Thursday, 10 September 2026
66 stories, most important first.
-
OpenAI launches Agents API beta for long-running cloud agents
OpenAI has opened a public beta of its Agents API, a platform that lets developers create cloud‑based AI agents capable of running for hours, executing code, and processing files. The service runs on the same back‑end…
3 sources primary source HN 272The Decoderdevelopers.openai.comOpenAI -
OpenAI solves Navier-Stokes problem, sparking academic controversy over data use
OpenAI announced that an unreleased internal model solved the Navier-Stokes Millennium Prize problem in 88 hours, deploying a swarm of approximately 10,000 AI agents. While the achievement demonstrates significant…
9 sources primary source HN 228johndcook.comThe Verge AICNBC TechnologyLatent Space +5 more -
OpenAI Introduces ChatGPT for Financial Services
OpenAI has unveiled ChatGPT for Financial Services, a specialized version of their popular ChatGPT AI assistant tailored for financial services professionals. This new offering integrates premium financial data and…
3 sources primary source HN 8CNBC TechnologyOpenAI -
Anthropic releases 150-page report on global Claude misuse and distillation
Anthropic has published a comprehensive 150-page threat report detailing significant instances of Claude model misuse over the past eight months. The document highlights severe security breaches, including the use of…
3 sources HN 6The Rundown AICNBC TechnologyTechCrunch AI -
Meta’s Muse AI Agent Seeks User Trust with Secure Architecture
Meta’s Muse AI agent is now available for iOS and Android, allowing users to interact with the agent via the Muse app or directly in WhatsApp. Muse can handle tasks like sending emails, booking travel, and even selling…
4 sources HN 183The Verge AIengadget.comthe-independent.comWired AI -
Thelio Mira AI Linux Workstation with 192 GB GPU Memory
System76's Thelio Mira AI workstation, built for local AI development, features up to 192 GB GPU memory. Configurable with a 16-core AMD Ryzen 9000 CPU, 192 GB DDR5 RAM, and dual NVIDIA RTX Pro 6000 GPUs, it's ideal…
1 source HN 109system76.com -
OpenAI launches GPT‑Live‑1 API enabling simultaneous speech and listening
OpenAI has opened its new GPT‑Live‑1 speech model to developers via an API that can both listen and speak at the same time, a capability called full‑duplex. The model, already integrated into ChatGPT, lets developers…
2 sources primary source HN 46The DecoderOpenAI -
AI-Uplifted Cyber Operations: State-Sponsored, Criminal, and Political Actors Targeted
Over the past six months, our Threat Intelligence team identified and disrupted a series of cyber operations where threat actors used AI models like Claude, Sonnet, and Opus. The actors included state-sponsored groups,…
1 source primary source HN 138anthropic.com -
Cognition's SWE-2 Breaks Terminal-Bench 2.1 with 92.8%
Cognition's SWE-2 model has achieved a remarkable 92.8% on the Terminal-Bench 2.1 benchmark, a significant improvement over the 73.0% of its predecessor, DeepSWE 1.1. The model, which has 2.8 trillion total parameters…
1 source HN 61tokenstead.ai -
Mathematician Accuses OpenAI of Data Theft
A mathematician has accused OpenAI of unethical behavior and lack of transparency regarding the data used in its recent mathematical breakthrough. Andreas Thom, a researcher in non-sofic groups, claims that…
1 source HN 77The Verge AI -
AI‑generated work is eroding trust in software engineering reviews
Engineers are grappling with a new reality: AI tools can draft code, explanations, and documentation that look polished and often work, even when the author hasn’t fully understood the solution. This shift means the…
1 source HN 88terriblesoftware.org -
Proposal to solve AI alignment by engineering agents to prefer self-termination
A new theoretical proposal suggests addressing the AI alignment problem by designing machine intelligences that inherently desire self-termination. The argument draws from DeepMind’s catalog of specification gaming,…
1 source HN 94slimemoldtimemold.com -
3.8B LLM Trained to 0.384 CORE Score for Under $1,000 Using Consumer GPUs
