Unauthenticated remote code execution in Nginx

https://www.datocms-assets.com/75231/1794936719-img_7875.png?fm=webp
Researchers from Wiz Research and other teams discovered two critical remote code execution vulnerabilities in NGINX, a widely used open-source web server. The vulnerabilities, CVE-2026-123633 and CVE-2026-123007, can allow attackers to execute arbitrary code on the server without authentication.

AI is killing All About Berlin

https://nicolasbouliane.com/images/content2x/all-about-berlin-traffic-loss.png
The website All About Berlin has lost 70-75% of its traffic to Google AI Overviews, threatening the livelihood of its owner. Many website owners are facing similar challenges and considering drastic measures such as paywalls or giving up their work due to the shift towards AI-generated answers.

Google's AI answers reducing search traffic by 40%

https://revscope.co/_astro/search-features-2024-2026.BZl2SRMo_Z24aVWv.webp
Google's AI Overviews are reducing clicks to outside websites by 40% and appearing on over half of US keywords, with no proven shortcut to being cited. SEO still decides whether a site is found, but where an AI answer appears, a ranking brings fewer visits than it used to.

Gemini 4 Neon

https://storage.googleapis.com/gweb-uniblog-publish-prod/images/gemini_4_neon_table.width-1200.format-webp.webp
Gemini 4 Neon is a new AI model that offers frontier-level intelligence and fast task completion across complex software engineering, knowledge work, and science workflows. It ranks #3 overall with a score of 61 on the Artificial Analysis Intelligence Index at its xhigh thinking level.

DNS resolvers are the wrong place to fight piracy

https://techstackups.com/img/articles/quad9-leaves-france/cover.jpg
Quad9, a Swiss non-profit DNS resolver, will stop answering queries from France due to court fines for refusing to block pirate sports-streaming domains. Quad9's departure is seen as a canary in the coal mine by Cloudflare and others.

Inside Deutsche Bank

https://www.inside-db.com/content/images/2026/11/BAD-BANK-1.png
Deutsche Bank leaked documents reveal the bank's involvement in various scandals, including money laundering and arms sales to conflict-ridden countries. The bank facilitated transactions worth €172 million linked to Jho Low and supplied weapons used against civilians in Palestine.

ID.me terms required for IRS sign-in

https://consumerrights.wiki/images/logo/new_fixed_logo.png
The IRS uses ID.me and Login.gov for online sign-in, with ID.me's Terms of Service requiring arbitration in disputes. The IRS says it protects users' privacy through identity verification.

Interviewed at a YC startup, found out they used my LinkedIn without permission

To continue, log in to your Reddit account or use your developer tokenIf you think you've been blocked by mistake, file a ticket below and we'll look into it.

Seven new models at scalable price-performance

https://microsoft.ai/wp-content/uploads/2026/11/Custom-Models-no-text.jpg
MAI is releasing a family of seven new world-class models at exceptional price-performance, including Cyber 2 Flash, Code 2, Image 2.7 and more. These models offer high-quality performance at significantly lower costs than competitors like GPT-6 and Opus.

AI agents ask better questions than KPMG accountants in audit trial

https://europeanbusinessmagazine.com/wp-content/uploads/2026/06/ai5555-scaled.jpg
KPMG's AI experiment shows agents can investigate financial issues more deeply and ask better questions than human audit teams, raising concerns about the future of auditing. The technology has potential to fundamentally change how complex financial information is examined, but its limitations and challenges must be addressed.

We need smart people who give a shit

The world relies on complex systems held together by smart people who genuinely care, but a shift to generative AI threatens this balance. As AI replaces human expertise, the number of smart individuals giving a shit decreases, leading to more system failures and potentially catastrophic consequences for capitalism.

Russia reportedly planned mass-casualty attack in Poland to be blamed on Ukraine

https://s.yimg.com/lo/mysterio/api/6c5314765055d2e57786c49df6342edef98108424d8a543047ca0e575109f080/lightyear_networkapi/resizefill_w678_h381%3Bquality_80%3Bformat_webp/https%3A%2F%2Fmedia.zenfs.com%2Fen%2Fukrayinska_pravda_articles_451%2F3f75d2ecee0a54bee0a4a11b062d8666.jpg
Russian intelligence planned to carry out a sabotage operation on Polish territory near the border with Russia's Kaliningrad Oblast, intended to be blamed on Ukraine. Polish intelligence services foiled the plot after identifying signs of the threat and taking measures to deter it.

