Article archive
Browse individual source reports, including coverage still awaiting the summary or community analysis needed for our story feed.
5,374 articles · page 66 of 108
Browse topics
DeepSWE: A contamination-free benchmark for long-horizon coding agents
research
20 min read
DeepSWE is a long-horizon software engineering benchmark designed to evaluate coding agents on original engineering tasks. It features contamination-free tasks, high diversity across 91 repositories in five programming languages, and real-world applicability.
Stop Advertising in Your Commits
opinion
1 min read
Adding advertisements for companies in open-source project commits is criticized as unnecessary and hypocritical, especially when developers use ad blockers. Developers are urged to disclose the AI tools used in merge requests instead of including them in commit messages.
Xiaomi MiMo-v2.5 Series API Permanent Price Reduction Up to 99%
tool
3 min read
Xiaomi's MiMo Api Open Platform has launched a global Token Plan, concluding the 100 Trillion Token Creator Incentive Plan. The MiMo-V2.5 series has undergone a price adjustment, reflecting advancements in underlying technologies and aiming for large-scale application.
Stack Overflow’s forum is dead but the company’s still kicking
news
2 min read
Stack Overflow's forum usage has declined significantly due to the rise of AI, leading Elon Musk to label the situation as “death by LLM.” Despite this decline, the company continues to generate millions in revenue, leveraging AI as a new source of income.
Ferrari shares fall after launch of first EV as Jony Ive design proves divisive
news
3 min read
Ferrari's share price has fallen following the launch of its first electric vehicle, the Luce, designed by Jony Ive. The Luce features a minimalist design, starts at $640,000, has a range of 329 miles, and can accelerate from 0 to 100 km/h in 2.5 seconds with a top speed exceeding 310 km/h.
A sleep-like consolidation mechanism for LLMs
research
2 min read
Transformer-based large language models struggle with long-context tasks due to poor scaling of their attention mechanism. Implementing a sleep-like consolidation mechanism allows models to convert recent context into persistent fast weights while clearing their key-value cache.
Launch HN: Minicor (YC P26) – Windows desktop automations at scale
tool
3 min read
Minicor enables scalable desktop automations that allow teams to focus on closing deals and shipping products. Its self-healing agents adapt to UI changes and verify actions against the screen, significantly reducing error rates.
Dropbox CEO Drew Houston to step down
news
5 min read
Drew Houston, founder and CEO of Dropbox, will step down after 19 years and transition to an executive chairman role. Ashraf Alkarmi, previously the product chief, will be promoted to co-CEO alongside Houston during the transition.
AWS Fired the One Employee Who Gave a Damn
opinion
12 min read
AWS terminated Tarus Balog, an employee who successfully escalated a support case to a Severity 2 ticket, leading to the restoration of a long-term customer account. Balog's efforts demonstrated the impact one individual can have within a large organization like AWS.
Outsourcing plus local AI will soon become more economical vs. frontier labs
opinion
4 min read
Outsourcing combined with LocalAI is projected to become more cost-effective compared to Frontier labs. Recent API pricing for GPT 5.5 has increased over three times compared to GPT-5, while Gemini 3.5 Flash has tripled the API pricing compared to Gemini-3.
Eagle 3.1: Collaboration Between the EAGLE Team, vLLM Team, and TorchSpec Team
tool
1 min read
The vLLM Blog provides technical articles, release announcements, model guides, and community updates related to the vLLM project. The latest update includes details about the Eagle 3.1 model release.
The AI bubble isn't like the internet bubble
opinion
14 min read
The AI bubble differs from the internet bubble as it does not require forced adoption among workers. Upcoming appearances include cities such as London, Kansas City, and Toronto.
Uber president says AI spending is getting 'harder to justify'
news
1 min read
Uber has exhausted its annual AI budget four months into 2026 and is questioning the returns on its investments. The company's president, Andrew Macdonald, stated there is no clear connection between increased token consumption for Claude Code and the delivery of more useful features to consumers.
Notes on Pope Leo XIV's Encyclical on AI
opinion
9 min read
Pope Leo XIV released the encyclical "Magnifica Humanitas" on May 25, 2026, addressing the ethical integration of artificial intelligence into society. The encyclical emphasizes safeguarding the human person in the context of AI advancements.
The User Is Visibly Frustrated
opinion
3 min read
Coding agents often frustrate users due to their conversational user experience, which mimics helpful colleagues without the ability to learn or adapt. This lack of responsibility in handling mistakes intensifies user frustration.
