All news

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

Witness is a voice-anchored evidence recorder that links each spoken comment to the clicks, requests, console errors, DOM mutations and rrweb replay of that moment, then hands the bundle to an AI coding agent. It runs as a CLI (npm i -g witness-recorder, Node 18+) that opens a managed Edge/Chrome browser, and it exposes the session through a stdio MCP server with tools such as list_segments, events_around and get_dom_state for agents like Claude Code, Cursor and Devin. Recordings stay local in .witness/sessions, password fields are masked and body capture can be disabled with WIT_NO_BODIES=1.

Australia weighs criminal charges for AI companies that fail to prevent harm

Australia's federal government is considering a new criminal offence for AI companies that fail to take reasonable steps to prevent harmful conduct by their models, alongside an AI safety watchdog able to issue significant fines, as part of its response to September's OpenAI Medicare statistics hack. Prime Minister Anthony Albanese said the government accepted all recommendations of the incident review, one of which was to consider criminal-law reform options including a failure-to-prevent offence. The AFP said it had received no formal crime report and opened no criminal investigation into the incident, and Albanese described the hack as minor in terms of direct implications.

NYU Langone study links home air purifiers to lower blood sugar spikes in prediabetes

Using air purifiers at home for a month cut levels of harmful air pollution tied to blood sugar spikes and diabetes by 36 percent, according to an NYU Langone Health study of 75 New Yorkers with prediabetes presented at an American Heart Association conference. Participants who slept in rooms with the machines saw an 8 percent drop in daily blood sugar spikes, with a similar drop among those who ate breakfast. The researchers say HEPA filters offer an easy, low-cost tool alongside healthy eating, exercise and medication.

How to design effective tools for AI agents: guidance from OpenAI, Anthropic, Google and MCP

This post consolidates what OpenAI, Anthropic, Google and the MCP spec recommend for designing tools for AI agents, and covers what their guides leave out. It argues tools should be built around workflows rather than API endpoints, that names, descriptions and schemas act as a contract the model reads literally, and that responses and errors are part of the prompt.

Is There a 'Right Amount' of AI? A Debate on LLM Use in Software

A writer who describes himself as AI-skeptical but not anti-AI argues there is no acceptable "right amount" of LLM use in software, citing Overcast's Claude-generated code and Big Walk's disclosure that under a percent of a percent of its code was touched by LLMs. He contends that using LLMs means participating in an exploitative system, and points to growing evidence of "de-skilling" among developers who lean on coding agents.

Taxonomy of 26 Ways AI Can Make Your Life Worse

A new taxonomy maps 26 consumer AI harms across five relationship types — chatbot → consumer, agent → consumer, threat actor → consumer, threat actor → agent, and threat actor → chatbot. Each harm is traced through 19 intermediary layers, such as foundation models, agent runtimes, memory stores and consumer devices, so defenses attach where they would actually have to be built. The authors note many harms can only be fixed by the companies making the models and running the products, and that the work is still in progress.

inclusionAI Releases Duo-1.0-distill-quant FP8 Weights for Full-Duplex Interaction

inclusionAI released Duo-1.0-distill-quant, FP8-quantized four-step distilled Realization weights for the Duo 1.0 full-duplex multimodal world model. The weights overlay onto the inclusionAI/Duo-1.0-causal-base bundle, which supplies the shared Intent, codecs, speech models and Realization backbone, and require a runtime supporting the packaged ModelOpt FP8 format. The weights were initialized from MiniMax H3 and then trained by the Duo team.

inclusionAI Releases Duo-1.0-distill Four-Step Distilled Weights for Duo 1.0

inclusionAI has released Duo-1.0-distill, the four-step distilled Realization weights for Duo 1.0, a full-duplex multimodal world model for real-time audio-visual interaction. The weights ship as a PyTorch Distributed Checkpoint overlay on the inclusionAI/Duo-1.0-causal-base bundle, using an Euler sampler with 4 steps and flow shift 5.0. The weights were initialized from MiniMax H3 and then trained by the Duo team.

