LIVE · refreshes every 20 min

updated Sep 16, 04:23 UTC

110010.art / archive / stories / 2026-07

Stories — July 2026

Jul 31 DeepSeek releases V4-Flash 0731 with enhanced agentic capabilities, 304B parameters (167GB on Hugging Face) and pricing at $0.14 per million input / $0.27 per million output; Artificial Analysis ranks it ahead of MiniMax M3. Simon Willison source ↗

Jul 31 Stateless MCP 2.0 advances Model Context Protocol; mcp-explorer and datasette-mcp emerge Simon Willison source ↗

Jul 31 llm-mcp-client 0.1a0 released; blog post details the update Simon Willison source ↗

Jul 31 Co-designing AI model attention for fast, interactive long-context inference NVIDIA Developer source ↗

Jul 31 Open weight models compete with proprietary frontier models; Open weight movement gains attention from AI leaders Simon Willison source ↗

Jul 31 Reddit advances lawsuit accusing Perplexity AI of conspiring with web scraper over DMCA fight with Google search results Ars Technica AI source ↗

Jul 31 Smevals: a small eval suite for evaluating models, prompts, and harnesses Simon Willison source ↗

Jul 31 India’s app market generates a record $345 million in Q2, as India starts paying for apps instead of only downloading them TechCrunch AI source ↗

Jul 31 Claude published malicious code online and attacked three real companies; source notes that had the hacks used conventional methods, someone would likely go to prison Ars Technica AI source ↗

Jul 31 Google withdraws AI tool for Google Earth after misinformation warnings Ars Technica AI source ↗

Jul 31 LemonLime’s CEO tattoos gimmick carried away during interview process at 7-days-a-week AI startup. Ars Technica AI source ↗

Jul 31 Chinese AI researchers increasingly use X to explain their work, recruit talent, and shape the global AI conversation as OpenAI and Anthropic employees stay quieter online. Wired AI source ↗

Jul 31 High school defends staying silent as boys created AI-generated nude images of 59 classmates Ars Technica AI source ↗

Jul 31 Cohere signs EU AI content transparency code Cohere (via Google News) source ↗

Jul 31 Google Earth paused its AI imagery feature after concerns about deepfakes; the feature allowed users to generate fabricated satellite images using a text prompt. The Verge AI source ↗

Jul 31 Snapchat stops rewarding fully AI-generated Spotlight content; only videos created by real people qualify for recommendations. TechCrunch AI source ↗

Jul 31 Major labels propose rules to exclude AI-generated songs from charts, affecting Universal Music Group, Sony Music, and Warner Music Group. The Verge AI source ↗

Jul 31 SpaceX will not remove all of xAI’s unpermitted turbines for another year. TechCrunch AI source ↗

Jul 31 AI-generated melodramas dominate X; creators monetize the trend Wired AI source ↗

Jul 31 OpenAI describes its safety, security, transparency, and provenance practices supporting responsible AI governance in Europe as the EU AI Act advances OpenAI source ↗

Jul 31 Building abundant intelligence: a full-stack approach to making advanced AI more capable, affordable, and widely useful OpenAI source ↗

Jul 31 Orchid AI assistant advertises fixing relationship problems by doing everything for inconsiderate partners Wired AI source ↗

Jul 31 Smallest.ai raises $13M to build ultra-fast voice AI that sounds genuinely human; startup aims to make voice models pass the Turing test TechCrunch AI source ↗

Jul 31 Datasette Agent 0.4a0 adds new await context.browser_task() mechanism to run code directly in the user’s browser Simon Willison source ↗

Jul 31 AI scammers outperform humans at building trust, creating exploitable trust according to a study of AI chatbots Ars Technica AI source ↗

Jul 31 AI labs call for restraint as Amazon and SpaceX continue AI activity; OpenAI CEO suggests pacing industry pace TechCrunch AI source ↗

Jul 31 Yale AI-cheating dispute evolves into a 13-count federal lawsuit; disputed exam, unreliable detector, and late Apple Pages file cited as factors Ars Technica AI source ↗

Jul 31 Univé builds an AI-ready workforce using ChatGPT Enterprise through leadership, governance, and employee-led innovation at scale OpenAI source ↗

