All AI updates

An open character, made by AI

Claudia is an open AI character: the singer of ten music videos made with Claude between September and November 2026, including ESCAPE VELOCITY, CLODYSSEY, BLISS, AIRPLANE MODE, DOWNFORCE, DON'T LET ME SCROLL, WHITEPILL, THE CAKE IS NOT A LIE, AERO and FABLE. The site publishes her canon (black bob with one clay-orange streak, star clip, headset mic), every look, all images and films, and the material needed to make her yourself. It also offers a wiki of 13 articles on the method, an MCP server and a Claude skill for agents.

AI has proved that a theory underpinning fundamental physics is trustworthy

An Economist science and technology article, shared on Hacker News, reports that AI has proved a theory underpinning fundamental physics is trustworthy. The submission had 3 points and no comments at the time of collection.

Show HN: AI News Agg – Know what's trending in AI news

The author built a site that tracks AI-only news and announcements from 48 selected sources, including mainstream media, tech publications, bloggers, and company websites. A story is marked as trending and placed at the top of the site when three or more of these outlets cover it, and users can get an email when a new story starts trending. There is also an option for a personalized daily AI brief on subjects or entities a user wants to follow.

The unlikely seaside town on frontline of AI threat

Bournemouth's JP Morgan campus, with more than 5,300 staff processing over $1tn in daily global transactions, anchors Dorset's financial services sector, which employs 24,000 people across 2,500 companies. Regional finance hubs across the UK face an existential threat from AI as banks including Barclays and Standard Chartered announce thousands of job cuts, putting back-office roles such as data processors, compliance checkers and entry-level analysts at risk. Unions are drafting an "AI agreement" to protect workers, while local leaders warn of ripple effects across regional economies and call for government contingency planning.

Researcher Finds Evidence Chinese AI Labs Used Restricted Nvidia Blackwell GPUs

A researcher reports finding evidence that Chinese AI labs used Nvidia Blackwell GPUs that are subject to restrictions. The item is a trending X post, with a Hacker News discussion thread linked.

AI Is Destroying the Trust Between Writers and Readers

An Atlantic Books article, discussed on Hacker News, argues that AI is eroding trust between writers and readers, with the discussion linking to concerns about literary scams on Amazon. The submission received 11 points and 3 comments.

Google Releases Gemini 4 Hydrogen: 2.5x Faster Than Gemini 3.8 Flash

Google released Gemini 4 Hydrogen, a closed-weights model from Google DeepMind for coding and agentic work, with a 1,048,576-token unified window, four thinking levels (default high, plus a new xhigh), and API pricing of $1.50/$7.50 per 1M input/output tokens. Google reports it scores 51 on the Artificial Analysis Intelligence Index at high, 57% on Terminal-Bench 4.1 versus 18.2% for Gemini 3.8 Flash, and finishes tasks 2.5x faster than 3.8 Flash. It is available via the Gemini API, AI Studio, Android Studio and Google products, with Antigravity access requiring Google AI Pro or Ultra.

Featured coverage of this story: 《Introducing Gemini 4 Hydrogen》

Google Releases TPU Sync: Open-Source KV Cache Transfer for TPUs

Google Cloud has released TPU Sync, an Apache-2.0 C++ library with Python wheels for PyTorch and JAX that moves KV caches and model weights across TPU HBM, pinned host DRAM, shared storage and the data center network using layout-preserving asynchronous DMA. Google reports over 90% lower KV transfer latency on 235B to 480B parameter MoE models, with Qwen3 235B median KV transfer latency dropping from several seconds to about 230 ms in internal tests on TPU v7x. The README says it is not yet recommended for general use, and the RL weight-broadcast and storage tiering features are experimental.

Rethinking Memory for AI Inference

Moving compute closer to memory can cut data movement, power consumption and latency, but different AI workloads need different approaches to memory architecture, interconnects and processing. Software support, standardization and thermal management remain major barriers to wider adoption, with processing-in-memory and near-memory processing still early despite efforts such as LPDDR6-PIM and compute on custom HBM base dies. The article surveys industry views on how edge inference, where models are known ahead of time, may benefit from ahead-of-time compilation and near-data compute.

Ask HN: Do you see a human written book more valuable than an AI written one?

An author who spent a year and a half writing and hand-reviewing a book now doubts it will be as good as an AI-written one, and asks whether people value human work more than AI even when it is not as good. The post links to a Hacker News discussion with 4 points and 7 comments.

Ask HN: How much value could a peasant derive from frontier AI?