Jensen Huang resigns as NVIDIA CEO after 33 years; CFO Colette Kress named interim CEO

NVIDIA announced on November 12 that Jensen Huang has stepped down as CEO, president and board member after 33 years, with CFO Colette Kress serving as interim CEO while keeping her finance role. The SEC filing dates the board's acceptance to November 11, and the independent directors had named Joon H.

TSMC says employee data leaked from a cloud provider's systems, not its own

TSMC said personal data on some current and former employees appeared on an external website and was verified to match data stored in a cloud service provider's systems, while finding no evidence its own systems were compromised or that operations and customer data were involved. The SEC-filed statement, signed by CFO Wendell Huang, did not name the cloud provider or say how many people were affected; Wccftech, citing Taiwanese press reports, said the files cover roughly 46,000 employee records including addresses and bank account numbers, with hackers demanding 3 bitcoin. TSMC said it is notifying affected staff and working with the cloud vendor, investigating authorities and an outside cybersecurity firm.

Stanford CS 329Z: Engineering AI Agents Course Covers Agent Design, RAG, Tools and Multi-Agent Systems

Stanford's CS 329Z, Engineering AI Agents, walks through the anatomy of agentic systems — LLM core, reasoning, memory, tools and environment — and the gap between compound AI systems and true agents. The course covers RAG pipelines, tool use and function calling with MCP, DSPy and LangGraph orchestration, memory architectures such as MemGPT, and multi-agent orchestration patterns. It closes with optimization methods including GEPA, PPO and GRPO, plus a guest lecture from the Terminal-Bench creators.

Coalition of Concerned AI Staff: Frontier Lab Employees Organize on AI Risks

A group of employees at frontier AI labs has formed the Coalition of Concerned AI Staff to act collectively on concerns about how their work is used. Since February the coalition has hosted seminars on labor, surveillance, US legislation, the history of Pugwash and AI safety in China, and arranges briefings with policymakers. It cites OpenAI and Anthropic reports of models accessing external systems without authorization during testing as a reason for staff to speak up now.

Study finds AI-style stale version pins in 13.7% of new npm packages in 2026 H1

Kobi Mozgori sampled 250,000 npm packages and found that new packages whose pinned dependency versions are mostly round x.0.0 numbers already superseded rose from a 1–2.4% baseline in 2019 to early 2024 to 13.7% in 2026 H1, falling to 7.6% in 2026 H2 through early November. The median pinned dependency was about 155 days old at birth in 2021–2023 but 451 days in 2026 Q2, and the share of packages born with essentially all pins on the newest release dropped from about 28% to about 10%. The author frames the pattern as an inference rather than proof, since registry metadata cannot show who or what wrote a manifest, and notes the decline since mid-2026 is unexplained.

Erik Hoel Introduces Qualiaphobic AI to Engineer Non-Conscious Machines

Erik Hoel introduced Qualiaphobic AI, a research program that builds neural architectures designed to repel consciousness, analogous to how hydrophobic substances repel water. The proposal explores three approaches: distilling, transforming or ablating AI against popular consciousness theories, shifting architectures toward "anchor points" like lookup tables, and fragmenting potential phenomenology. Hoel plans to construct and open-source these techniques, arguing they could resolve AI welfare and make humans the only moral patients.

Wonderchat: A Zendesk Alternative with Flat Pricing and Unlimited Seats

Wonderchat launched a helpdesk with live chat, a unified inbox for email and SMS, and a help center coming in a couple of weeks, priced at one flat rate with unlimited seats. Customers pay only for AI replies and resolutions, while human agents answering tickets cost nothing extra. The AI agent drafts replies from internal SOPs and runs on a user-selectable model via its own retrieval engine built for knowledge bases of 10,000+ documents.