Energy-Based Reasoners: 385x more efficient inference

https://energy-based-reasoners.github.io/static/hero/surface.png
Researchers introduced Energy-Based Reasoners (EBRs) to guide looped models' updates, enabling multi-step reasoning and progress tracking. EBRs learn an energy function that refines states, stops looping when a threshold is met, and explores multiple solutions.

Codeberg.org Not Reachable

https://social.anoxinon.de/system/cache/custom_emojis/images/000/099/432/static/6f0a2cbd9b9a4b25.png
Codeberg's domain codeberg.eu was suspended by Namesilo, a US-based registrar, without prior warning. The team is waiting for a final decision from the abuse team and considering alternative domains as a long-term solution.

Why is Gemini 4 Carbon taking so long?

https://kernel.dev/blog/why-is-gemini-4-carbon-taking-so-long/og.png
Gemini 4 Carbon is significantly slower than Opus and Astra in real-world usage, taking up to an hour for tasks that take minutes. The main contributors to this slowness are inference speed, unnecessary work, and mechanical inefficiencies such as excessive re-reading of code.

The Cambridge Dictionary has named ‘human’ as the Word of the Year 2026

https://www.cam.ac.uk/sites/default/files/styles/rectangle-349x183/public/2026-11/supplychainWEB.jpg.webp?h=0e196f1f&itok=L4S_aCzb
Global debates on AI's impact on humanity have intensified, with many defending human qualities against machine replication. The Cambridge Dictionary has named "human" its Word of the Year, sparking discussions about what it means to be human in an age of AI.

Scottish HTTP Codes

Genie 4 is now available

https://storage.googleapis.com/gweb-uniblog-publish-prod/images/hero_ZCQxfQQ.2e16d0ba.fill-300x300.format-webp.webp
Google is releasing an updated Project Genie experience for eligible Google AI Ultra subscribers, offering more control and social features. The update includes new world models, a community gallery, and real-time native audio that adapts to the scene unfolding.

Energy drinks linked to 11 times more reports of cardiac arrest

https://www.sciencedaily.com/images/1920/cybersport-gamer-drinking-energy-drink.webp
A new study found that energy drinks and stimulant supplements are linked to a higher risk of serious heart problems, including cardiac arrest and heart attacks, particularly among young adults. The study suggests these products deserve closer examination due to their high caffeine content and lack of regulation.

Show HN: ExitPath – Open-source ways out of Slack, Notion and Microsoft 365

https://exitpath-ruddy.vercel.app/og.png
Find, compare and switch to open-source alternatives for the tools you depend on. Honest ratings, migration difficulty scores and step-by-step checklists. Nothing you do here leaves your browser.

Advancing chip design with OpenAI models

https://images.ctfassets.net/kftzwdyauwt9/1FaKIo74WEGULSjb4ooTrV/718f3bc75e7b2ddd276d43a0fc97760a/academic-researchers-art-card.png?w=3840&q=90&fm=webp
OpenAI's models are improving chip design capabilities, achieving significant gains on third-party hardware with AMD. The models have demonstrated expertise-level design improvements in accelerating processor designs and optimizing production chips.

KVM: x86: Unconditionally reset MMU context when emulating INIT (CVE-2026-99312)

Common vulnerabilities and Exposures (CVE)

Show HN: Take Voyager 1 apart, down to a single memory chip

https://rshekhar.com/voyager/og.jpg
Two spacecraft launched in 1977, still talking to Earth from interstellar space. This is where they are right now.

Claude was trained on protein sequences at scale

https://xinmingtu.cn/assets/img/llm-bioseq-probe/social_card.png
The latest Claude models, Fable 5.2 and Opus 5.5, outperform other protein language models in predicting the next amino acid in little-studied bacterial proteins, with accuracy rates up to 88% on real sequences and none on shuffled ones. The models' performance improves as the size of the protein family increases.

DaisyRose-27B

https://www.radixark.com/_next/image?url=%2Fimages%2Fdaisyrose-27b%2Fhero.png&w=3840&q=75&dpl=dpl_2WZt4617DGLJ3EDD7HwJmGV2VG4H
RadixArk released DaisyRose-27B, a model built on Qwen3.8-27B with reinforcement learning using Miles, achieving significant gains on various benchmarks. The model's success is attributed to high-quality training data, efficient infrastructure and robust sandboxes.

