571
OpenAI’s policies hinder reproducible research on language models
LLMs have become privately-controlled research infrastructure
Top posts
Every blog post that took off on Hacker News, ranked by the votes of its best submission. 6,324 posts.
572
The Matrix Calculus You Need for Deep Learning
Most of us last saw calculus in school, but derivatives are a critical part of machine learning, particularly deep neural networks, which are trained by optimizing a loss function. This article is an attempt to explain all the matrix calculus you need in order to understand the training of deep neural networks. We…
573
How to stop Claude from saying load-bearing
Hack the text output of Claude Code to make life a little bit sillier.
574
Grok 4.7
SpaceXSI's most powerful model for coding and knowledge work. Twice as fast, at half the price of comparable models.
575
Deep learning gets the glory, deep fact checking gets ignored
AI, science, education, & ethics
576
Claude Code uses Bun written in Rust now
In Rewriting Bun in Rust Jarred Sumner made the following claim: Claude Code v2.1.181 (released June 17th) and later use the Rust port of Bun. Startup got 10% faster on …
577
Llama.ttf: A font which is also an LLM
llama.ttf is a font file which is also a large language model and an inference engine for that model.
578
A guide to local coding models
What you need to know about local model tooling and the steps for setting one up yourself
579
Ollama releases Python and JavaScript Libraries
The initial versions of the Ollama Python and JavaScript libraries are now available, making it easy to integrate your Python or JavaScript, or Typescript app with Ollama in a few lines of code. Both libraries include all the features of the Ollama REST API, are familiar in design, and compatible with new and previous…
580
AlphaGenome Atlas: a high-resolution map of human DNA
We’re introducing AlphaGenome Atlas, a database predicting the effects of every possible single nucleotide variant in the human genome.
581
Disputing a Parking Fine with ChatGPT
A use case for ChatGPT
582
Nano Banana 2: Google's latest AI image generation model
Our latest image generation model offers advanced world knowledge, production-ready specs, subject consistency and more, all at Flash speed.
583
Fly.io has GPUs now
Documentation and guides from the team at Fly.io.
584
Native JSON Output from GPT-4
When integrating LLMs in your products, you often want to generate structured data, like JSONs. With the help of function calling (released June 13th 2023), this process has become much simpler! In this post I will explore the new API. Thanks for reading Simon’s Substack! Subscribe for free to receive new posts and…
585
Data Exfiltration from Slack AI via indirect prompt injection
Authors: PromptArmor
586
A one-year moratorium on police use of Rekognition
We’re implementing a one-year moratorium on police use of Amazon’s facial recognition technology. We will continue to allow organizations like Thorn, the International Center for Missing and Exploited Children, and Marinus Analytics to use Amazon Rekognition to help rescue human trafficking victims…
587
DeepMind and Blizzard Open StarCraft II as an AI Research Environment
DeepMind's scientific mission is to push the boundaries of AI by developing systems that can learn to solve complex problems. To do this, we design agents and test their ability in a wide range of environments from the purpose-built DeepMind Lab to established games, such as Atari and Go.
588
Scott’s Supreme Quantum Supremacy FAQ
You’ve seen the stories—in the Financial Times, Technology Review, CNET, Facebook, Reddit, Twitter, or elsewhere—saying that a group at Google has now achieved quantum computation…
589
Unsupervised sentiment neuron
Stay up to speed on the rapid advancement of AI technology and the benefits it offers to humanity.
590
Evolving OpenAI's Structure
An update from the OpenAI board on transitioning its for-profit entity to a Public Benefit Corporation, reinforcing its mission-driven structure under nonprofit oversight while enabling greater impact and long-term alignment with the public good.
591
Gemma 3 QAT Models: Bringing AI to Consumer GPUs
Explore Gemma 3 models now offering state-of-the-art AI performance on consumer GPUs with new int4 quantized versions optimized with Quantization Aware Training (QAT).
592
AI surveillance should be banned while there is still time
All the same privacy harms with online tracking are also present with AI, but worse.
593
An animated introduction to Fourier series
This article will be part of an extensive series in which I plan to explore various aspects of Fourier Mathematics. I will take notes, create some visuals (a good pretext to learn more about graphics), and hope that it will be helpful to anyone looking for a clear, visual introduction to these concepts. The article…
594
595
4TB of voice samples just stolen from 40k AI contractors at Mercor
Lapsus$ posted Mercor on its leak site. Voice biometrics plus ID documents in one bundle is a deepfake-ready kit. Here is what to do.
596
Mistral Large
The most powerful AI platform for enterprises. Customize, fine-tune, and deploy AI assistants, autonomous agents, and multimodal AI with open models.
597
A dumb reason computer vision apps aren’t working: Exif Orientation
I’ve written about lots of computer vision and machine learning projects like object recognition systems and face recognition projects. I…
598
A Jev that can ask Opus for help gives Opus accuracy at 4x lower cost
We tested Jev and GPT-6 Luna as judges for identity resolution. A hybrid of Jev and Claude Opus matched frontier accuracy at over 4x lower cost.
599
Deep learning algorithm diagnoses skin cancer as well as seasoned dermatologists
In hopes of creating better access to medical care, Stanford researchers have trained an algorithm to diagnose skin cancer.
600
Qwen3.6-Plus: Towards real world agents
Qwen offers comprehensive functionality spanning chatbot, image and video understanding, image generation, document processing, web search integration, tool utilization, and artifacts.