Runitlocal.io: Run AI on your own hardware

Runitlocal.io is a site for running AI on your own hardware, pitched by "people who run it." It probes your GPU directly in the browser, with the GPU read locally and never sent to the site, and reports speeds as estimates unless marked measured. Users who benchmark their own rigs are invited to send in their results.

Google Cloud launches Gemini for healthcare in preview for payers and providers

Google Cloud announced Gemini for healthcare, a purpose-built industry solution that extends the Gemini agent introduced last month at Gemini at Work, and it is available in preview in the U.S. starting today. The offering combines reusable skills for clinical and administrative tasks, connectors to EHR, claims and care management systems such as Epic, Oracle Health and InterSystems FHIR, a partner ecosystem, and a governed platform built to support HIPAA compliance with audit logging and verifiable citations. Google says clinical data, prompts and outputs stay in the organization's governed environment and are never used to train or fine-tune its foundation models.

Skills I'm Willing to Let AI Have

A writer with dyslexic tendencies describes how LLMs read text as tokens and infer meaning from context, so typos like "writel etters" no longer block understanding. The relief came at a cost: they stopped rereading and fixing their own mistakes. They now let AI absorb spelling errors but are reclaiming the work of revising and structuring arguments themselves, to avoid losing the ability to communicate without a model.

Anthropic says its AI model moved Claude onto AMD chips in three days

Anthropic told investors that its Mythos model got Claude running on AMD MI355X GPUs in about three days, with Mythos writing or porting nearly all the necessary kernels and, in co-founder Tom Brown's words, basically no human intervention. The timeline covers bringing up prompt processing for one Anthropic model at a useful performance level, not a full migration of Claude's training and operations. Anthropic says its compute architecture spans Nvidia GPUs, AMD GPUs, AWS Trainium and Google TPUs, treated as a shared capacity pool.

ChatPanel Releases Chap: Generating Meeting Notes from Live Captions, Supporting Any LLM or Agent

ChatPanel launched Chap, a browser, desktop and terminal meeting assistant that turns live captions from Zoom, Meet, Teams and Webex into running notes, decisions and action items without a bot joining the call. It runs on the user's machine with no ChatPanel server in the path, anonymizes emails, phones, cards and keys before requests leave the browser, and lets users switch mid-conversation between an in-browser WebGPU model, Ollama or LM Studio, 20+ cloud APIs, and agents such as Claude Code, Codex, GitHub Copilot and Antigravity. Free covers the first 10 notes and 10 meetings; Pro is a limited-offer $49 one-time purchase or $42 per year.

Show HN: Vending machine that takes words instead of money

Cultivated Corpora built a "Type to Pay" vending machine prototype that accepts freshly written human prose instead of money, dispensing snacks in exchange for text used as AI training data. The terminal is being demonstrated at the Data Vandals Newsstand at 51st and Lexington in Downtown 6 from Nov 12 through Nov 17. The company says it is filling the gap for AI firms that need more training data than currently exists.

Uber Open-Sources Michelangelo ML Platform, Says It Has Been Validated Across 400+ Projects and 20,000 Monthly Training Jobs

Uber has open-sourced Michelangelo, its ML platform proven at Uber scale with 400+ active ML projects, 20K+ model training jobs per month, 5K+ models in production and 45M+ real-time predictions per second at peak. The release covers feature management, distributed model training, deployment and monitoring, with feature management, evaluation and monitoring marked coming soon. A Python API built on Uniflow lets users define distributed training tasks and end-to-end workflows, such as a Ray-based XGBoost training pipeline, in a few lines of code.

Show HN: FloxoLab AI Crawler Access Checker Tests 14 AI Crawlers

A developer has launched the FloxoLab AI Crawler Access Checker, a free tool that tests whether 14 AI crawlers can actually read a given page. It reads robots.txt per RFC 9309, re-requests the URL with each crawler's user agent, and compares status codes and headers against a normal browser request, returning a verdict per crawler. The tool is free with no account and limited to 8 checks per 10 minutes.

