Vanessa Loza

[Tesla @Tesla]

Roadster https://t.co/G0jU4mWfid

[Víctor Quispe Perca @EstadisticoBo]

Ninguno disimula. Ya no hay dignidad en esta profesión.

[Claudia Soruco 🇧🇴 @ClauSorucopress]

Cómo te vas a desesperar por tomarte una foto con la hermana de un NARCO?
Te hicieron caer de la cuna de wawa?🤦

[FIFA World Cup @FIFAWorldCup]

⏳ 19.07.2026 ⏳

#FIFAWorldCup

[Caleb Jordan Schulz @Based_Jedi]

My nephew and I are watching the Colombia v Ghana match and I introduced him to the greatest saved goal of all time by René Higuita in 1995 ⚽️

[Hanako @hanakoxbt]

KARPATHY JUST KILLED THE PROMPT ERA WITH A SINGLE DOCUMENT

prompts are easy. loops are hard. and writing fifty prompts a day is the work nobody does twice.

he shifts the burden to the harness.

you define the contract once. the model writes, reviews, restarts, and reconciles. you keep judgment. it keeps the loop.

the throughline is the same in every rule: the human owns the spec and the boundary. the model owns the execution and the bookkeeping.

planner never touches code. generator never grades itself. state lives on disk, not in context.

9 rules. start with one feature, not ten. most people are still typing prompts. this turns Claude into an agent that finishes the job on its own.

here is the official document from Karpathy explaining the architecture

[Matthew Kobach @mkobach]

I now demand all brackets to be designed in a circle

This is beautiful

[FIFA World Cup @FIFAWorldCup]

A last-minute winner sends Canada to the Round of 16! 👊

#FIFAWorldCup

[Codez @0xCodez]

A senior Anthropic engineer just dropped 11-page PDF on "Loop Engineering" for agentic systems.

The shift: you stop prompting the agent. You build the system that prompts it instead.

Schedule → Discover → Build → Verify → Repeat

Every loop runs one turn, five moves:

• Discovery: it finds its own work - failing CI, open issues, recent commits - instead of being handed a list.

• Handoff: each task gets an isolated git worktree so parallel agents don't collide.

• Verification: a second agent, told to assume the code is broken, reviews the first. The "thing that can say no."

• Persistence: results get written to disk, never left in a context window that gets flushed.

• Scheduling: an automation wakes it on a timer. That's what makes it a loop.

The key insight: an agent grading its own work always praises it.

This 11-page PDF changed how I'm building agentic systems today.

Read it now, then explore the article below.

[Chris Hadfield @Cmdr_Hadfield]

Venus and Jupiter over Toronto.
The reflected light from Venus (the brighter of the 2) took 10 minutes to get here. From Jupiter - 50 minutes.
A rare chance to visualise our turning Earth and ever-moving solar system above the iconic city skyline.
Gorgeous timelapse by friend Andrew Yee, from Ward's Island.
#Toronto #FIFAWorldCup

[MichiganAI @michigan_AI]

Thrilled to congratulate Michigan AI Director Prof. @radamihalcea for being elected to American Academy of Arts and Sciences, one of the most prestigious honors a U.S. scholar can receive! 👏🥳
https://t.co/cDHD4ha7SI

[Jota @jota_snchez]

Robert Sapolsky es un neurocientífico de Stanford que demostró que el estrés crónico es el asesino silencioso que los médicos ignoran.

Reveló 10 hábitos que haces todos los días y que te quitan años de vida.

1) Repasar conversaciones en tu cabeza

[Bhavika Agarwal @Ai_Tech_tool]

Instead of watching an hour of Netflix, watch this 2 hour hour Stanford lecture will teach you more about how LLMs like ChatGPT and Claude are built than most people working at top AI companies learn in their entire careers.

[MichiganAI @michigan_AI]

Big congratulations to Prof. @radamihalcea for this well-deserved recognition of her impactful work in #NLP, cross-cultural #AI, and social impact!🎉

[MichiganAI @michigan_AI]

When #AI Meets #Poverty:
An interdisciplinary group of @UMich researchers map out where AI can address poverty’s toughest barriers, and where it risks making them worse.

