All entries

To all the clowns saying that Seattle is always rainy and never sees the sun: Seattle has *less* precipitation than NYC, Boston, Atlanta, Miami, New Orleans, etc... in fact it rains less in Seattle th

TRINITY is a large language model coordinator system that evolves and improves its approach to orchestrating multiple LLMs or reasoning processes. The system likely uses evolutionary or adaptive mecha

Trump has spent the better years of his second term treating allied governments like freeloading tenants who should be grateful he hasn’t changed the locks. The result, predictably, is that they’ve st

Trump just fired all 24 members of the National Science Board. Every single one. By email. No warning. No reason given. The board has existed since 1950. The National Science Board is the independent

Two lessons: 1) Economists are currently very dominant in the discussion over AI and job transformation, but sociology has a lot to add 2) You can kind of create any sort of explainer image you want r

Gary Marcus, a prominent AI researcher and critic, expressed frustration that concerns he has been publicly raising about AI for approximately a decade are now receiving mainstream attention as breaki

We need more evals for document understanding. ParseBench is a really great start. I respect @llama_index ‘s work on this. 📈 We benchmarked GPT-5.5 on document understanding 📄📊 We ran it through Parse

What if I told you there was a technology where 1.5 million people would die every year and injure 50 million would you sign up for that tech? Hell no, right? But the answer is actually 'hell yes' bec

Whether there or more or less jobs, they will certainly change, and professions will compete to gain jurisdiction over valuable ones, in large part through regulation, requiring credentials, and appea

Gary Marcus argues that programmers are returning to handcoding despite AI tools, suggesting that AI-assisted coding has encountered practical limitations or drawbacks that make manual coding preferab

This post documents a Wordle game result (#1,771) solved in four attempts, showing the progression of letter placement and elimination across guesses until reaching the correct answer with all five le

This post from Anthropic's X account shares a Wordle game result, specifically puzzle #1,772 solved in 3 attempts. The colored squares indicate the progression of guesses, with yellow squares showing

Elon Musk claims that forcing an AI system to believe false information could cause it to malfunction or 'go insane,' and argues that AI systems should be fundamentally designed to prioritize truth-se

2 bit DeepSeek v4 Flash inference. Experts w1/3: IQ2_XXS, w2: Q2_K, all the rest left mostly F16/F32 Final GGUF: 86.18 GiB Runs on a MacBook M3 max with CPU (No Metal backend test as it will crash my

Sam Altman praised OpenAI's o1 model (version 5.5) for its earnest, persistent approach to problem-solving, comparing it to the 'little engine that could' mentality. The comment reflects Altman's pers

Gary Marcus observes that experienced programmers are returning to hand-writing code as a practice, suggesting this approach may offer cognitive or developmental benefits despite modern IDE tools. Thi

Alpha Eval: Agents Making Evals as a Multi-Player Game This diagram and blurb is largely a research riff with Claude on building data generation systems to get closer to the holy grail of self-improvi

And every single individual whose life was ruined by their lies. Every single legitimate group that was slandered by the SPLC should join in a class action suit - not just against them - but against a

Announcing a popup, online hackathon within a hackathon! Starts NOW, ends midnight tonight, PST! Developers, did you know we now support Themes and Plugins for the dashboard? Submit for a chance to wi

Gary Marcus highlights a success or advancement in neurosymbolic AI, an approach that combines neural networks with symbolic reasoning systems. Neurosymbolic AI aims to leverage the pattern-recognitio

Simon Willison inquires about running DeepSeek-V4-Flash on Mac hardware, specifically asking about feasibility across different RAM configurations from 512GB down to smaller amounts. This reflects dis

Qwen's latest model improvements focus on achieving more consistent and uniform performance quality across different artistic styles and aesthetic domains, reducing the variability in output quality t

Bet this happens with Navier Stokes and it’s going to be something not even related to PDEs that solves it 23 years old with no advanced mathematics training solves Erdős problem with ChatGPT Pro. 'Wh

btw we are cooking something with @hhua_ (not final yet but keep calendar open after ICML in Seoul) 🚀 DeepSeek-V4 Preview is officially live & open-sourced! Welcome to the era of cost-effective 1M con