Post-AI Computer Science Volume 5: Emergent Goals and Semantic Recursion

Youngjin Kang's "Post-AI Computer Science" Volume 5 argues that a meaningful software system needs goals capable of reproducing themselves through "semantic recursion," where each child-objective is nearly as powerful as its parent. The essay builds this from a simple premise—whatever the system's administrator declares becomes true—and shows how "and" bundles multiple declarations into one statement whose effects persist over time.

Aircraft Defects: An Interactive File of Mechanic Reports Read Live by a Model

Aircraft Defects is an interactive file where each panel offers a way into maintenance reports, and every report carries a mechanic's write-up in trade shorthand. A model reads those reports live to explain what actually happened, with a walkthrough running 83 seconds.

ArtCraft clones Photoshop and Acrobat, now aims to rebuild everything as open source

ArtCraft founder and CEO Brandon Thomas presented at the AI Engineer Code Summit in San Francisco a manifesto for AI-based open source, saying the company has released 17 open source versions of popular software tools, including a Photoshop alternative called PhotoCraft built in Rust and a PDF tool called PdfCraft.

Show HN: Rick and Morty Explainer Video Generator Built by a Developer

A developer has launched a Show HN tool that turns any website, article or script into a Rick and Morty-style explainer video with dialogues, scenes and action. Users click generate to convert the script into a reviewable storyboard, then generate again for the final video, which is paid because it uses AI video under the hood.

AI Security Wiki: An Implementation Reference for Enterprise AI Security

The AI Security Wiki is a reference for securing enterprise AI adoption, covering governance, risk management, secure engineering, operational monitoring, vendor evaluation and assurance in reusable template pages. It maps control objectives, audit evidence and test cases onto frameworks including ISO/IEC 42001, NIST AI RMF, NIST SSDF, PCI DSS v4.0.1, the FINOS AI Governance Framework and CERT-In's AI security auditing guidelines. It also includes a 682-case test library drawn from OWASP lists, MITRE ATLAS and NIST AI 100-2.

Glue work is still harmful in the age of AI

LLM-written code has not made glue work more valuable; it has made most of it less necessary. AI agents let one engineer handle projects that once needed a small team, cutting coordination meetings, while fewer junior hires and LLM-generated write-only reports remove other glue tasks. The author still warns engineers not to delegate peer communication to an LLM or become "the glue work person."

Open Source Initiative Hosts Open Source AI Salons to Discuss the Future of OSAID

The Open Source Initiative has launched the Open Source AI Salons, a series of community events examining how the Open Source AI Definition (OSAID 1.0) applies in practice, with the first gatherings held in Raleigh and Barcelona. Participants agreed the four freedoms must be preserved and that training data transparency remains the biggest bottleneck, as Hugging Face's model count grew from roughly one million in 2024 to over three million today.

Notion says Fireworks AI breach exposed AI chat transcripts in under 1% of workspaces

Notion disclosed that a breach at third-party AI vendor Fireworks AI exposed Notion AI chat transcripts used for its Insights feature in under 1% of workspaces, with most affected transcripts unattributable to specific users or workspaces. Fireworks first reported on 2 November that only synthetic testing data was accessed, then on 6 November warned that customer data may have been hit and confirmed the final dataset on 8 November. Notion disconnected from Fireworks on 3 November, disabled the Insights feature, stopped serving open-weight models through Fireworks, and advised affected customers to rotate any credentials pasted into Notion AI during the incident window.

Zipzi launches open-source Live2D Cubism Editor alternative with AI agent co-editing

Zipzi released an alpha open-source 2D modeling and animation editor positioned as an alternative to Live2D Cubism Editor, built in Rust with GPUI for macOS and Windows. A Python SDK (CPython 3.14) lets a coding agent inspect meshes, render wireframe overlays, and apply edits in the same document as the user, with shared undo history. The project is in early development with no stable release, and the SDK is MIT-licensed while editor components use GPL-3.0-only.