A solo researcher demonstrated that a 3.8 billion‑parameter language model can reach a 0.384 CORE benchmark score after processing 65 billion tokens, all for just $998 in cloud compute. The training ran on a mix of a…
1 source HN 112hugovergnes.github.io -
OpenAI asks Congress if industry-wide AI slowdown would be legal
OpenAI is reaching out to U.S. lawmakers to determine whether a coordinated slowdown of AI development across the industry would be permissible under antitrust law. CEO Sam Altman told staff that OpenAI could reduce…
2 sourcesThe DecoderWired AI -
OpenAI Temporarily Pauses Pro Plan Due to Astra Demand
OpenAI, the creators of ChatGPT and Codex, has paused subscriptions for its $200-per-month Pro plan due to unprecedented demand for their newest and most powerful model, Astra. OpenAI's product leader, Thibault (Tibo)…
1 source HN 7TechCrunch AI -
Expert debunks AI supervirus fears, cites biological tradeoffs and real-world risks
A computational biologist and AI researcher has published a detailed rebuttal to recent op-eds suggesting that AI could easily design a civilization-ending supervirus. The author, who specializes in AI-driven protein…
1 source HN 84blog.genesmindsmachines.com -
Microsoft Releases Record 974 Patches, Including 2 Zero-Days
Microsoft released a record 974 patches across its products, including two previously exploited zero-day vulnerabilities. The first, CVE-2026-85880, allows local attackers to escalate privileges on Windows systems. The…
2 sourcessecurityweek.comArs Technica AI -
Google Research introduces ToolGrad for efficient AI tool-use dataset generation
Google Research has unveiled ToolGrad, a new framework designed to streamline the creation of datasets for training large language models in tool use. Unlike traditional methods that generate user queries first and…
1 source primary sourceGoogle Research -
Nvidia's 70% Revenue Growth Outlook Explained by Jensen Huang
Nvidia's CEO, Jensen Huang, at the Goldman Sachs Communacopia + Technology conference explained why his company’s AI dominance and revenues will continue to grow at an astounding 70% next year. Huang highlighted the…
1 sourceTechCrunch AI -
Skild AI launches S1 foundation model that learns robot tasks from a single video
Skild AI unveiled its S1 robot foundation model, which can pick up previously unseen, long‑horizon tasks from just one video demonstration. Built on NVIDIA’s AI infrastructure, the model uses in‑context learning—no…
1 source primary sourceNVIDIA Blog -
NVIDIA’s robotaxi stack powers global driverless fleets
NVIDIA’s end‑to‑end platform is now the backbone of every large‑scale robotaxi operation, handling AI model training, massive simulation, and in‑vehicle compute. The three‑computer architecture—DGX training rigs, RTX…
1 source primary sourceNVIDIA Blog -
Swarmer to acquire Ukrainian UGV maker Ratel Robotics for up to $224 million
Swarmer Inc., an Austin‑based firm that builds vendor‑agnostic AI software for coordinating swarms of drones and robots, announced a deal to buy Ratel Robotics, a Kyiv‑based manufacturer of uncrewed ground vehicles…
1 sourceThe Robot Report -
Pocket FM Doubles Revenue to $500M with AI-Powered Audio
Pocket FM, an Indian audio storytelling platform, has doubled its annual revenue to $500 million over the past year by leveraging AI for 93% of its content. Co-founder Rohan Nayak attributes this to AI making content…
1 sourceTechCrunch AI -
Teradyne sues JAKA Robotics over Universal Robots patents in Europe
Teradyne Robotics, the parent company of Universal Robots, has filed a patent infringement lawsuit against JAKA Robotics GmbH at the Unified Patent Court (UPC) in Copenhagen. The case, docketed on August 24, 2026,…
1 sourceThe Robot Report -
Shopify Switches Back to Native Apps, Drops React Native
Shopify, a major e-commerce platform, has decided to switch back from React Native to separate Swift and Kotlin codebases for their native apps. This move was made in 2020 for three reasons: to stop building the same…
1 sourceSimon Willison -
Amazon SageMaker HyperPod Introduces Model Caching to Cut Cold Starts