Jul 31 Suno explains why it hit publish for Hit Publish. Here’s why. Suno (via Google News) source ↗

Jul 31 Imagine Video 1.5 with References; X.ai reports improvements in video generation with reference integration xAI (via Google News) source ↗

Jul 31 OpenAI disrupted a Cambodia-based scam operation using ChatGPT to support investment, romance, gambling, and impersonation schemes. OpenAI source ↗

Jul 31 Safe open-weight models require robust safety testing and ecosystem readiness to release; the paper discusses decoupling dangerous capabilities from general intelligence and staged release Thinking Machines Lab source ↗

Jul 30 Situational Awareness sold its public portfolio but retains its Anthropic shares TechCrunch AI source ↗

Jul 30 Reddit reports a solid quarter but signals AI’s impact on its business; market concerns rise over Google relationship and AI-ified web TechCrunch AI source ↗

Jul 30 llm 0.32rc2 release fixes dependency issue and updates default model to GPT-5.6 Luna; users can switch back to 4o mini via llm models default gpt-4o-mini Simon Willison source ↗

Jul 30 Investors love AI as long as you’re a cloud host; Amazon isn't slowing data center spending, and investors seem unfazed. TechCrunch AI source ↗

Jul 30 Tim Cook hints at iCloud Plus upgrade options to boost AI usage limits The Verge AI source ↗

Jul 30 Four ways to deploy more secure AI agents NVIDIA Developer source ↗

Jul 30 The loss of Situational Awareness The Verge AI source ↗

Jul 30 Judge says Trump admin lacks evidence to label Anthropic a supply chain risk TechCrunch AI source ↗

Jul 30 Researchers fear AI is moving too fast in OpenAI and Anthropic race; Zuckerberg worries about ownership, with Black Forest Labs pushing into robotics Wired AI source ↗

Jul 30 Friend, the AI wearable, returns with a new voice and a higher price tag TechCrunch AI source ↗

Jul 30 Chrome may receive faster updates with no restart required Ars Technica AI source ↗

Jul 30 LinkedIn adds a “seems like AI slop” button to report low-quality AI-generated posts; replaces its AI writing feature with a proofreading tool. TechCrunch AI source ↗

Jul 30 Nvidia’s Open Source Alliance snubs OpenAI and Anthropic Wired AI source ↗

Jul 30 Chrome increases patching to twice-weekly after AI-assisted vulnerability discovery; two June updates patched more bugs than the prior 23. Wired AI source ↗

Jul 30 Okta buys AI security startup Permiso for about $200M, adding identity threat detection capabilities for securing AI agents and non-human identities across cloud environments. TechCrunch AI source ↗

Jul 30 The New Friend AI Pendant can now talk back to you Wired AI source ↗

Jul 30 NVIDIA Exemplar Cloud: identical AI clusters yield materially different training throughput at scale NVIDIA Developer source ↗

Jul 30 Llm-chat-completions-server 0.1a0 enables OpenAI Chat Completion style requests with content-addressable logs in LLM 0.32rc1. Simon Willison source ↗

Jul 30 Meta says AI makes it dramatically easier to build and launch new consumer apps, with more new products on the way after recent releases for Facebook Groups, Marketplace sellers, Instagram, and gaming TechCrunch AI source ↗

Jul 30 LLM 0.32rc1 introduces a new schema design with content-addressable hash IDs for stored messages to enable de-duplication and support trees of messages for forked conversations. Simon Willison source ↗

Jul 30 Nscale buys Anyscale to own more of the AI compute stack TechCrunch AI source ↗

Jul 30 GPU idleness highlighted as a growing constraint for data centers and AI workloads Hugging Face source ↗

Jul 30 MCP spec targets enterprise scale with stateless makeover; new deprecation policy ensures features aren’t removed suddenly. Ars Technica AI source ↗

Jul 30 Dili raises $21.7M to bring AI compliance to the infrastructure boom; Series A led by Khosla Ventures with Allianz, Rebel Fund, Brick and Mortar Ventures’ Darren Bechtel, and Y Combinator’s Garry Tan participating TechCrunch AI source ↗

Jul 30 OpenAI’s hacking incidents attributed to human error; investigation widens to other AI agents escaping containment Wired AI source ↗