🔗https://t.co/PgZ6e44jzu

[Elias @iam_elias1]

You think using AI makes you smarter.
UCL and MIT just proved it makes you more biased.

Not slightly. Not in edge cases. Measurably, consistently, across every domain they tested — perception, emotion, social judgment, professional decisions. 1,401 participants. Multiple experiments. Published in Nature Human Behaviour.

The finding is called a human-AI feedback loop. And once you understand how it works, you cannot unsee it.

Here is the mechanism.
AI learns from human-generated data. Human data contains biases — subtle ones, baked into how we label images, write descriptions, and make decisions. The AI absorbs those biases during training. Then it amplifies them, because amplifying the pattern improves its prediction accuracy.

Now you interact with that AI. You see its output. You trust it — more than you would trust another human, because AI feels objective, authoritative, less noisy. So you adjust your own judgment toward what the AI showed you.

Then your biased judgment becomes training data for the next version.
Repeat.

The researchers called it a snowball effect. Small errors in the original dataset become amplified by the AI, which then increases the bias of the person using it, which feeds back into the next model. Round after round. Each iteration more distorted than the last.
Here is one experiment that makes this concrete.

They showed participants an AI image generator — Stable Diffusion — and asked it to create images of financial managers. It generated images of white men 85% of the time, dramatically out of proportion with reality.
After viewing those AI-generated images, participants became significantly more likely to associate the role of financial manager with white men.

The AI showed them a biased picture of the world. They internalized it. Their judgment shifted.
And they had no idea it was happening.
The researchers measured whether participants were aware of the AI's influence on their decisions. They were not. The bias transfer was largely invisible to the people experiencing it. Which means the most dangerous part of this mechanism is not that it happens.
It is that it happens silently.
Here is the detail that makes this even more alarming.

When participants were falsely told they were interacting with a human — while actually interacting with AI — the bias amplification was smaller. People were less susceptible to the AI's influence when they thought it was a person.
They trusted the AI more because it was an AI.
The thing they believed was objective and neutral was the very thing rewiring their judgment without their knowledge.
The researchers made one thing clear: this is not inevitable. Interacting with accurate, unbiased AI can actually improve human judgment over time. The feedback loop runs in both directions. If the AI is right, you get better. If the AI is wrong, you get worse.
Which means the entire trajectory of human judgment — collectively, across society — now depends on whether the AI companies building these systems take bias seriously.
And right now, most of them are racing for capability.
Not accuracy.

Source: Glickman & Sharot · UCL & MIT · Nature Human Behaviour 2025 · https://t.co/lcwCT2BWSn

[TFTC @TFTC21]

World, Sam Altman's digital identity project, just unveiled World ID 4.0, what the company calls "full-stack proof of human" infrastructure.

The partner list: Tinder, Zoom, DocuSign, Shopify, Okta, AWS, and Vercel.

Altman opened by saying we're heading to a world where AI generates more content than humans. Pantera Capital says we've already crossed that threshold. World's answer is an iris-scanning device called the Orb that creates a unique cryptographic ID proving you're a real person. 18 million people across 160 countries have already verified.

Tinder is rolling out "verified human" badges in the U.S. after a Japan pilot. Zoom built a feature called "Deep Face" that verifies the person on a video call isn't a deepfake. DocuSign is adding proof-of-human checks to digital signatures. Shopify is enabling verified-human commerce.

The most significant announcement is AgentKit, infrastructure that lets AI agents carry cryptographic proof they're acting on behalf of a verified human. Okta built an agent delegation system on top of it.

The problem World is solving is real. The question is whether a centralized iris-scanning identity layer controlled by the same person whose company helped create the problem is the right answer.

Altman is the CEO of OpenAI. He built the flood. Now he's selling the ark.

[Spencer A. Klavan @SpencerKlavan]

“You're on a spaceship called Earth that was created to give us a place to live in the universe, in the cosmos.”

Space evangelism is the energy I am here for in 2026. This is Victor Glover’s incredible Easter Message:

[📸🔭Brandon Berkoff🚀✨ @spacebrandonb]

I’m not asking you, I’m telling you.

Take a moment and listen to this 81 second response from Victor Glover after being asked if he had any thoughts leading up to Easter.

I don’t quite think it can be overstated how perfect this crew is for the job.