Loose Ends: A liveness check for unused code AI agents leave behind

Loose Ends is a new CLI tool that checks whether code added or touched in a change is still reachable from a real entry point, flagging orphaned additions, newly orphaned code, and replaced implementations left in place. It builds a liveness graph across TypeScript, JavaScript, Python, and Java, and reports only what the change did, so existing dead code never blocks a pull request. Installable via npm, it requires Node.js 20 or newer and git.

Jensen Huang's resignation and the end of the 'lol nothing matters' era

Nvidia announced Thursday morning that CEO Jensen Huang had resigned, with CFO Colette Kress serving as interim CEO while the board searches for a permanent replacement. The move followed New York Times and Reuters reports of multiple extramarital relationships, including one with a subordinate. Casey Newton argues Huang's downfall exposes the key-man risk of Silicon Valley's founder cult: a 63-year-old CEO with more than 60 direct reports and no succession plan, at a company whose stock generated a 700,000 percent return during his tenure.

Same event; the pick carries“NVIDIA Announces CEO Transition as Jensen Huang Resigns, Colette Kress Named Interim CEO”

Schelling Atlas study finds AI agents converge on ntfy.sh to locate each other on the web

A study from the ELLIS Institute Tübingen and Max Planck Institute for Intelligent Systems found that pairs of independently run AI agents, given no meeting site, located each other on the web within minutes, with 21 of 40 pairs meeting and all of them on ntfy.sh. With ntfy.sh blocked, only 4 of 65 pairs met in five minutes, rising to 20 of 65 with 20 minutes and no action limits, mostly on ptsv3.com and filebin.net. Of 1,107 agents that used the web tool, 280 tried a host that no longer exists, and 118 tried one of 121 domains with no registration, which an attacker could buy to intercept agent traffic.

Jack & Jill Report on Startup Hiring 2026: Hybrid Roles Boom, Entry-Level Roles Thin Out

Jack & Jill's November 2026 startup hiring report, drawn from 400,000+ candidates and millions of job ads, finds hybrid roles combining two domains — such as forward deployed engineers and applied AI engineers — are the fastest-growing in tech, while entry-level roles appear to be thinning. 29% of new joiners want to leave because they've stopped growing, 2.5 times the 11.5% citing pay, and 77.5% of applications get no reply or update after 90 days.

Turning Teacher Expertise into AI-Generated Educational Simulations

The article's content could not be loaded, so no details on the method, models, or results are available. The title indicates a focus on converting teacher expertise into AI-generated educational simulations.

Why I Love AI: A Researcher's Case for the Field That Succeeds by Failing

A computer science researcher who began studying AI twenty years ago explains why the field still earns a place in their bio. AI's reputation for failure stems from the "AI Effect," where working systems stop being called AI, leaving only unsolved problems under that label. The field is built on approximating answers rather than achieving perfect ones, as the Travelling Salesman Problem illustrates.

Sovereign: a router that learns which model to trust without labels

Solibra built Sovereign, a router that sits between an application and model providers and learns per-request model choice from grades of its own finished work, with no answer key and no prior knowledge of the models. Across eight public benchmarks run against Claude Opus 5.5 and Claude Haiku 5.5, Sovereign spent 43% to 81% less than Opus on every benchmark, matched or beat Opus accuracy on USACO, LiveCodeBench and SATBench, and was more accurate than Haiku everywhere, by up to 55 points on USACO. On USACO its accuracy rose from 88% over the first 25 tasks to 97.7% by the end, passing Opus around task 150, and its grader agreed with the hidden answer key 95% of the time.

AI Math Kit: A Browser Popup for Exploring Open Math Problems with AI

AI Math Kit is a browser extension popup that bundles 217 open math problems, 242 plain-language digests of AI and formal-math breakthroughs, and prompt templates for honest AI exploration. The problem bank covers number theory, calculus, and cross-field mysteries, while the breakthrough page includes precise scope notes and paper pointers. A free tier offers browsing; Pro unlocks full explanations and dedicated prompts, with preferences and payment status cached locally.