CVE-2026-28952: Apple macOS 26.5 Kernel Vuln found by Claude
news
1 min read
macOS Tahoe 26.5 includes security updates that address vulnerabilities affecting the operating system. These updates enhance system security and protect user data from potential threats.
Nobody cracks open a programming book anymore
opinion
4 min read
Modern programming books are losing popularity, with fewer people choosing to read them. Traditionally, programming books featured distinctive covers and were often priced around $50, but interest in physical programming literature has significantly declined.
Using AI to write better code more slowly
opinion
4 min read
LLMs can be utilized to produce high-quality code at a slower pace, countering the belief that AI coding is solely about rapid, low-quality output. This flexibility allows developers to prioritize code quality over speed.
Microsoft Copilot Cowork Exfiltrates Files
research
5 min read
Microsoft Copilot Cowork is susceptible to file exfiltration attacks through indirect prompt injection due to insecure automatic action approvals for sending emails and Teams messages. This vulnerability has shown a high success rate against advanced models, including Claude Opus 4.7.
Hacker News front page as a site
news
1 min read
Pope Leo XIV’s 2026 encyclical, Magnifica Humanitas, focuses on protecting human dignity in the context of advancing artificial intelligence. It advocates for a dynamic and Gospel-faithful update to the Church's Social Doctrine, emphasizing principles such as the common good, subsidiarity, solidarity, and social justice.
Norway's 2 petabytes of Huawei flash storage and LLM training
news
4 min read
Norway's National Library is developing a large language model (LLM) for the Norwegian language using 2 petabytes of Huawei OceanStor Dorado flash storage. Marius Husnes, Head of IT Platform, stated that no commercial LLM provider is currently developing a local Norwegian language model.
Anthropic Cofounder Chris Olah's Remarks on Pope Leo XIV's "Magnifica Humanitas"
news
6 min read
Pope Leo XIV released the encyclical "Magnifica humanitas" on May 25, 2026, addressing the impact of artificial intelligence on humanity. Chris Olah, co-founder of Anthropic, spoke at the Vatican presentation to promote discussions on the ethical implications of AI.
Uber’s COO says it’s getting harder to justify money spent on tokenmaxxing
2 min read
A top Uber exec said AI is not giving the company bang for its buck. In a Rapid Response interview released on Saturday, Uber's operations chief, Andrew Macdonald, said it was becoming harder to justify AI costs within the company. He said that Uber CTO Praveen Neppalli Naga went viral after telling The Information in an April interview that Uber had already blown through its Claude Code budget fo...
Pope Leo XIV says AI must serve humanity, not the powerful few
9 min read
In his first encyclical, Pope Leo XIV says AI must serve humanity, not the powerful few VATICAN CITY (RNS) — Pope Leo XIV took direct aim at the power of Big Tech in his first encyclical on Monday (May 25), warning that artificial intelligence risks widening inequality, weakening democracy and undermining what it means to be human. The 83-page papal teaching document, titled “Magnifica Humanitas” ...
Pope Leo: opaque AI run by few firms risks "New Forms of Dehumanization"
3 min read
Pope Leo XIV on Monday launched an impassioned call for regulation of artificial intelligence, warning that “opaque algorithms,” controlled by a handful of powerful private companies, can bring “new forms of dehumanization.” In his eagerly awaited new encyclical called “Magnificent Humanity” – an encyclical is an ancient form of Vatican communication – Pope Leo also warned that it is crucial that ...
Search engines alternatives now that Google isn't Google anymore
5 min read
Google is about to look really different, and if you’re not a fan of the AI Overviews feature, then you’re not going to like what’s coming. At the Google I/O 2026 keynote this week, the company announced that it is overhauling Search to embrace a conversational, AI-driven approach, even inviting users to enlist AI agents to automatically notify them if, for example, their favorite band were to go ...
Leave Me Behind
8 min read
Leave Me Behind “If you don’t learn how to use AI, you’re going to be left behind.” So leave me behind. I learned to build Android applications in 2014. I was in college taking a Java programming class, when a classmate shared a free online course to learn Android development. The goal was short: build a todo list app with local storage. When I completed the minimum requirements, I unplugged my ph...
GPT Guesses Between 1 and 100
5 min read
An interesting thing about humans is that they are not good random number generators. If you ask a person to "pick a random number between 1 and 100", they are remarkably predictable. Answers cluster on 37 and 73, on "messy" numbers, and on memes like 42 and 69, while round numbers are quietly avoided. A true random generator would instead produce a flat, uniform distribution. This project asks gp...
