Article archive
Browse individual source reports, including coverage still awaiting the summary or community analysis needed for our story feed.
5,357 articles · page 17 of 108
Browse topics
Spymarks, Not Watermarks
1 min read
Watermarks that spy on users are no mere watermarks
I don't want to read what you didn't write
15 min read
I Don’t Want to Read What You Didn’t Write People who rarely produced original writing are suddenly producing extensive design proposals, business plans, documentation, presentations, tickets, pull requests, blog articles, and meeting summaries, all generated by AI. It is all unreadable. AI is helping me write better and faster, but I’m fed up with reading almost anything written by AI. In this es...
The US government opens the door to working with voice AI platforms
1 min read
Phase Ongoing Certification Status FedRAMP Certified As of 8/25/2026 Authorizations 0 Certification Profile - Type - 20x - Path - Program - Class - Class A (Pilot) Certified Since N/A Ongoing Certification FedRAMP Certified As of 8/25/2026 0 N/A
Xiaomi MiMo v2.6
1 min read
Introducing the MiMo-V2.6 series: frontier intelligence, all the modalities, built in public.
Transformers Explained Visually
12 min read
What is a Transformer? Transformer is a neural network architecture that has fundamentally changed the approach to Artificial Intelligence. Transformer was first introduced in the seminal paper "Attention is All You Need" in 2017 and has since become the go-to architecture for deep learning models, powering text-generative models like OpenAI's GPT, Meta's Llama, and Google's Gemini. Beyond text, T...
AI coding has made CI a bottleneck, so we reworked ours to keep up
10 min read
AI coding has made CI a bottleneck, so we reworked ours to keep up Earlier this year, I opened Linear to find that Tuomas, our CTO, had assigned an issue to me, titled “CI costs are high.” While I was at it, he also wanted me to make CI faster. Agents have made it exponentially faster to ship code, but validating those changes hasn’t quite kept up at the same rate. Every PR still has to pass throu...
The Advisory Group on Mathematics and Artificial Intelligence
2 min read
[This is a guest post by the Advisory Group on Mathematics and Artificial Intelligence. This blog post was initially written in a different file format and converted using AI. — T.] We would like to use this guest post to announce the creation of the Advisory Group on Mathematics and Artificial Intelligence hosted at the Institute for Advanced Study (Princeton) and online at agmai.org. The rapid a...
Wall Street Is Growing Skeptical of the Data Center Boom
1 min read
Supported by Wall Street Is Growing Skeptical of the Data Center Boom Several companies tied to the data center industry have delayed their initial public offerings amid the increasing public backlash to these energy guzzling facilities. The backlash against data centers that is shaping elections and kitchen table debates across the country is now spilling onto Wall Street. Companies that house th...
Frontier AI on Your Own Hardware
19 min read
In one of my classes I asked the question I was afraid to ask but I just needed the answer to: “Who is afraid of not getting a job after graduating?” About eighty percent of the 150 people in the room raised their hands. That is roughly 120 students answering, in one motion, that they do not believe there is a place for them in the future. The other story arrives by email. PhD students who cannot ...
US halts flights at busy East Coast airports, says fiber line cut
1 min read
Posted by allanbreyes. Score: 181 points. Comments: 101.
Why does mathmain need an encrypted loader?
13 min read
Why Does an npm Math Library Need an Encrypted Loader? On this page We found a remote access implant hidden inside [email protected] , an npm package that copies the popular mathjs library. The malicious code ships encrypted. It stays dormant until a program solves a specific equation with the library. That equation is the key. When the key matches, the package decrypts a payload and runs it. The ...
Turn off and restrict access to Apple Intelligence features on Mac
1 min read
On your Mac, you can turn off Apple Intelligence features like Siri AI, summarization features, and more.
Advisory Group on Mathematics and Artificial Intelligence
2 min read
Advisory Group on Mathematics and Artificial Intelligence On August 28, we began training a new internal model. In addition to resolving the Navier–Stokes Millennium Prize problem, this model has now resolved more than 100 long-standing open problems across most areas of mathematics. The pace of its progress in mathematics has surprised the mathematicians within OpenAI. This has led to internal di...
Amazon blocks Meta’s new Muse AI agent from shopping on amazon.com
4 min read
Amazon began denying Muse access to Amazon.com on Sunday, telling users that continued activity by an unauthorized agent violates the conditions tied to their accounts, according to GeekWire. Meta introduced Muse on Sept. 8 as a personal agent that can open a browser, fill out forms and check out for the person using it. Starting Sunday, people who asked it to shop on Amazon started getting a popu...