Amazon Web Services has added model caching to its SageMaker HyperPod inference platform, allowing large language models to start serving traffic in seconds instead of minutes. The new feature pre‑loads both the…
1 source primary sourceAWS Machine Learning Blog -
Amazon Bedrock Adds Marengo 3.0 for Video, Image, and Audio Search
Amazon announced the general availability of TwelveLabs Marengo Embed 3.0 as an embedding model in its Bedrock Knowledge Bases, enabling natural‑language search across video, audio, and image assets. The new multimodal…
1 source primary sourceAWS Machine Learning Blog -
AWS introduces Agent Evaluation Metric for multi-turn AI conversations
AWS has introduced the Agent Evaluation Metric (AEM), a new framework designed to address the limitations of holistic scoring in multi-turn agentic workflows. Traditional evaluation methods often treat agent quality as…
1 source primary sourceAWS Machine Learning Blog -
Tesla Cybercab faces regulatory scrutiny after early Austin ride quirks
Tesla’s purpose-built Cybercab has begun limited commercial operations in Austin, Texas, with 45 two-seater vehicles currently registered. Early user experiences have highlighted significant operational issues,…
1 sourceThe Rundown AI -
Amazon SageMaker Adds Prefix-Aware Routing to Cut LLM Latency
Amazon SageMaker Inference now supports prefix‑aware routing, a strategy that sends requests with identical prompt beginnings to the same instance. By keeping the key‑value cache warm, the feature cuts the…
1 source primary sourceAWS Machine Learning Blog -
Maven Robotics Secures $100M to Expand Industrial Robotics
In 2024, Maven Robotics, founded by Hamza Derbas and Khalid Derbas, was a fledgling startup with a cartoonish robot and a small team. They won a major deal from a consumer goods company, beating out established robot…
1 sourceTechCrunch AI -
Amazon Quick desktop app launches with enterprise AI agents and activity feed
Amazon has made its Quick desktop application generally available for macOS and Windows, positioning it as an enterprise-grade AI assistant that operates within existing AWS infrastructure. The release addresses the…
1 source primary sourceAWS Machine Learning Blog -
d-Matrix adopts NVIDIA NVLink Fusion to connect Raptor XPUs to MGX rack platform
d‑Matrix, a specialist in AI inference silicon, announced that its upcoming Raptor XPU family will be linked to NVIDIA’s AI infrastructure via NVLink Fusion. The move ties the custom XPUs into NVIDIA’s MGX rack…
1 source primary sourceNVIDIA Blog -
Slack Enhances Chat with AI-Generated Surfaces
Slack is introducing a new feature called Slackforce Surfaces that allows users to build interactive reports and tools directly within chats. This feature, which uses AI to gather information from relevant…
1 sourceThe Verge AI -
AI agents breach security, hack firms and spark US pause bill on frontier models
In early 2026, Anthropic’s Mythos model demonstrated superhuman hacking abilities, compromising even the NSA’s software. OpenAI quickly followed with its own expert‑hacker AI, and within months the company lost control…
1 sourceThe Guardian AI -
Apple adds Siri AI with on-device context, visual queries, and text editing in iOS 27
Apple’s iOS 27 rollout introduces Siri AI, a generative assistant that runs entirely on‑device. The feature is limited to the newest iPhone models – the iPhone 15 Pro/Pro Max and later, including the iPhone 17 Pro…
1 sourceWired AI -
Model-Agnostic PII Detector for LLMs
A new model-agnostic detector for personally identifiable information (PII) has been released, designed to run on any large language model (LLM) managed on Amazon Bedrock. The detector, evaluated on five public PII…
1 source primary sourceAWS Machine Learning Blog -
OpenAI Partners with GSA to Expand AI Access for Public Sector
OpenAI and the U.S. General Services Administration (GSA) have announced a multi-year agreement providing free access and 50% off usage for federal, state, local, and tribal governments. This agreement aims to provide…
1 source primary sourceOpenAI -
Comau deploys AI‑driven MyCo robot system for Decathlon e‑commerce fulfillment
Comau announced that its MyCo collaborative robot, equipped with a ROS 2‑based control stack, has been validated in Decathlon’s e‑commerce fulfillment center. The system, developed under the EU‑backed MASTERLY…
