatuto marco se resume a esta viñeta de Mafalda que no conocía y que me han pasado

todo era más divertido cuando le pedías a la IA un listado de banderas de países de Europa y te devolvía lo siguiente:

the password to the louvre surveillance server was "louvre" www.thesocialpost.it/2025/11/02/f...

We now have multiple visualizations of this phenomenon from various news organizations. Which one is the best?

The next time people complain that user research just slows things down, show them this towering monument of hubris. Quite staggering.

recently discovered that wikipedia volunteers have a hilariously high bar for what constitutes "unusual death"

Eso de Martín Fowler disfrazado del chaos monkey no me lo vi venir

The #1 criteria for my browser is that I trust it, and it’s secure. It’s logged into my email, Stripe account, stores my passwords and credit cards. A browser is useful if it does all these. These AI browsers are interesting but… with prompt injections, no way I trust them…

Todo bien con el nuevo Tahoe y las curvaturas 🥲

Un poco de supersmooth disco para animar la tarde soundcloud.com/slightlytran...

This photo accompanying a news story about the heist at the Louvre is perfection.

Cuando te traen la tapa de pulpo.

Cada vez que aparece el "Whats heavier, a kilogram of steel or a kilogram of feathers?" en alguna conversación es autoretweet 😂 www.youtube.com/watch?v=-fC2...

Remember this one? “Humans aren’t very efficient movers—until you put us on a bicycle, when we become some of the most energy-efficient land travelers in the animal kingdom.” Via Scientific American @sciam.bsky.social #CityMakingMath

Microsoft hates computers AND everyone who uses them, whereas Apple only hates computers, and Linux only hates the people who use them. In this essay I will

my latest investigation for @consumerreports.org is based on months of reporting and 60+ lab tests of leading protein supplements we found that most protein powders and shakes have more lead in one serving than our experts say is safe to have in a day (🧵) www.consumerreports.org/lead/protein...

We should be asking this c o n s t a n t l y.

“El futuro va a ser una pasada” El futuro:

Quicir, todos entendemos el sentido de tener una tele y radio públicas ¿No? ¿Por qué no una red social pública? A estas alturas del partido creo que tiene sentido. Podemos discutir todos los "cómos" , pero el por qué empieza a ser evidente.

Chatbots and AI slop are going to ruin countless brains. Yours doesn't have to be one of them. Study history. Read poetry. Get acquainted with Bach, Shakespeare, George Eliot. In a time when people are rapidly getting dumber, we need communities devoted to the life of the mind.

Let’s talk about AI art. theoatmeal.com/comics/ai_art

la newsletter de @spavel.bsky.social es canela fina, pero es que su último envío tiene un primer párrafo para enmarcar:

he puesto de moda en mi squad el "corrígeme si me equivoco" 😅