Fable 5 – Median thinking declined in August
1 min read
After Anthropic made Fable 5 permanently available in subscription plans, I noticed a large drop in performance. The model felt dumber, and I couldn't explain why. Measured five different ways, August delivered dramatically fewer thinking tokens than July.
How do Traffic Signals Work (2019)
10 min read
How Do Traffic Signals Work? If you live in a major city, I can take a pretty good guess at one of your most common frustrations: traffic. In city driving, the journey is rarely better than the destination. In most cases, we just want to get where we’re going. Traffic is not just frustrating, but it has consequences to the environment as well. All those idling vehicles have an impact on air qualit...
Grok 4.7
2 min read
SpaceXAI's most powerful model for coding and knowledge work. Twice as fast, at half the price of comparable models. Grok 4.7 is our most capable model for coding and knowledge work. It works longer on difficult tasks, checks its own work more carefully, and comes with our best-calibrated safeguards to date. Served at the same price and speed as Grok 4.6, it is highly competitive in its class. On ...
Good people refuse to do bad things
5 min read
Good people refuse to do bad things · 5 min read I resigned from Anthropic today. I spent the last three years doing pretraining research at both OpenAI and Anthropic. Neither company is acting responsibly. They are racing straight to self-improving superintelligence and gambling with our lives. More thoughts below. — Jacob Coxon (@hilbertspaess) September 9, 2026 This is how Jacob Coxon said good...
The Claude Delusion
18 min read
Today's links - The Claude Delusion: What if we're the ones having hallucinations? - Hey look at this: Delights to delectate. - Object permanence: 9/11 v sex; 9/11 v flags; US customs v diplomatic Vegemite; CIA torture survivors speak; HP's inky timebomb; "The Raven" pop-up; Forex v Zimbabwe's internet; Shirky v "expert Wikipedia": McSweeney's v RIAA; Italy's internet disconnection law; Facehugger...
Attention is all you have
3 min read
Attention is all you have The Tetris effect is one of psychology’s most easy to reproduce experiments. Simply spend a bit of time playing the eponymous game every day for a few weeks. After a little while, you’ll start recognizing familiar Tetromino shapes in clouds, buildings, and everyday objects. You might even see them appear before your eyes when you start falling asleep. Attention hijacking ...
Noodle Gallery – Self-hosted photo and video manager forked from Immich
1 min read
Self-hosted photo and video manager forked from Immich, adding multi-owner Shared Spaces, global people search, S3 storage, and Google Photos import. Noodle Gallery is a community fork of Immich that rebases onto every upstream release, so all of Immich's fixes and features land automatically, while adding a substantial set of its own on top. The headline addition is Shared Spaces: multi-owner col...
macOS 27: Workaround to avoid downloading AI models and save storage
1 min read
Workaround to avoid downloading AI models and save storage Update for 27 public release: there are multiple ways to handle this reported by the community. Here are two verified ones, you only need to pick one: - Siri language: Turn off Siri, then pick a language that doesn't match your system and (preferably) isn't supported by it. - Low Data Mode: the original workaround from this post, simply en...
Trump brutally mocked over bizarre effort to rename AI 'superior' or 'extreme' intelligence
2 min read
Trump brutally mocked over bizarre effort to rename AI 'superior' or 'extreme' intelligence It comes as President Donald Trump has criticized efforts to slow the fast-moving industry's development amid growing global concerns about AI, particularly in the last week, with several industry figures raising the alarm about its risks Donald Trump was slammed Sunday over a bizarre poll on Truth Social a...
M5 Ultra Mac Studio Review
16 min read
For the past few days, I’ve been testing the (currently) top-of-the-line M5 Ultra Mac Studio with 256 GB of RAM. I’ll cut to the chase: the M5 Ultra Mac Studio is a dream machine for local AI agents. This computer makes it possible to run personal assistants powered by local models with great performance and no additional cloud costs. If you’ve been skeptical of testing OpenClaw or Hermes Agent wi...
Python Workers are now generally available
10 min read
Python Workers are now generally available Gyeongjae Choi, Dominik Picheta, and Hood Chatham We introduced Python Workers two years ago, providing a way to run Python applications in the Cloudflare Workers runtime. Our goal was to make it as simple to write Workers in Python as it is in TypeScript, and to make the ecosystem of Python packages and frameworks âjust workâ. Today, Python Workers ar...
The End Of Upward Mobility – AI is coming for the meritocracy
15 min read
In April 1941, with the Wehrmacht in Paris and the Ribbentrop–Molotov pact still in force, the political theorist James Burnham published “The Managerial Revolution,” which famously declared that capitalism was finished. His claim was not that it ought to be finished, but that it was, as a matter of observable fact, being replaced. The successor system would not, however, be socialism, as he had h...