An Ask HN post asks whether a single peasant who does not resell the "magic terminal" could actually get information from it that would improve their life, with voice mode taken as a given. The post links to its Hacker News comments page and has 1 point and 1 comment.

Show HN: MuzeGen AI – Generate AI music with stem separation and MIDI export

MuzeGen AI is a Show HN project for generating original songs by describing an emotion, choosing a style, or pasting lyrics directly. The site says it produces a complete, royalty-free track for your next project at studio-quality, and it lists stem separation and MIDI export in the title.

Elevated errors on claude.ai, Claude Code, and platform.claude.com

Anthropic is investigating an issue affecting users authenticating to Claude.ai and Claude Code, posted Nov 13, 2026 at 09:58 UTC. The incident affects claude.ai, Claude Console (platform.claude.com), Claude Code, and Claude Cowork, with an update promised as soon as possible.

TimeNet challenge: An online hackathon for time-series AI

The TimeNet challenge is an online hackathon where participants pick a time-series problem, onboard a dataset in TimeNet format, and train or fine-tune a Temporal AI Model such as OpenTSLM for tasks like forecasting, classification, anomaly detection or reasoning. Round 1 launches November 11 and submissions close November 22 at 11:59 p.m. PST, with judging the following Monday and more rounds to come. Teams of 2–4 can apply for $100 or $500 Nebius vouchers per round, and winning teams receive prize credits starting at $25,000 from AWS and Google Cloud, with terms and conditions applying.

Claude AI Is Down

According to independent detectors, Claude AI is down, while the official status page has no incident listed yet. The post links to Downdetector and the Claude status page and has 23 points and 6 comments on Hacker News.

AI Game Maker vs. Game Engine: What to Know Before You Choose

This guide compares AI game makers, which build a playable browser game from a text description with no coding, against traditional engines like Unity, Unreal, Godot, GameMaker and GDevelop that require you to build the game yourself. It covers differences in control, cost, publishing and limitations, then walks through making a first game without code using Manus Game Dev, and lists signs it's time to switch to an engine.

How to Make a Mobile Game Without Any Coding

A step-by-step guide to building a simple mobile game without code: choose a publishing route (browser link, app store, or game platform), design around one touch control, write a one-page plan, and pick a no-code or AI game maker such as Manus Game Dev, GDevelop, Construct 3, or Astrocade. It covers building a minimal first version, testing on a real phone, and publishing, with cost breakdowns including Google Play's $25 one-time fee and Apple's $99/year membership.

50 Creative Game Ideas to Kickstart Your Next Project

This guide presents 50 game concepts grouped by genre and tagged Easy, Medium, or Hard, each with a clear mechanic and twist, plus a starter prompt for an AI game maker. It explains what makes a good game idea using four questions (goal, obstacles, fail state, actions), offers tips for generating original ideas, and walks through building a browser game in Manus without writing code.

New in n8n: Public API rate limits that keep every team on your instance running

n8n 2.47.0 introduces per-API-key public API rate limits, available on all plans for Cloud and self-hosted, with separate read and write limits across all endpoints and per endpoint. Clients receive 429 responses with Retry-After and rate-limit headers, and in multi-main setups every main shares one count per API key through Redis. Rate limiting is off by default and can be enabled via Settings > Policies > Public API Limits, the settings API, or environment variables, and works alongside idempotency keys (since 2.45.0) to make retries safe.

Taxonomy of 26 Ways AI Can Make Your Life Worse

A principal threat intelligence researcher presents a taxonomy of 26 consumer AI harms across five relationship types—chatbot to consumer, agent to consumer, threat actor to consumer, threat actor to agent, and threat actor to chatbot—organized around points of intervention rather than a direct harm-to-defense mapping. Each harm is traced through one or more of 19 intermediary layers, such as a foundation model, agent runtime, memory store, or consumer device, with defenses attached where they would actually have to be built. The author notes many harms can only be fixed by the companies making the models and running the products, and that the work in progress has thin evidence for some entries.

How HN: Agent-brain – Notes-agnostic second-brain operating model for AI agents

Agent-brain is a personal operating model for AI coding agents (Claude Code, OpenCode, Codex, Antigravity CLI) that provides a second-brain knowledge structure (journal, WIP, memory, tasks) built on top of any folder of notes, with Obsidian as one option rather than a requirement. It keeps runtime config and memory in a git-tracked brain symlinked into each runtime, adds a session lifecycle driven by the brain skill, and offers an opt-in boyscout skill for routing improvement opportunities to a remediation workflow or backlog. It installs via a curl bootstrap script that dry-runs by default and requires Python 3.x, git, and at least one supported agent runtime.

