At long last. This is great

All AI-powered code editors, maybe except Github Copilot.

Letter of the day (in the Times)

🔥🔥🔥 Wout van Aert DROPS Tadej Pogačar in the last climb up Montmartre ! 🔥🔥🔥

#TDF2025 | #TDFLive

Surely I can't be the only one who thinks of them like this, given how attention works by considering the entire context? And just the practical experience of long discursive conversations going completely to hell.

Prompt summarize Attachments image/jpeg: /var/folders/x6/31xf1vxj0nn9mxqq8z0mmcfw0000gn/T/llm_pdf_to_images_rw_qkmgz/page_001.jpg image/jpeg: /var/folders/x6/31xf1vxj0nn9mxqq8z0mmcfw0000gn/T/llm_pdf_to_images_rw_qkmgz/page_002.jpg Response This document outlines various types of political regimes and their characteristics: Democracy: A political regime where the government is accountable to the people and power is exercised directly or through representatives. It operates under guaranteed civil liberties, political equality, and periodic, competitive elections. Liberal or Representative Democracy: Features popular support and limitations on state power, with a functioning constitution, competitive elections, rule of law, and protections for civil liberties. It requires institutions like legislatures, courts, and political parties. Illiberal Democracy: Elections occur, but civil liberties and political rights are often limited, with the possibility of electoral processes manipulated by elites or authoritarian tendencies. Authoritarianism/Dictatorship: Characterized by the concentration of power in one or a few leaders, limited political pluralism, lack of democratic elections, and restricted civil liberties.

I built a new LLM plugin that can turn a PDF into an image-per-page for feeding into vision models, and in testing it found that GPT-4.1 mini hallucinates WILDLY if you feed it a blank white rectangle followed by a blank black rectangle simonwillison.net/2025/May/18/llm-pdf-to-i...

I heard this described as Dubai Final Boss…

Lost in AGI dread and gleeful anticipation is the fact that, while the technologies driving the discussion around AGI are novel and new, the stories we're hearing about them are not. The AGI story today's elites embrace is a retelling of the AI story embraced by the elites of yesteryear and the automation story of their predecessors. It's fair to say we're in uncharted technological territory, but that is also a truism. Where else would we be? As workers or as politicians, regulators, industrialists, or members of the curious general public, we have all previously confronted technological breakthroughs and moments of political and economic uncertainty. All those moments created opportunities for narrative inver tion, too. In its more cynical deployments, AGI, like Al before it, is a marketing story masquerading as a descriptive term; it's a bid to rebrand - and narrow down and direct - our collective idea of the future.

AGI, like G-less Al, automation, and even mechanization, are indeed stories, but they're also sequels: This time, the technology isn't just inconceivable and inevitable; it's anthropomorphized and given a will of its own. If mechanization conjured images of factories, automation conjured images of factories without people, and AI conjured humanoid machine assistants, AGI and ASI conjure an economy, and a wider world, in which humans are either made limitlessly rich and powerful by superhuman machines or dominated and subjugated (or perhaps even killed) by them (Indu trial Revolution 3: The Robot Awakens). In imagining centralized machine authoritarianism in the future, AGI creates a sort of authoritarian, exclusionary discourse now. A narrative emerges in which the decisions of AGI stakeholders - Al firms, their investors, and maybe a few government leaders - are all that matter. The rest of us inhabit the roles of subject and audience but not author.

AGI is a story, not a technology nymag.com/intelligence...

my hope is that AI can empower the Dumbest, Least talented slobs i know to replace everything i ever loved with One Million Years of Content

Guy behind me at this coffee shop just typed “had a dream my mom died what does it mean” into ChatGPT

1.5 yrs ago, we set out to answer a seemingly simple question: what are we *actually* getting out of RL in fine-tuning? I'm thrilled to share a pearl we found on the deepest dive of my PhD: the value of RL in RLHF seems to come from *generation-verification gaps*. Get ready to 🤿:

This is what they mean by "feeling the AGI"

Tweet: "Have you said thank you once?" -@VP JD Vance - pic of jd Vance looking oddly symmetrical. Why are his pants so tight?

I can't stop laughing at this man I'm sorry. This is the kind of shit the biggest loser in the world would say to their girlfriend about buying her flowers on Valentine's Day once, and he's doing so in an argument that started because he called her a bitch

An engineer once explained to me that an efficient system always has need available when a resource becomes free, and an effective system always has a resource available when a need arises. Mind was blown.

NEW: updated long-run gap in voting between young men and women in Germany:

Gender divide continues to widen, but contrary to what is often assumed, young men continue to vote roughly in line with the overall population, while young women have swung sharply left.

www.ft.com/content/29fd...

This weekend I've decided to finally read Balaji Srinivasan's terrible fucking book, "The Network State."

I've put this off for as long as I possibly could. Here goes.

thenetworkstate.com/book/tns.pdf

fr fr fr. So good! Love a surprise k. album, turns out

Gladiator II at IMAX last night was a perfect piece of escapism

Thinking I'd like to go to an AI conference in the heart of all the madness next year (SF). Anyone recommend good ones?

Admire the dedication. I used to be all in on vim but then eventually my config lapsed the current hotness (around LSPs) and I couldn't be bothered anymore and just gave in to vscode

I've long though that "taste" is key in almost all disciplines, and barely discussed, let alone taught.

I was just thinking about hacking together an ios app in swift but then realised using cursor might not be fun for it. This could be v interesting

Are We Too Impatient to Be Intelligent? - by Rory Sutherland - Behavioral Scientist There are things we need to deliberately and consciously slow down for our own sanity and for our own productivity. If we don't ask the question about what those things are, we might get things terrib...

This is as much a design problem as a tech problem. It's definitely making us lazy if just put suggestions in front of people and circumvent any thought, see: behavioralscientist.org/are-we-too-i...