Don't Use AI to Write
3 min read
Don't use AI to write Writing requires thinking, don't let an AI do the thinking for you. Tools like Claude, Codex/ChatGPT and Gemini are great. I use them all day (both at work and personally) for coding, research, digging through data, collecting information etc. But what I don’t use them for is writing, and I strongly suggest you do the same. Don’t skip the thinking! Writing forces you to deepl...
Jev-Leftpad
1 min read
Left pad a value with Jev. Could this be one line with padStart() ? Yes. Does it need a model call? No. Anyway: npm install jev-leftpad import leftPad from 'jev-leftpad'; const result = await leftPad('jev', 8); console.log(JSON.stringify(result)); // " jev" (probably) Set TYPESAFE_API_KEY before using it. jev-leftpad uses jev-latest through TypeSafe's native @typesafe-ai/sdk . It requires Node.js ...
Kev: Tiny Jev-like family of decision models built on top of Qwen3.5
1 min read
tiny Jev-like family of decision models built on top of Qwen3.5 you can train and run on your own - jaredpalmer/kev
AI chatbots give wrong answers to financial queries 'most of the time'
1 min read
AI chatbots give wrong answers to financial queries ‘most of the time’ Save over 40% on Standard Digital was $540 now $299 for your first year.Save now on essential digital access to trusted FT journalism on any device. Savings based on annualised monthly price. Explore more offers. Trial Then $75 per month. Complete digital access with exclusive insights and industry deep dives on any device. Can...
FBI Director Kash Patel says that AI use at the FBI has "increased by 605%" since he became director — claims that every major tech player is "embedded" in the agency
6 min read
Kash Patel says that AI use at the FBI has 'increased by 605%' since he became director — claims that every major tech player is 'embedded' in the agency A seemingly unsupported figure that nevertheless may be grounded in truth. FBI Director Kash Patel just stated in an interview that he's responsible for a "605% increase" in the bureau's usage of AI. The problem is that while the pattern-recognit...
Why Backprop Goes Backward (2018)
4 min read
Why Backprop Goes Backward Backprogation is an algorithm that computes the gradient of a neural network, but it may not be obvious why the algorithm uses a backward pass. The answer allows us to reconstruct backprop from first principles. The usual explanation of backpropagation (Rumelhart et al., 1986), the algorithm used to train neural networks, is that it is propagating errors for each node ba...
Amiga Unix, Again
3 min read
Year of the Amiga Unix Desktop 2026 Amiga Unix, again Amiga Unix â Amix â was Commodore's System V Release 4 for the Amiga: shipped in 1990â92 for the A2500UX and A3000UX, then left where it stood. amigaux.org is an unofficial community project that picks it up again: Amix on 68040 and 68060 machines and on today's accelerator hardware, with a modern toolchain, a package manager, and drivers for c...
Google's Open Agentic Orchestrator
1 min read
Declare an agentic task in YAML. AX sandboxes it, wires up its workspace, fences its network, and helps running it at scale.
I stopped drinking the AI Kool-Aid
6 min read
I stopped drinking the AI Kool-Aid As a child, you drink Kool-Aid. Then somewhere along the way, you stop. Maybe you switch to sugary sodas your parents never let you drink. Realizing it's just sugar water, you switch to a healthier option. For me, it took a little longer to grow up. I had the Kool-Aid jug and matching cups, and even the cycling hat which I wore until the flip up brim fell off. Th...
Frontier Labs Are Selling Garbage to Fools in Washington
8 min read
Selling snake oil to the United States Congress is an ancient American craft, and the frontier artificial intelligence industry is currently attempting the most audacious hustle in modern corporate history. Every few weeks, another tech billionaire in an expensive suit glides into a Senate hearing room, sits opposite lawmakers who struggle to operate an office microwave, and explains with a straig...
Why MCP Was Always a Bad Idea?
5 min read
Why MCP Was Always a Bad Idea Why MCP Was Always a Bad Idea Recently I went to an all-day event centered around the latest and greatest in the MCP world. While all of the presenters were awesome and seemed to be passionate about the work they were doing, I’m honestly tired of MCP. It’s a horrible protocol built for a time when LLMs weren’t that smart, and we’ve outgrown it. A Brief History MCP was...
I turned Jev into a (lousy) chatbot
5 min read
We know Jev. jevchat turns that into a chat model. At every step it asks Jev one question: Given the user's question and the reply written so far, which symbol comes next? The options are an alphabet plus an option to stop emitting. Jev returns a probability for each one, and the sampler draws the next symbol from that normalised distribution. Append, repeat, and stop when STOP is drawn. There are...