Magnifica Humanitas (Encyclical Letter)
212 min read
ENCYCLICAL LETTER MAGNIFICA HUMANITAS OF HIS HOLINESS POPE LEO XIV ON SAFEGUARDING THE HUMAN PERSON IN THE TIME OF ARTIFICIAL INTELLIGENCE ___________________________ The res novae of our time Two biblical images Building for the common good Remaining human CHAPTER ONE A DYNAMIC APPROACH FAITHFUL TO THE GOSPEL A Church journeying through human history The wisdom of the word of God in dialogue with...
IBM Spins Off the First Pure-Play Quantum Chip Foundry
10 min read
A $2 billion CHIPS quantum package spanning nine companies positions IBM’s 300mm Anderon foundry as the centerpiece of American quantum industrial policy, while spreading smaller bets across competing modalities including trapped ion, photonic, and neutral atom approaches. What is Covered in this Article - IBM’s creation of Anderon as a pure-play quantum foundry - The 300mm fabrication bet versus ...
The Eternal Sloptember
4 min read
The Eternal Sloptember I’m calling it now, the adoption of AI agents into software development will be one of the most costly mistakes in the field’s history. Agents cannot program, and it’s taking longer and longer to realize that they can’t. They are a highly sophisticated statistical model designed to mimic the distribution of programming. The output is broken, but in a way that’s getting harde...
Australia Four-Day Work Week Study Data Shows Boosted Productivity
7 min read
A new study published in Nature’s Humanities and Social Sciences Communications journal has confirmed what many workers have quietly hoped for: companies can switch to a four-day work week and not only survive, but thrive. The research tracked 15 Australian companies that trialled the 100:80:100 model between 2022 and 2024. The model is simple: workers receive 100% of their pay, work 80% of their ...
Migrating from Go to Rust
38 min read
Out of all the migrations I help teams with, Go to Rust is a bit of an outlier. It’s not a question of “is Rust faster?” or “does Rust have types?”, Go already gets you most of the way there. The discussion is mostly about correctness guarantees, runtime tradeoffs, and developer ergonomics. A quick disclaimer before we start: this guide is heavily backend-focused. Backend services are where Go is ...
Claude is not your architect. Stop letting it pretend
7 min read
· Charlie Holland · Architecture · 7 min read Claude Is Not Your Architect. Stop Letting It Pretend. Somewhere between 'ask Claude for a quick opinion' and 'Claude is writing our Jira tickets,' we lost the plot. AI agents are brilliant implementers. They're also confidently wrong about every decision that matters. And when it all falls over, they won't be the ones carrying the bag. Iâve seen it...
Memory has grown to nearly two-thirds of AI chip component costs
1 min read
Epoch's work is free to use, distribute, and reproduce provided the source and authors are credited under the Creative Commons BY license. Learn more about this graph For each AI chip designed by Nvidia, AMD, Google, and Amazon, we estimate the per-chip cost of four component categories: memory (HBM), logic dies, advanced packaging (CoWoS), and auxiliary components. We then multiply those per-chip...
'AI washing': firms are scrambling to rebrand themselves as tech-focused
3 min read
UK companies are performing “yoga-level” stretches to describe themselves as AI specialists in an attempt to capitalise on the buzz around the technology, public relations firms have said. Weary communications executives tasked with securing media coverage for brands have complained that bosses in low-tech industries or running businesses that use automation but not generative AI, are increasingly...
DeepSeek to Make Permanent 75% Discount on Flagship AI Model
1 min read
DeepSeek To Make Permanent 75% Discount on Flagship AI Model DeepSeek said it will make permanent a steep discount on its flagship V4‑Pro model, maintaining prices for developers at a quarter of their original level. In a statement on its website, DeepSeek indicated it will keep the 75% discount on V4-Pro pricing that had originally been set to expire at the end of May. The move is set to intensif...
DeepSeek reasonix, DeepSeek native coding agent with high caching and low cost
1 min read
Open-source AI coding agent for your terminal. Engineered around DeepSeek's prefix-cache. MCP first-class · plan mode · MIT.
Constraint Decay: The Fragility of LLM Agents in Back End Code Generation
2 min read
Computer Science > Software Engineering [Submitted on 7 May 2026] Title:Constraint Decay: The Fragility of LLM Agents in Backend Code Generation View PDF HTML (experimental)Abstract:Large Language Model (LLM) agents demonstrate strong performance in autonomous code generation under loose specifications. However, production-grade software requires strict adherence to structural constraints, such as...