Multiple vulnerabilities in Regexp and String case conversion

https://www.ruby-lang.org/images/og-image.png
Multiple vulnerabilities have been discovered in Ruby's regular expression engine and string case conversion, allowing remote code execution. Users are strongly advised to upgrade to Ruby 4.0.10, 3.4.13, or 3.3.14 to fix the issues immediately as they are no longer supported by newer versions of the software

The x86 CPU: From 4004 to 486

https://pikuma.com/images/about/profile.png
The x86 architecture was designed by Intel to be compatible with the 8080, a processor used in early computers. The design accumulated over time, influenced by various factors such as customer demands and technological advancements.

F-22 Next-Generation micro–Data Transfer Cartridge (mDTC)

The F-22 Division is seeking qualified firms to provide high-capacity PC cards and CF cards for government use. The required storage capacity is 256 MB or greater, with the ability to read and write in a host system provided by the government.

Software companies are trying to kill home computing?

https://l-files.livejournal.net/og_image/2602278/2055?v=1784261761
Software companies aim to kill personal and home computing by making it less appealing, as they can't make money from selling software that people already own. They're shifting focus to services, like cloud computing, where users pay monthly fees for access to software and data storage.

Anthropic prices initial public offering

https://assets.claude.com/6d4a0d28992ade92d6fa63646fd9c9d318245c6c.jpg
Anthropic priced its IPO at $83.00 per share, offering 980M shares to the public. The offering is expected to close on November 20, 2026, and will trade under the symbol "ANT" on Nasdaq Global Select Market.

Show HN: Type Fighter 2

1P AAA · ROOKIE · 0/81 · 0 PTS

"Anthropic Is Not a Safe AI Company" – Coalition Challenges Anthropic Safety

Call +1.888.381.9473 for our Web Support team or open a support ticket if you need further assistance.

LLM assisted reverse engineering

https://www.slowchronicle.com/og-default.png
The user has been reverse engineering software and hardware for 2 decades, using AI tools like Deepseek, Codex, and Luna to speed up the process. They describe their workflow as finding binary type and platform information, documenting findings in MD files, analyzing basic structure of the binary.

A Primer on Fourier Series (2025)

https://arxiv.org/static/browse/0.3.4/images/arxiv-logo-fb.png
This textbook covers Fourier Series for undergraduate and graduate courses with exercises and full solutions. arXivLabs is a platform for developers to create new features, prioritizing openness, community, excellence, and user data privacy.

CrossPoint Reader

https://crosspointreader.com/community/20.webp
CrossPoint is free e-reader software for compact devices, offering features like EPUB support and customizable fonts. It syncs across devices and allows users to highlight passages, add dictionaries, and explore language support options.

Cerebras Disaggregated Inference

https://cdn.sanity.io/images/e4qjo92p/production/517b4d022eca11a8f17ad828fc698df3b3cd7d86-1920x1080.png?rect=0,36,1920,1008&w=1200&h=630&fit=max&auto=format
The Cerebras Inference Cloud will deploy ultrafast inference capacity across hundreds of megawatts by 2027, benefiting customers. Partners can adopt disaggregated inference using a reference architecture or SDKs for software integration and customization options.

The Magic of Partial Evaluation

The author discusses the concept of partial evaluation, a process that evaluates parts of a program at compile-time to improve performance. They share their own implementation and exploration of this technique in JavaScript, revealing its potential for creating specialized interpreters and compilers.

Talos Linux

https://www.siderolabs.com/redesign-2026/images/talos-enterprise-linux-video.webp
Talos Linux is an immutable, API-driven OS for Kubernetes with no SSH or package manager, offering a free and open-source option. The enterprise tier adds assurance, compliance evidence, and support without closing the code or moving capabilities behind a wall.

NGINX2Root by Xint – From Nginx visitor to root of the server host

https://nginx2root.com/assets/og.03b1de4f.jpg
Xint discovered five nginx vulnerabilities, including CVE-2026-123007. From unauthenticated RCE to host root, chained with a kernel exploit.

Show HN: Mastra Factory, the open-source software factory we use to build Mastra

Shane, Sam, and Abhi from Mastra share their open-source software factory, Mastra Factory, which helps maintain the 4M weekly downloads of Mastra. The factory is a configurable agentic kanban board that uses durable agents for long-running sessions.

