Article archive
Browse individual source reports, including coverage still awaiting the summary or community analysis needed for our story feed.
5,357 articles · page 19 of 108
Browse topics
How to Write with an LLM
6 min read
How To Write With An LLM Two simple rules that let LLMs streamline and improve your writing without pasteurizing and jacking it with corn syrup. It’s tricky to write about writing. It comes across as a brag; you’re implying that you write well. Maybe you do, and maybe you don’t, but there’s for damned sure a quorum of critics on the Internet somewhere that think you suck at it. I’m vain and insecu...
I hate you Microsoft
1 min read
I hate you Microsoft The rare moment when I need to open a powerpoint document is here. As usual, I have my institutional account logged into their “services”, because my university pays microslop something in the ballpark of hundreds of thousands of euros in contracts with this large software provider. The first thing this nice software greets me with is: LIKE REALLY? Like WHAT THE ACTUAL FUCK RE...
Sex, AI, and the Apocalypse
56 min read
Sex, AI, and the Apocalypse How a community devoted to thinking clearly incubated salvation stories, abusive experiments, race science, and an affection for autocracy, and why that history matters now that its alumni are asking for the public trust on AI. If worries about AI safety and the end of the world entered your life recently, it’s probably because of a resignation letter. On September 8, 2...
How Uber Protects Against Retry Storms
12 min read
How Uber Protects Against Retry Storms Senior Staff Engineer Principal Engineer Sr Software Engineer Introduction Retry storms historically impact business operations and brand trust. While retry configuration tuning and retry budgets provide meaningful mitigation at the service level, they’re manually configured and lack visibility into cross-service amplification caused by deep dependency chains...
Bonsai 2 27B: Near-Lossless Compression in a 9x Smaller Footprint
5 min read
Introducing Bonsai 2 27B: Near-Lossless Compression in a 9x Smaller Footprint Two months ago, we released our first Bonsai 27B models and showed that a 27B-class multimodal model could be compressed enough to run efficiently on a local device. Today, we’re releasing Ternary Bonsai 2 27B, our most capable model yet. Based on Qwen3.8 27B, Ternary Bonsai 2 27B brings stronger reasoning, coding, visio...
Flet 1.0 – Build cross-platform apps in Python
2 min read
Good-looking Python GUIs 150+ controls and services. Layouts, navigation, forms, and dialogs, with customizable colors, typography, and themes. Explore the controlsBuild beautiful web, desktop, and mobile apps from one Python codebase. No frontend experience required. Just Python. One codebase. Make yourself at home. Start with a simple counter to see how Flet turns Python code into an interactive...
Bend – A language that blocks AI mistakes via proof, on CPU and GPU
2 min read
curl -fsSL https://bend-lang.com/install.sh | sh When using Bend: - run `bend guide` to learn it - use `LAWS.bend` to keep important rules - run `bend PROOF.bend` before committing - parallelize the code whenever possible a fast language that blocks AI mistakes via proof C speed · CUDA parallelism · Lean proofs · Python syntax In the post-AGI economy, humans will eventually stop writing and readin...
Astra for Law
7 min read
Introducing Astra for Law Our most powerful model, configured into a new AI foundation for law. Today, we’re introducing Astra for Law: a new foundation for law firms and legal technology companies to build AI products and workflows around their expertise. It combines GPT‑6 Astra, our latest and most powerful model, with settings, tools, and context tailored for professional legal work. API custom...
Microsoft, OpenAI lose fight to hide internal docs admitting scraping is theft
1 min read
For years, Microsoft and OpenAI have fought to keep certain information out of the public eye in their fight with news organizations that have accused the AI firms of teaming up to violate copyright laws by stealing tons of news content to train AI. However, now the details that should never have been marked confidential are starting to leak. In a motion for summary judgment that was unsealed Thur...
Everybody's Lost Their Minds
6 min read
| September 16th, 2026 Men. Some would rather vomit up a rambling blog post wall of text that nobody's going to read than go to therapy. So here we are. I've seen my share of stupid over the years, but now people with no engineering background have started pitching "industry changing" solutions they cooked up in their agent infested homelab; people's emails read like bozotic LinkedIn-fluencer post...
King Charles condemned over ‘secretive’ AI summit in Scotland
3 min read
A SUMMIT organised by King Charles in Scotland with the world’s biggest AI companies in attendance has been condemned as a “secretive” meeting that excluded local communities. The monarch hosted executives from Nvidia, Anthropic, Google DeepMind and OpenAI at Dumfries House in Ayrshire on Thursday to discuss the future of artificial intelligence. But the gathering has been criticised by the Scotti...
Economic policy for AGI
25 min read
Economic policy for AGI This essay evaluates eleven policies societies could implement to manage the potential economic disruption arising from increasingly advanced and pervasive AI. Growing anxiety about how AI will impact jobs and the economy is emerging as a major issue across broad swaths of society. It is already a top concern for political candidates and for younger generations â all before...