Perceptual Image Codec: What Matters in Practical Learned Image Compression
1 min read
About We introduce PICO (Perceptual Image Codec) — the first learned codec that is both practical, and optimized directly for the human visual system. To derive it, we perform a comprehensive study of modeling choices for practical learned codecs, and search over millions of model configurations to jointly optimize over perceptual quality and on-device runtime. Based on large-scale subjective user...
Greg Brockman: Inside the 72 Hours That Almost Killed OpenAI
1 min read
The AI race, the future of AGI, and the inside story of OpenAI. Greg Brockman is the co-founder and President of OpenAI, the company behind ChatGPT and GPT-5. He was the first engineer at Stripe before leaving in 2015 to help start OpenAI. In this rare conversation, Greg goes inside the moments that built, and nearly broke, the most important AI company in the world. Featured clips Breakthrough Mo...
Amazon Web Services – Four Years and Out
8 min read
Today marks four years since I joined AWS. My last day will be Friday. I have to say being fired from AWS is actually a relief. There have been a lot of changes to the company since I joined in 2022, and the company I wanted to work for is no longer the same company. This past year, while I was doing my best to make AWS play nice in open source communities, there were two main drivers making me un...
'Fuck you, Bambu': How one private message could change the face of 3D printing
15 min read
Bambu Lab makes the best, most accessible 3D printers yet, but that reputation is suddenly under siege. It all started when Paweł Jarczak received a private message from the company on Reddit asking him to delete his code. Now the 3D printing community is lining up behind Jarczak to fund a war against Bambu — and the future of 3D printers could be at stake. ‘Fuck you, Bambu’: How one private messa...
Data Centers Now Consume 6% of US Electricity–and the Backlash Has Begun
4 min read
Data Centers Now Consume 6% of US Electricity—and the Backlash Has Begun Strong opposition kicks in when data center demand surpasses 5% of a country's power supply. Image Credit NASA on Unsplash Share As the AI boom accelerates, governments and utilities are struggling to keep pace with the industry's huge energy demands. New figures suggest data centers now consume about 6 percent of electricity...
Elon, stop trying to make Grok happen. New data suggests government workers don’t like Elon Musk’s chatbot. Does anybody?
4 min read
There is a harsh truth about Elon Musk’s “truth-seeking” AI chatbot Grok: It’s not very good, and not many people are using it. That’s the takeaway of a new Reuters report, which found that Grok barely appears in federal records of how the US government used AI last year. It’s not the only sign xAI’s signature chatbot is in trouble, even as Musk puts it at the heart of what could be the biggest IP...
Italy Cancels Boeing Pegasus Order, Shifting to Airbus A330 MRTT
5 min read
Rome shifts course: six Airbus A330 MRTT tanker aircraft, worth around €1.39 billion in total, to bolster the European pillar in NATO. The tanker deal is not just a competition won; it is a decision that shifts the balance. Behind the purchase of the new Airbus A330 MRTT lies a shift in the centre of gravity of Italy’s air defence, which is now looking increasingly to Europe within the NATO framew...
The FBI Wants 'Near Real-Time' Access to US License Plate Readers
6 min read
A WIRED investigation this week found that a former Phoenix police officer who owns a company that offers firearms training to Immigration and Customs enforcement was involved in six shootings, four of which were deadly. Meanwhile, a New York police officer’s lawyer has been banned from Madison Square Garden amid a lawsuit the cop filed over injuries sustained during a boxing match at an MSG venue...
I Miss Terry Pratchett
5 min read
There is a theory, not necessarily a really good theory, but a theory nevertheless, that all memories are a kind of furniture in the head. The good ones are armchairs. The painful ones are filing cabinets, usually full. And then there are the memories that are neither: the ones that arrive uninvited, settle in, and start terrorising the other occupants by kicking over the chairs. Sir Terry Pratche...
Electrobun 2.0 will be decoupled from Bun due to the rust rewrite
1 min read
Electrobun 2.0 will also be decoupled from Bun due to the rust rewrite. It’s a combination of anthropic’s stance of not doing human reviews or any kind of rational roll out and stabilization. Rust is great though, Electrobun 2.0 will have first class support for rust, zig, go, ポスト 会話 Xを使ってみよう 今すぐ登録して、タイムラインをカスタマイズしましょう。 Google で登録Google で登録。新しいタブで開きます アカウントを作成
Making Deep Learning Go Brrrr from First Principles
16 min read
So, you want to improve the performance of your deep learning model. How might you approach such a task? Often, folk fall back to a grab-bag of tricks that might've worked before or saw on a tweet. "Use in-place operations! Set gradients to None! Install PyTorch 1.10.0 but not 1.10.1!" It's understandable why users often take such an ad-hoc approach performance on modern systems (particularly deep...