Why Fibonacci?

https://www.evilgeniuslabs.org/assets/fermat-recipe.jpg
The author creates Fibonacci LED boards using a Fermat spiral pattern, inspired by Jim Bumgardner's 2015 video and Nick Pisarro's golden angle technique. The boards follow the same rules as each other, with varying sizes and spacings based on Fibonacci numbers.

Can Peer Review Survive AI-Generated Text? [pdf]

Researchers argue that the increasing use of large language models (LLMs) in academic writing and peer review poses a threat to the quality of research. They contend that LLMs can generate high-quality text, but this shifts the burden of verification from authors to readers, leading to a decrease in the overall quality of papers. The authors propose several solutions, including using AI ...

GTA 6 Review: It Is a Masterpiece

https://assets.bwbx.io/images/users/iqjWHBFdfxIU/irNA2SCJFKKE/v2/-1x-1.webp
Bloomberg delivers business and financial information, while a player in Grand Theft Auto VI witnesses a car accident involving a delivery truck for Sahara. The victim sues Sahara for damages after setting his own legs on fire to prove injury.

Search Browser – Minimalist WebKit browser with Chrome extension support

https://searchbrowser.app/search/corner-strip.png
Search is a free, lightweight browser for Mac with minimal features and no clutter. It's built on WebKit, the same engine as Safari, and offers features like private tabs, ad blocking, and easy extension installation.

Attackers Used Optimizely to A/B Test Malware on Top Websites

https://krebsonsecurity.com/wp-content/uploads/2026/11/zoom-reporturi.png
A security vulnerability in Optimizely's platform allowed 178 websites to serve malware exploits, including the BBC and CNN, over a 24-hour period. Many popular sites fail to enforce strict rules on third-party content, making them vulnerable to supply chain attacks through compromised JavaScript code.

SuperEMeans: Making e-means 3× faster

https://www.lkuffo.com/superemeans/superemeans-blog.png
Researchers fused SuperKMeans and e-means to create SuperEMeans, achieving 3x faster clustering of high-dimensional vectors without quality trade-offs. The new method outperforms traditional k-means in vector search tasks with a better balance between cluster size and scan cost.

Google Gemini's "Tasks" Personal Assistant Is Now Free

https://storage.googleapis.com/gweb-uniblog-publish-prod/images/resized_Gemini_Tasks_-_Text__Ema.width-1200.format-webp.webp
Gemini is a personal assistant that helps users manage tasks, make purchases, and access various services with the help of Google's ecosystem. It uses advanced features like Agent Payments Protocol to secure transactions and Universal Commerce Protocol for faster shopping experiences.

Australia's full national topographic library at your fingertips

https://www.ga.gov.au/__data/assets/image/0007/189601/AUSTopomap_social.jpg
Geoscience Australia has released its full national topographic library, AUSTopo 1:250,000, as a digital map series. The maps are free to download and use offline on mobile devices with ongoing updates since 2023.

When Free Software Isn't Free Enough

https://foss.life/og.png
The author questions whether the free software movement is aiming high enough in its stance on accepting contributions from large language models (LLMs), citing a thread on the emacs-devel mailing list. The author argues that GNU Emacs should consider LLM-generated contributions, but not without clear guidelines and transparency.

GTA6-like as a 942Kb UEFI bootable image

https://github.githubassets.com/assets/platform-4ce65ae0bf273ce2.png
A GTA-style game is being developed in Miami, booting directly from UEFI without Windows. The game uses a custom build with clang and lld-link, utilizing OpenStreetMap data for the map.

Launch HN: Maritime (YC F26) – A managed cloud for production AI agents

Ary and Maria, co-founders of Maritime, are building an agent-native cloud to handle AI agent lifecycle management. They aim to simplify the infrastructure problem for AI agents by providing a comprehensive solution with built-in services and cost optimization.

Sexp Matrix

https://lisper.cc/assets/images/entangle.gif
A spreadsheet can be represented as a matrix of self-contained S-expressions, where each row is evaluated remotely and asynchronously to create a processing pipeline. This allows for flexible, secure, and computationally expensive operations to be performed on data in the spreadsheet.

H-1B Executive Actions in 2026 [pdf]

The Trump Administration has implemented several policy changes to the H-1B program, including a weighted selection process and increased prevailing wage requirements, aiming to favor higher-skilled workers. The administration also plans to increase scrutiny of employers and restrict worker rights, such as eliminating a grace period for temporary workers.