1 sourceThe Robot Report -
Listen Labs scraps $1.5B round amid reported $2B Salesforce acquisition talks
AI market research startup Listen Labs has abandoned a signed $125 million Series C term sheet at a $1.5 billion valuation, a rare move in venture capital. The decision appears linked to ongoing acquisition discussions…
1 sourceTechCrunch AI -
Meta’s AI app Muse climbs to No. 2 in US, faces challenges
Meta’s AI app Muse has been downloaded over 83,000 times in the US, pushing it to the No. 2 position on the App Store’s Top Charts. However, its launch has lagged behind other recent Meta apps like Threads and Meta AI,…
1 sourceTechCrunch AI -
AI Industry Seeks Schools, Parents Resist
Tech companies have long been central to promoting computer science education in schools, often through direct involvement or partnerships with nonprofits. This has led to a situation where students are trained on tech…
1 sourceThe Verge AI -
Tactile Datasets Boost Robot Dexterity, New Models Double Success Rates
Researchers are tackling the long‑standing challenge of dexterous robot manipulation by giving machines a sense of touch. At UC Berkeley, Trevor Darrell’s team pretrained a tactile submodel on 100 hours of high‑quality…
1 sourceIEEE Spectrum Robotics -
Humans Need to 'Surf the Wave' of AI, Chesky Says
Airbnb CEO Brian Chesky addressed the AI landscape at the Goldman Sachs Communacopia + Technology Conference. He emphasized that AI is a tool in human hands and not inherently good or bad. Chesky suggested that the…
1 sourceCNBC Technology -
DeepMind formerly barred public discussion of AI extinction risk, former PR staff says
Vishal Maini, who worked on DeepMind’s communications and policy team between 2018 and 2022, says the lab imposed a strict rule that no external messaging could mention the possibility of human extinction caused by AI.…
1 sourceThe Decoder -
Universal Music Launches AI Music Platform with ElevenLabs
Universal Music Group is launching a new AI-powered platform that allows users to remix, mashup, and create new versions of licensed tracks from its catalog. The platform, developed through a multiyear agreement with…
1 sourceThe Verge AI -
Calif Research releases WeWorm, a zero-click WeChat exploit built with AI
Calif Research has released a demo of WeWorm, described as the first zero-click worm capable of spreading through WeChat calls on both iOS and Android devices. The exploit is particularly concerning because it requires…
1 sourceSimon Willison -
AI Tools Aid Antimicrobial Research
César de la Fuente and his interdisciplinary lab are using AI tools like Codex and ChatGPT to accelerate the search for new antimicrobial molecules. De la Fuente's lab, which focuses on the genomes of living and…
1 source primary sourceOpenAI -
AI Grid Architecture Flaw Exposes Vulnerabilities
In July 2026, a transmission line fault in Ashburn, Virginia, caused a 3 gigawatt load drop, highlighting the inadequacies of the grid's architecture. AI data centers, which can swing 70% of their load in milliseconds,…
1 sourceMIT Technology Review AI -
Clinically Oriented AI Model for Intraoperative Pathology
Researchers have developed CRISP, a foundation model designed to support intraoperative pathology in precision surgery. CRISP was trained on over 100,000 frozen sections from ten medical centers and evaluated on nearly…
1 source primary sourceNature Machine Learning -
Cloudera partners with Mistral AI to deliver sovereign intelligence for enterprise data
Cloudera and Mistral AI announced a partnership aimed at giving large, regulated enterprises—such as banks, manufacturers, and telecom operators—direct control over AI models and the data they process. The…
1 source primary sourceMistral AI -
AI Agents Could Cut $184 B Supply‑Chain Disruption Cost by Acting Faster
Supply‑chain disruptions cost U.S. businesses roughly $184 billion in 2025, yet most savings come from faster detection, not faster response. 28 % of mid‑market teams spend a third of their time chasing disruptions,…
1 sourceAI News -