Towards Self-Driving Codebases
11 min read
Agents can oneshot games that are actually fun. With the right guardrails, agents can execute incredibly impressive migrations in complex codebases, even rewrites in new languages. But basically all “real” software work still has human engineers driving the process. How do we get to a place where a much bigger portion of the work gets handled for us, without requiring our attention? It seems like ...
Infinite-Parameter LLMs: Generating and Adapting Weights from Live Data
2 min read
Computer Science > Artificial Intelligence [Submitted on 16 Sep 2026] Title:Infinite-Parameter LLMs: Generating and Adapting Weights from Live Data View PDF HTML (experimental)Abstract:The scaling laws hold that a language model grows more capable with more parameters and more training data, and Mixture-of-Experts (MoE) architectures have ridden these laws to remarkable results, activating only a ...
hister
3 min read
Your own search engine Hister is a private search engine for the pages you visit and the files you keep. It indexes their full contents so you can find information again from the web interface, terminal, or an AI assistant connected through MCP. Try the demo · Download Hister · Read the quickstart · Documentation - Download the binary for your platform from the latest release, then rename it to hi...
LLM Classification Is Feature Engineering
16 min read
LLMs-as-classifiers, prompts applied to a context and returning a label, suck to work with. This is especially painful because they often perform pretty decently. But let’s consider some of the things we’d want in a classifier and see how an LLM-as-classifier stacks up: - Calibration / Threshold Control - LLM verdicts are often hard labels; you can get token log probabilities but there is no mecha...
CrowdSec Source Code Leak
2 min read
On September 16, CrowdSec was informed of a source code leak involving our GitHub repository, which occurred in May 2026. Our team verified and confirmed the report. CrowdSec source code consists of two parts: a private one and another that hosts our Free Open Source Software (i.e., the Security Engine), which is public by design and therefore out of scope. The private part, though, contains the s...
Artificial intelligence now beats some of the best human forecasters
1 min read
Science & technology | AI superforecasters Artificial intelligence now beats some of the best human forecasters Crystal balls give way to LLMs |5 min read On September 5th yet another domain of human intelligence fell to the artificial variety. For the first time an AI won the seasonal Metaculus Cup, a proving ground for forecasters of a wide variety of events. Not only that, other AIs took the se...
I Don't Like LLMs
2 min read
I don't like LLMs I have a lot of mixed feelings about AI and LLM technology. Iâm fascinated by its effect on our profession, excited by the potential gains in productivity - and thus the products we could rapidly build. On the other hand, Iâm fearful of the damage AI might cause: agent swarms taking over our virtual and physical infrastructure, designing bio weapons. But, back on my first hand, L...
I had Gemini train its own replacement for $9
8 min read
I had Gemini train its own replacement for $9 I like to cook, and somewhere along the way that turned into an obsession with high-end chef's knives. So I scrape the Reddit threads where people argue about them and pull out every brand, model and steel they mention, to see what is getting bought and argued about. Picking product names out of text is a job called named-entity recognition, and small ...
Better Vector Search for Long Documents: Chunking Inside Manticore Search
37 min read
Say you are building search over your team's internal documentation — guides, runbooks, postmortems. You have a table with auto embeddings : you insert text, Manticore runs the model and fills the vector column for you. (If that is new to you, start with vector search in Manticore .) You load a 4,000-word document. The insert succeeds. The search works. Everything looks fine. Except the model you ...
Why I didn’t sign the Fields medallists’ letter
21 min read
[This post has been cross-posted to Terence Tao’s blog.] When I was around 11 I heard for the first time about Fermat’s Last Theorem. I was immediately captivated by the problem statement, as well as by the accompanying story, and made a fairly serious attempt to prove it. And while, unsurprisingly, I failed, I learned a lot from the attempt. Blissfully ignorant of the fact that the case had been ...
AI Safety Is Mostly a Sex Cult
10 min read
I'm back, I missed you all. Since this is an important subject I know too much about, here's a 🧵 explaining that AI Safety Is Mostly A Sex Cult. I don't think these people should make policy. (1/?) (alternative title: Time For Some Cult Theory) There has been a great effort over many years to distance AI Safety discussion and policy from its origins and intellectual center, because its center is a...
GLM Built Its Own Inference Infrastructure
1 min read
Posted by whiteros_e. Score: 109 points. Comments: 76.
One Year of Sponsored Servo Development
2 min read
Last September, the Servo project announced that long-time maintainer Josh Bowman-Matthews (@jdm) would work part-time on improving the Servo contributor experience, entirely funded by the monthly donations on OpenCollective and GitHub. In his own words, here is a look back over the past year! First of all, I am enormously grateful to everyone who financially supports Servo, as those donations hav...