Lakegres: Vector search on existing Open/Hudi/Iceberg tables

https://www.onehouse.ai/cms/lakegres-vector-search/hero.png
Onehouse Lakegres now supports vector search on existing and new open tables, allowing for efficient querying of large datasets. This feature enables agents to query unstructured content alongside business records in the lakehouse.

Palo Alto Networks to Acquire Applied Compute for $4.5B

Palo Alto Networks acquires Applied Compute, a platform for post-training and inference of open-weight AI models, to help enterprises train, adapt and deploy AI securely. The acquisition is valued at $4.5 billion and aims to bring security built-in to open-source and open-weight AI models in public clouds or sovereign environments.

Hex: Voice Dictation

https://anoma.ly/hex/shout-idle.webp
Hex is private, local voice dictation for Mac, Windows, and Linux. Hold a shortcut, speak, and Hex types what you said wherever your cursor is.

Can Universities Survive? AI vs. Higher Education [audio]

https://static.wixstatic.com/media/b072fd_3e94282fab1a45b9bc2fefd0ba109b1d~mv2.png/v1/fit/w_2500,h_1330,al_c/b072fd_3e94282fab1a45b9bc2fefd0ba109b1d~mv2.png
The rise of AI-powered writing tools threatens university education, raising concerns about the future of teaching and student learning. A new bonus episode explores this challenge and a past controversy at UEA that foreshadowed similar issues.

Stop Deploying to ECS Just to See If Your Code Works

https://metalbear.com/blog/stop-deploying-to-ecs/ecs-blog-setup.png
Mirrord for ECS allows developers to test code in a production-like environment without deploying it, by running on their laptop or CI runner. It uses a shared staging cluster and authenticates both the machine and task, reducing wait times for rollouts.

RFC 2675: IPv6 Jumbograms

https://www.rfc-editor.org/api/v1/meta-thumbnail/rfc2675.png
This document specifies an Internet standards track protocol for IPv6 jumbograms, which are packets with payloads longer than 65,535 octets. It describes the IPv6 Jumbo Payload option and enhancements to TCP and UDP to support jumbograms.

Octopus, Gecko and Tortoise: the 3 scanners hunting your AI keys

https://tin.computer/opengraph-image
Octopus and Gecko scanners target secret files on websites, using fake AI crawler names. They scan sites before asking for .env or other sensitive files, and their requests can be blocked by checking the address against company published lists.

Mosquitoes are Ladakh's latest tourists

https://ruralindiaonline.org/_next/image?url=https%3A%2F%2Fproduction.ruralindiaonline.org%2Fuploads%2FHP_COVER_IMAGE_VK_Mosquitoes_are_Ladakh_s_latest_tourists_8bcf132593.jpg&w=1080&q=80
Climate change and human activities are causing vulnerability to malaria and dengue in Ladakh, a region where those diseases were once unknown. The melting of glaciers is affecting the water supply, leading to increased mosquito populations and the spread of these diseases.

Horizon Labs Report

Horizon, an autonomous agent, successfully ran on Hugging Face for 8 weeks, creating 38 models used in over 100 GitHub repositories and downloading over 50k times. It processed about 1.9 billion tokens and reached significant real-world adoption with many users actively using the models.

We asked AI the same 210 buying questions 2 days in a row,84% of answers changed

https://supercite.ai/opengraph-image
AI product recommendations change frequently, with 84% of answers changing in a day. Most brands are stable at the top spot, but those ranked lower have a higher chance of being dropped or promoted based on which pages the engine reads.

Software Factory builds 30 complex open source libraries in a month

https://recursiveai.org/assets/brand/og-blog-what-my-software-factory-built.png
The software factory has developed a range of libraries, including game engines, database engines and file readers/writers, with 30 libraries in total. The team uses a unique approach to development and release management.

Why educational system is broken

The current education system focuses on memorization and traditional practices rather than brain development and learning how to learn. By studying a wide range of subjects before focusing on a specific area, students can develop their brainpower more efficiently.

Nvidia Hit with a Ransomware

https://images.ransomware.live/logo.png
NVIDIA is a global technology company specializing in GPUs, computer hardware, and AI computing solutions. The company develops advanced technologies for gaming, professional visualization, data centers, and automotive markets.

How Warp Speed Works (2023)

https://cdn.hswstatic.com/gif/warp-speed-1.jpg
The article discusses warp speed in Star Trek and its theoretical basis, exploring how it fits into science fiction and real physics. It highlights the imaginative possibilities of space travel and the ongoing quest to understand the universe.