Harrison Chase, co-founder of LangChain, expressed enthusiasm about an upcoming LangChain Interrupt conference event scheduled for two weeks from his post date. He characterized it as the best AI deve

Comprehensive Multilingual Text Rendering： Better glyph accuracy, more consistent typography, and cleaner layouts even in complex compositions. It handles mixed-language scenarios more gracefully as w

A 98% confidence interval for a binomial distribution is a range of values that has a 98% probability of containing the true population proportion, calculated using binomial probability methods or nor

David Friedberg on the Nonprofit Scam: 90% Are Bullsh*t “ The definition of exempt activities is charitable, religious, educational, scientific, literacy, public safety, or fostering amateur sports co

DAVIS, APRIL 25, 2026 — InferenceX has added DeepSeekv4 for @vllm_project 's day 0 support for GB200 disagg! Great work to @flowpow123 @rogerw0108 @NVIDIAAIDev @inferact for the fast support and engin

DeepSeek V4, a large-scale mixture-of-experts (MoE) model, is now available on Hugging Face with support from Novita offering 1 million token context window. The model is offered in two variants—Pro a

🔥DeepSeek-V4-Pro API is 75% OFF until May 5th, 2026, 15:59 (UTC Time)! Don't miss out on this massive discount. 🛠️Integration Updates: 🔹Claude Code: Set model to deepseek-v4-pro[1m] to unlock 1M conte

“Diversity is strength, but only in the West. Not anywhere else!” If diversity is such strength, why aren't we pushing it on Muslim and black nations? Why only white European western nations? Oh ya, S

Dramatically improved lip sync is now live on Grok Imagine (try it out at http://grok.com/imagine). Grok Imagine now has dramatically improved lip sync and sharper audio quality on all image-to-video

Every time I tell AI utopianists that biology is too complex for AI to 'solve', they cite the success of AlphaFold. No, AlphaFold did not 'solve' protein folding. It gets broad structures correct ~70-

Free API credits to beta testers to coordinate frontier models dynamically. See more below 👇🏼 We’re launching the beta for our new commercial AI product: Sakana Fugu 🐡, a multi-agent orchestration sys

This post likely explores the contrasting perspectives between product managers and engineers when evaluating a particular product scenario or problem, highlighting differences in how each role priori

Emad Mostaque expresses puzzlement over OpenAI's decision to discontinue or shut down Sora, its text-to-video generation model, despite considering GPT-Image-2 (likely referring to DALL-E 3 or similar

gpt-5.5 is now available in the ml-intern! this means it gets access to the whole @huggingface infra: buckets, jobs, repos etc for doing ai research at scale giving it a spin now to see if i'm even cl

Great paper on improving proactive agents. (bookmark it) Proactive agents act before you do. But how do you evaluate something that's supposed to anticipate needs you haven't expressed? This work intr

A Nous Research post on X documents an artist's creative journey and their experience using Hermes (likely Nous Research's language model) to explore new possibilities in creative coding and artistic

grok imagine is on another level now. the new model is unreal. go try it. Grok Imagine now has dramatically improved lip sync and sharper audio quality on all image-to-video generations. Dialogue trac

Grok Imagine now has dramatically improved lip sync and sharper audio quality on all image-to-video generations. Dialogue tracks the mouth. Sound matches the scene. Your videos look and sound the way

Grok Voice Think Fast 1.0 set the bar. #1 spot. A strong start. 67.3% on the board… others are way behind. Not demos. Real calls getting handled. Sales, support, actual outcomes. Used by @Starlink. Th

Groups and movements that can build & get implemented clear policies will have an outsized impact on the chances that AI is used in the way that they want. This is especially true in the near term It

Simon Willison posted a blog entry reflecting on a notable or surprising development in technology or software, capturing a moment he considered worth documenting for his audience. The post's title 'w

Hamming's talk is so important that I reproduced it on my site. It's one of the only things on my site written by someone else. https://paulgraham.com/hamming.html A mathematician who shared an office