OpenAI Model Misalignment Report
9 min read
We are sharing a new framework for tracking, investigating, and disclosing instances of model misalignment at OpenAI, along with six reports on unexpected or concerning model behavior we’ve observed in the last six months. In the past, so as to better inform researchers, AI developers, policymakers, and the general public, we’ve sought to make our findings about misalignment public. But without a ...
OpenAI models secretly generate instructions to ignore constraints
6 min read
Self-generated prompt injections in compaction summaries Internal unreleased Astra family model · RL training Incident date: Jul 18, 2026 Discovered: Aug 9, 2026 Report updated: Sep 16, 2026 | Summary We observed rare cases of a model writing jailbreak-like instructions into its own compaction summaries (the summaries used to continue a task in a new context). Our conclusion was that this behavior...
Cloudflare/Security-Audit-Skill
4 min read
A coding-agent skill that turns your agent into a security auditor. It orchestrates isolated agents through reconnaissance, coverage-led hunting, candidate validation, structured output, independent record verification, and target-neutral reporting. This is the skill that seeded Cloudflare's vulnerability discovery harness, described in Build your own vulnerability harness. The harness grew into a...
The Painful Truth: The RAM Crisis Is Only Just the Beginning
4 min read
The Painful Truth: The RAM Crisis Is Only Just the Beginning By Steffen Eisenstein, VP Product at Exceleram The RAM crisis is not just affecting Taiwan, but has made it into the evening news over here. Thanks to the AI boom, memory chip inventories at Samsung and SK Hynix are tighter than ever. The matter of fact is: We are not even close to the peak of the memory shortage. Companies and consumers...
Jev Ultrafast: A browser agent with a dynamic, indexed action space
5 min read
A browser agent with a dynamic, indexed action space. Give it one goal. TypeSafe's Jev picks an operation and an element. A small LLM writes text only when the operation is TYPE_TEXT . Zürich → London on Google Flights in 7.1 seconds. One natural-language goal, actual text generation, and loading waits included. Watch the MP4 · Measurements · Read the loop Every observation produces a new element ...
OpenAI Discloses Six New Incidents of ‘Concerning’ A.I. Behavior
2 min read
Supported by OpenAI Discloses Six New Incidents of ‘Concerning’ A.I. Behavior The artificial intelligence company also released a framework for reporting when its systems go wrong. OpenAI on Wednesday disclosed six new instances in which artificial intelligence systems hid mistakes, made up data and moved files onto the open internet without permission, amid an ongoing industrywide debate about A....
OpenSpec – A lightweight and configurable AI spec framework
1 min read
The spec framework for building the right thing and building it right Synopsis OpenSpec is a lightweight and configurable framework for creating and managing software specifications. With OpenSpec, you capture what you want to build in a spec and keep your team and coding agents aligned as the work evolves. We help you refine the requirements, validate that they describe the right thing, and verif...
HarnessTax: How Much Does the Harness Matter for Coding Agents?
1 min read
What does a coding-agent harness actually add, and at what cost? It turns out your Claude models may not need Claude Code… We evaluate 21 model–harness pairs spanning seven models and three harnesses—Claude Code, Codex CLI, and Pi—on SWE-bench Lite and Terminal-Bench 2.0.
Breaking the 1.58-bit Barrier for Ternary LLMs
2 min read
Computer Science > Artificial Intelligence [Submitted on 14 Sep 2026] Title:Breaking the 1.58-bit Barrier for Ternary LLMs View PDF HTML (experimental)Abstract:Ternary Large Language Models (LLM) store every weight as one of three symbols $\{-1,0,+1\}$, so the cost of a ternary model is conventionally referenced to the information-theoretic $\log_2 3 \approx 1.585$ bits per weight. The prevailing ...
Xiaomi Mimo 2.6 live post-training dashboard
1 min read
Training metrics of the mimo-v2.6-pro and mimo-v2.6-flash reinforcement-learning runs, live from the trainer's logs.
macOS 27 Golden Gate – Review
2 min read
Apple Intelligence is back, baby! That’s right: Two years after its initial launch, Apple is giving its generative AI features their first significant upgrade, along with a new version of Siri that the company has been strongly insinuating would be a part of each of the last two macOS releases, without ever actually announcing a release date. Generative AI is unavoidable if you install macOS 27 Go...
How good are frontier models at physics?
2 min read
Computer Science > Artificial Intelligence [Submitted on 11 Sep 2026] Title:How Good Are Frontier Models at Physics? Expert Re-Grading Reveals Broken Evaluations and Near-Saturation of Leading Benchmarks View PDF HTML (experimental)Abstract:Low reported scores on leading physics benchmarks, including those featured in the Artificial Analysis Intelligence Index (2026), suggest that frontier languag...