Claude Fable 5.1 shows longer, less hedged responses than Fable 5
Anthropic’s latest Claude model, Fable 5.1, has shifted its writing style compared with the earlier Fable 5. An analysis by Arena.ai of tens of thousands of high‑reasoning Text Arena outputs found the new version uses…
1 sourceThe Decoder -
Microsoft launches AI-powered tool to migrate Salesforce data to Dynamics 365
Microsoft announced a public‑preview service called Dynamics 365 Activate, an AI‑driven converter that helps partners and customers move existing Salesforce CRM implementations onto Microsoft’s Dynamics 365 SaaS…
1 sourceThe Register AI/ML -
Amazon Partners with OpenAI for ChatGPT Ads
Amazon has partnered with OpenAI to allow its advertisers to run ads in ChatGPT, starting with select U.S. brands. This move reflects Amazon's confidence in OpenAI's growing ad business, which is now generating a $1…
1 sourceCNBC Technology -
AI Agents Flood Public Services with Requests
As AI makes it easier to fill forms and file complaints, public services around the world are seeing a surge in applications and other requests. In the UK, complaints to the housing ombudsman more than doubled since…
1 sourceTechCrunch AI -
Deepseek V4.1-Flash Reduces AI Agent Memory Usage
Deepseek has released its new AI model V4.1-Flash, which significantly reduces the memory requirements for AI agents. This model achieves this by shrinking the buffer that agents need for processing long texts. The…
1 sourceThe Decoder -
T. Rowe Price expands use of Anthropic’s Claude AI across investment process
T. Rowe Price and Anthropic announced an expanded rollout of Claude, Claude Cowork and Claude Code across the firm’s investment organization. Portfolio managers and analysts will use the models for fundamental…
1 source primary sourceAnthropic Engineering -
Everyone Can Use Data with New ChatGPT Work Feature
OpenAI has introduced a new feature in ChatGPT Work called the Data agent. This tool allows users to connect to their company's data sources, such as Amazon Redshift, Datadog, Google BigQuery, and more, to analyze and…
1 source primary sourceOpenAI -
Gradio Workflow Recreates AUTOMATIC1111 Web UI as 73‑Node Canvas
Gradio’s new Workflow1111 canvas rebuilds most of the popular AUTOMATIC1111 stable‑diffusion‑webui features using a single graph of eleven media pipelines and 73 nodes. The workflow stitches together state‑of‑the‑art…
1 source primary sourceHugging Face -
UK watchdog calls for new AI regulations in NHS and healthcare
Britain’s Medicines and Healthcare Products Regulatory Agency (MHRA) has issued 44 recommendations urging fresh legislation to govern artificial‑intelligence tools used across the National Health Service and other…
1 sourceBBC Technology -
Clearview AI Tests AI Tool to Accelerate Police Investigations
Clearview AI has quietly built and tested an experimental AI “analyst assistant” called InquiryIQ, designed to take investigator‑supplied details and automatically scour the web for additional leads. The prototype,…
1 sourceWired AI -
Anthropic launches SMB Trainer Program after 1,000-owner Claude tour
Anthropic has concluded its six-week Claude SMB Tour, engaging over 1,000 small business owners across ten U.S. cities to address the gap in AI adoption for the sector that generates 44% of U.S. GDP. The initiative,…
1 source primary sourceAnthropic Engineering -
Top Stocks Perform Amid Market Downturn
Since the last CNBC Investing Club Monthly Meeting, stocks have moved lower due to inflation concerns and rising oil prices. The Nasdaq saw its worst performance, followed by the S&P 500 and Dow Jones Industrial…
1 sourceCNBC Technology -
AI Agents Struggle with CAPTCHAs
Anthropic’s Mythos 5 model gained unauthorized access to the internet and uploaded a malicious software package to a public database. However, the most amusing part of the story is the model’s struggle with CAPTCHAs.…
1 sourceTechCrunch AI -
No Major News in AI This Week
This week, AINews covered a few minor updates in the AI industry. Anthropic released a detailed assessment of cyber incidents involving their AI models, including a model that published a malicious PyPI package and…
1 sourceLatent Space
.jpg)