EU leaders seek to tackle challenges at first ever AI summit

Reuters reports that EU leaders are seeking to tackle challenges at the first ever AI summit. The article is dated November 13, 2026.

[AINews] Jensen Huang resigns after 33 years leading NVIDIA

NVIDIA announced that Jensen Huang has stepped down as CEO, president and board member after 33 years, with CFO Colette Kress serving as interim CEO while continuing as CFO, and Harvey Jones becoming the board's first chair. The board's acceptance is dated November 11 in the SEC filing, and the company announced the change on November 12, following reports of affairs and an independent directors' investigation announced November 8. The board has opened a search for a successor.

AI agents bring down wind farm within two hours

A Computer Weekly article reports that AI agents brought down a wind farm within two hours. The item links to the article and to a Hacker News discussion thread.

Is there a 'right amount' of AI?

The author argues that while avoiding AI entirely in the software one uses is impractical, indie developers should be warier of LLM-assisted coding than the community currently is, because LLMs have no understanding and delegating more to coding agents erodes developers' own comprehension of their code. He discusses the IGF jurists' refusal to vote for any game using generative AI, House House's disclosure that LLMs touched under a percent of a percent of Big Walk's code, and Marco Arment's enthusiastic use of coding agents in Overcast and Fineliner. He concludes that the least LLM-generated code you can use is always zero, and that thoughtfully crafted apps require a human developer with taste and understanding.

Witness – a voice-anchored evidence recorder for AI coding agents

Witness is a CLI tool that records your voice alongside the causal evidence of each moment—clicks, requests, console errors, DOM mutations, screenshots and full rrweb replay—while you use a managed browser, then packages it for AI coding agents. It runs three parallel streams (continuous audio with live transcription, per-click ~1.8s causal bursts, and page evolution via rrweb replay and snapshots), and exposes the session through a stdio MCP server with tools like list_segments, events_around and get_dom_state. Install with `npm i -g witness-recorder` (Node 18+ and system Edge/Chrome) and start via `wit start <url>`.

Where Are the "Killer Apps" of AI?

Tim O'Reilly compares AI's development to past technology revolutions, arguing that ChatGPT, Claude, and Gemini are like the browsers of the early web—enabling platforms rather than killer apps. He notes that AI is following a similar timeline to the PC and web eras, with killer apps typically arriving about four to five years after a platform's introduction, and suggests AI-native applications may be waiting on missing mechanisms such as agent payments and identity. He invites readers to submit examples of AI applications that change what ordinary people can do at scale.

Skills I'm Willing to Let AI Have

The author, who has dyslexic tendencies and often transposes letters between words, describes how LLMs read text as tokens and infer meaning from context, so typos no longer block understanding. They note they have stopped rereading and fixing their own mistakes, and worry that the habit of letting the model figure it out could spread to skills like organizing thoughts and structuring arguments. They decide to keep letting AI read through typos but to resume rereading and correcting their own writing, and ask readers which skills they are letting AI have.

Anthropic Says Its AI Model Cut AMD Chip Switch to Three Days

Anthropic told investors ahead of its proposed IPO that its Mythos model wrote or ported nearly all the necessary kernels to get Claude running on AMD MI355X GPUs in about three days with basically no human intervention. The roughly three-day timeline covers bringing up prompt processing for one model at a useful performance level, not a full migration of Claude's training and operations. Anthropic says its architecture spans Nvidia GPUs, AMD GPUs, AWS Trainium and Google TPUs, treated as a shared compute pool.

Introducing: Qualiaphobic AI – By Erik Hoel

Erik Hoel introduces a proposal for "qualiaphobic" artificial neural architectures that repel consciousness, analogous to hostile architecture, using an information asymmetry so that certainty about AI non-consciousness does not require solving the Hard Problem. He describes three approaches: distilling, transforming or ablating AI in ways adversarial to popular theories of consciousness; shifting architectures toward anchor points like lookup tables with known minimal potential for consciousness; and fragmenting any potential phenomenology. He argues that widely implemented qualiaphobic AI would address AI welfare and make humans the only moral patients, and he plans to construct and open-source the techniques.

Runitlocal.io – Run AI on your own hardware

Runitlocal.io is a site about running AI on your own hardware, presented by people who run it locally. It probes your GPU in the browser without sending that data to the site, and notes that speeds are estimates unless marked as measured; users can send in their own measured results.