Trying the Software Factory Pattern
4 min read
Trying the Software factory pattern. One of the interesting challenges of the AI ecosystem in 2026 is that new, effective patterns emerge faster than I can adopt them. I’ll find a handful, get back to work, and realize a month later that I’d missed four or five more. The adoption cycle for Imprint this year has been something like: - January: get every engineer onto Claude Code every single day - ...
Flock is rolling out a voluntary severance program
2 min read
Around this time last year, most average Joes had no idea who or what Flock Safety was, as the company largely operated under the radar, successfully marketing itself to HOAs and police departments as a simple, ordinary tool for catching car thieves and other criminals. Obviously, privacy advocates and civil rights groups like the Electronic Frontier Foundation (EFF) and the ACLU were already soun...
I am often wrong
2 min read
I am often wrong September 19, 2026 [I shared this note with my team earlier this week, and am posting it here as well. I hope it is interesting or helpful for others working on building product in the age of AI.] Something that people learn quickly when they work with me is that my approach to pretty much every problem is: - Understand the available information - Gather missing information - Defi...
Prompts aren’t Real
13 min read
Everyone is driving themselves crazy exchanging text with LLM’s these days. I would like to propose an alternative. We should all go insane building interlocking evaluation and optimization pipelines, instead. The path to victory in our war with AI is to ignore the prompts, and bring even more AI firepower to bear. Like my friend Coda says all the time: "the textual nature of prompts leads us to t...
Laya (OS Jev) on Mac M4 CoreML Offline (45 decisions per second)
1 min read
Last active September 20, 2026 23:41 - - Save fordnox/e592d0f68b543fd044be8e6d040863a0 to your computer and use it in GitHub Desktop. Laya on Mac m4 CoreML Offline https://github.com/mizorewww/laya-coreml This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode ...
ChatGPT now knows what you do on other websites via ad collector
6 min read
ChatGPT now knows what you do on other websites via ad collector OpenAI's ad collector at bzr.openai.com sets a cookie called __obi , scoped to .openai.com . The value is while you are on ChatGPT and tied to your ChatGPT account. __obi is then sent to OpenAI from ordinary websites you visit. Any company that buys ads on ChatGPT installs a small piece of OpenAI code on its own site, the same way re...
Pirate Face Rescues LLM Models from Deletion
6 min read
Turn AI into torrents that live forever. Open models - LLMs, image, audio, datasets - as 🧲 magnet links that can never be taken down. No single owner or point of failure. Trending Models Trending Apache-2.0 & MIT models, synced live from Hugging Face - each becoming a checksum-verified torrent. The permanence layer for sovereign AI. Pirate Face is decentralized infrastructure for sovereign AI. Eve...
Teen Social Media Bans Miss the Point
6 min read
Teen Social Media Bans Miss the Point Late last year, Australia became the first country in the world to enact a ban on social media for teens under age 16. The move was celebrated at the time by parents, child-safety advocates, and politicians across the political spectrum. But less than a year in, it suggests the ban is failing: A state-sponsored survey of parents indicated that seven in 10 teen...
Qwen-Image-2.1: Compact, efficient, and unified image creation
1 min read
Qwen offers comprehensive functionality spanning chatbot, image and video understanding, image generation, document processing, web search integration, tool utilization, and artifacts.
Chat-based Large Language Models replicate the mechanisms of a psychic's con
26 min read
For the past year or so I’ve been spending most of my time researching the use of language and diffusion models in software businesses. One of the issues in during this research—one that has perplexed me—has been that many people are convinced that language models, or specifically chat-based language models, are intelligent. But there isn’t any mechanism inherent in large language models (LLMs) th...
The Millennium Problems for Biology
8 min read
Demonstrate the emergence of life from chemical precursors in a laboratory setting. Specifically, demonstrate the unassisted emergence of self replicating RNA- and protein-based cells from a plausible primordial soup with a plausible energy source. A “cell” may be any compartment with a defined boundary. To be considered successful, the following conditions must be met. Firstly, it must be shown t...
If AI coding is lowering your code quality, you're not managing quality right
6 min read
One common take on the coding agents that I see goes something like this: “Sure, AI helps you output more code, but won’t the quality suffer?” It certainly will if you just blindly merge the PRs and send them off to prod. But if you take a thoughtful, layered approach to managing quality, I find that it’s possible to not just keep the number of bugs stable but actually reduce it—while still increa...