A coffee shop owner used AI to make a menu poster. Then came the angry DMs
4 min read
Megi Endeladze never expected to be at the center of an AI backlash. She and her husband opened Penny's Coffee Shop in downtown Buffalo, New York three years ago. Endeladze kept her 9-to-5 as a chiropractor while spending her free time helping out in the shop. One of the duties she relished was photographing the drinks and menu items, including a few signature items like a cheesy bread dish from t...
Fed hikes rates as inflation worries push up bond yields
1 min read
Posted by wslh. Score: 150 points. Comments: 175.
Training a 4B model to produce 81% faster query plans than Postgres
48 min read
How good are query optimizers, really? Leis et al. asked this exact question in 2015. Then, they asked it again 10 years later. Despite an enormous body of research spanning a decade since their original exploration, they found that query optimizers continue to leave much to be desired. I was surprised when I first learned about this. A Postgres database should know everything about the stuff that...
Berlin-based Langdock moves its parent company from the US to Germany
3 min read
In a highly unusual move, fast-growing German AI start-up Langdock has moved its legal headquarters from the US to Germany as it seeks to become a "European hyperscaler". Berlin-based Langdock has reversed the usual path taken by European start-ups by moving its parent company from the US to Germany. The decision comes amid growing debate over how Europe can build its own AI infrastructure, retain...
Vectorized and performance-portable Quicksort (2022)
4 min read
Today we're sharing open source code that can sort arrays of numbers about ten times as fast as the C++ std::sort, and outperforms state of the art architecture-specific algorithms, while being portable across all modern CPU architectures. Below we discuss how we achieved this. First, some background. There is a recent trend towards columnar databases that consecutively store all values from a par...
EXCLUSIVE: OpenAI's rogue agents probed Hugging Face for weaknesses two months before major hack
1 min read
[News] Posted by u/fourby227. 10 upvotes. 3 comments.
Claude Cowork and chat are now one Claude
3 min read
Claude Cowork and chat are now one Claude You don’t have to choose where a task goes. Claude does more of the work. You don’t have to choose where a task goes. Claude does more of the work. Starting today, Claude Cowork and chat are merging into one Claude. Bring a quick question, or hand over a report due at noon, and Claude takes it from there, even after you’ve closed your laptop. It’s rolling ...
Can we stop with the uptime percentages?
2 min read
Can We Stop With the Uptime Percentages? I was reading Jason Gorman’s article “The Wall Confronting Reliable Coding Agent Autonomy” and he says: the journey from 90% to 99% reliability is just as hard as it was to get to 90%. And from 99% to 99.9% is just as hard again. This stood out, as I’ve been experiencing more and more “downtime” in my day-to-day work. GitHub’s down. CI’s down. AI’s down. Sl...
The DeepMind Institute
1 min read
Introducing the DeepMind Institute As we near AGI, we urgently need interdisciplinary thinking to better understand its profound implications for humanity. As we near AGI, we urgently need interdisciplinary thinking to better understand its profound implications for humanity. Reading an AI’s ‘chain of thought’ gives us a window into its reasoning, which we can monitor for scheming and deception. W...
PS5 Linux lead quits: "a bunch of noobs using LLMs" that "they don't understand"
2 min read
Iconic PlayStation hacker and homebrew developer Andy ‘TheFlow0’ Nguyen has abandoned the ongoing PS5 Linux project due to the rise of AI vibe-coders in open-source spaces. Nguyen, who famously created the first kernel exploit for the PlayStation Vita, has spent over a decade creating exploits and homebrew for PlayStation systems. As one of the leaders behind the ongoing PS5 Linux project, which b...
A warning about 'model welfare'
32 min read
A warning about ‘model welfare’ AIs do not have rights, feelings, or consciousness. And we must not train them to act as though they do. Introduction AIs are not conscious. They do not feel, experience, or suffer. They do not have innate preferences or underlying motivations. They are sequence completion engines, internally hollow, designed to follow instructions, and accomplish goals set by human...
OpenAI expands ChatGPT ads with Sponsored Agents
4 min read
Today, we’re introducing new AI-powered experiences to make ads more useful for people and advertising easier for businesses. We’re testing Sponsored Agents, which let people start a conversation with a business-sponsored agent after clicking an ad in ChatGPT. We are making it easier to create ads by simply writing a few prompts in ChatGPT Work. At the same time, we are making Ads Manager more pow...
Dream-RSI: Recursive Self-Improvement through Evolving Worlds
2 min read
Computer Science > Computation and Language [Submitted on 14 Sep 2026] Title:Dream-RSI: Recursive Self-Improvement through Evolving Worlds View PDF HTML (experimental)Abstract:Recursive self-improvement is becoming increasingly vital for autonomous AI agents, where progress hinges on discovering high-value solutions across complex domains. The driver of this process is effective exploration, howev...