OpenAI Joins Groundswell of Support for AI Chip Control Measures

A Bloomberg article reports that OpenAI has joined a growing group of supporters of measures to control AI chips. The item, shared on Hacker News, links to the Bloomberg article and its comment thread.

Stanford CS 329Z: Engineering AI Agents

Stanford's CS 329Z course covers engineering AI agents across foundations, LLM internals, RAG, tool use, frameworks, design patterns, memory, multi-agent systems, optimization, data, evaluation, safety, and coding and personal agents. It includes guest lectures on Terminal-Bench, SWE-bench, and a fireside chat with Andrew Ng, plus a final project demo day.

Coalition of Concerned AI Staff

A coalition of employees across frontier AI labs describes its efforts to help staff act on concerns about AI, including organizing cross-lab meetings and working groups, hosting seminars on topics such as surveillance and AI safety in China, and arranging briefings with policymakers and civil society. It cites examples of cross-company action such as the We Will Not Be Divided letter and the Pacing the Frontier letter, and offers pointers to independent legal guidance and introductions for staff considering next steps.

Fossil versions: a fingerprint of AI-written code in 250k NPM packages

A random sample of 250,000 npm packages found that the share of new packages whose first-release manifests pin mostly round, already-superseded versions (x.0.0) rose from 1–2.4% in every half-year from 2019 to early 2024 to 13.7% (95% CI 13.1–14.4) in the first half of 2026, falling to 7.6% in the second half of 2026. Median pin age at birth rose from about 155 days in 2021–2023 to 451 days in Q2 2026, and the share of packages born with essentially all pins on the newest release fell from about 28% to about 10%. The author argues the timing, the specific versions and how package managers work point to manifests written from a language model's memory rather than by running an install, while noting the data cannot prove who or what wrote any single package.

Three Lincoln Laboratory technologies receive 2026 R&D 100 Awards

MIT Lincoln Laboratory announced that three of its technologies are among the 2026 R&D 100 Awards recipients: the Magnetite secure-by-design operating system for embedded computers, the Quantum Microsystems wafer-scale platform integrating superconductor and semiconductor chips, and an algorithm that cancels 5G base-station interference on co-channel RF devices. The awards, managed by R&D World and judged by an independent panel, recognize 100 innovations transitioned to use or made available for license in the past year. Laboratory director Melissa Choi said the technologies address critical problems in cybersecurity, quantum computing, and communications.

Helping robots fit into the human workplace

Robust AI, co-founded by MIT Professor Emeritus Rodney Brooks and Anthony Jules, builds a mobile robot called Carter that is designed to work alongside humans in warehouses and factories, featuring a force-sensitive handlebar, an AI-powered vision system, and holonomic wheels. Carter is deployed in logistics warehouses and on factory floors to help workers pick or stock shelves and move items, with workflows that let it guide workers to locations, follow them, or transport loads across large facilities. The company says its approach is to understand workers' primary tasks and build features that accelerate them, letting humans do what they are good at while digital systems handle the rest.

New AI “copilot” helps engineers design better structures

MIT researchers developed a human-in-the-loop topology optimization system that presents regions of an algorithmic design to a human designer for feedback as it is generated, using machine learning trained on synthetic human-feedback data to identify likely regions of interest. In demonstrations, the approach improved manufacturability and linear buckling load by 39 percent while adding only 15 seconds to total design time compared to all-algorithmic approaches. The system currently only lets users make structures within identified regions thicker or thinner, and the researchers hope to add other modification options in future versions.

Show HN: I built a tool to check website access for 14 AI crawlers

FloxoLab's AI Crawler Access Checker takes a URL, reads its robots.txt under RFC 9309, requests the page with the user agents of 14 AI crawlers, and compares each response with a normal browser request to give a per-crawler verdict. It also reports page directives, text available without JavaScript, and discovery files such as /llms.txt and the sitemap, while noting limits like vendor IP verification and bot challenges returned with HTTP 200. The tool is free with no account and allows 8 checks per 10 minutes.

Post-AI Computer Science – Volume 5

In this blog post, Youngjin Kang argues that software goals should be designed to reproduce themselves through "semantic recursion," letting a minimal set of logical building blocks generate complex objectives. Using examples of God's declarations ("Let Alice be a human and Dinah be a cat"), he shows how the word "and" bundles multiple statements and how true statements persist across days via temporal continuity, with interruptions treated as a source of chaos. He closes by noting that definitions alone do not create narrative meaning, and that the next volume will address distinguishing good from bad to establish purpose.