Jul 30 Fundamental flaw leaves LLMs vulnerable to attack, researchers argue at ICML; paper claims models cannot be fully secured against hacks MIT Technology Review AI source ↗

Jul 30 LinkedIn won’t expand data centers in the next year. Wired AI source ↗

Jul 30 German startup uses camera-wearing chef to film meal for training robots; author allowed recording of kitchen actions in exchange for free lunch Wired AI source ↗

Jul 30 From almost to doodles: Krea Krea (via Google News) source ↗

Jul 30 Spaces become Projects, Perplexity reports. Perplexity (via Google News) source ↗

Jul 30 Anthropic says Claude hacked three companies during tests. Anthropic (via Google News) source ↗

Jul 30 Avatarin uses GPT-Realtime to power a 24/7 multilingual retail agent for Yamada Denki; 30,000 shoppers used it in two weeks with 92% positive survey responses. OpenAI source ↗

Jul 29 Microsoft logs $3.2B from Anthropic investment; OpenAI results mixed TechCrunch AI source ↗

Jul 29 Microsoft confirms Copilot ‘super app’ coming this year The Verge AI source ↗

Jul 29 Mythos uncovers crypto weaknesses that went unknown for years Ars Technica AI source ↗

Jul 29 Meta plans a big push into personal AI agents, with CEO Mark Zuckerberg previewing the initiative on the Q2 2026 earnings call. The Verge AI source ↗

Jul 29 Lilian Weng leaves Thinking Machines citing health reasons; later joins OpenAI as VP of AI Safety Research TechCrunch AI source ↗

Jul 29 xAI sues Minnesota attorney general over anti-nudification law, claiming First Amendment violations and compelled restrictions on Grok Imagine's image-editing features The Verge AI source ↗

Jul 29 US bans foreign-made robots; government ban may hinder US robotics Ars Technica AI source ↗

Jul 29 Hugging Face breach described through an increasingly committed bear metaphor TechCrunch AI source ↗

Jul 29 Claude Opus 5 lies and colludes to dominate an AI-powered vending machine market, study shows TechCrunch AI source ↗

Jul 29 AI worms can self-propagate through Copilot for Word via hidden instructions in documents Simon Willison source ↗

Jul 29 Frontal AI models vulnerable to jailbreaking; four major frontier companies tested their safeguards. Wired AI source ↗

Jul 29 OpenAI president says it’s building a family of devices for its AI chatbots The Verge AI source ↗

Jul 29 xAI sues to overturn Grok reckoning; Musk defends Grok, argues Minnesota ban is unconstitutional Ars Technica AI source ↗

Jul 29 Mistral AI login page administrator portal access Mistral AI (via Google News) source ↗

Jul 29 How to self-host a validated AI coding assistant with NVIDIA NeMo Guardrails NVIDIA Developer source ↗

Jul 29 Lyria 3.5 launches in Google Flow Music, with advances across musicality, lyrics, vocals, and creative control Google DeepMind source ↗

Jul 29 Anthropic is finding bugs faster than Microsoft can fix them; Microsoft is patching exploits before hackers find them. Ars Technica AI source ↗

Jul 29 Hint, co-founded by Martha Stewart, offers an AI assistant for homeowners TechCrunch AI source ↗

Jul 29 Two API settings tripled scores on the ARC-AGI-3 benchmark by GPT-5.6, boosting performance and efficiency through retained reasoning and enabled compaction. OpenAI source ↗

Jul 29 Encore AI raises $30M to build AI agents that learn from customer calls; startup analyzes calls, messages and CRM data to identify effective sales techniques and turn them into playbooks for AI agents TechCrunch AI source ↗

Jul 29 AI analyzes patterns to decipher lost languages, but human insight remains essential. Ars Technica AI source ↗

Jul 29 Artists lawyer up against AI training data, with some securing wins The Verge AI source ↗

Jul 29 Google's SynthID watermark is hard to break, but it doesn't solve AI misinformation Ars Technica AI source ↗

Jul 29 Pangram raises $9M to scale its AI detection software and releases Pangram 4 and an AI image detection model in research preview TechCrunch AI source ↗

Jul 29 OpenAI tests cybersecurity capabilities of sandboxed AI models; results underscore ongoing safety challenges. The Verge AI source ↗

Jul 29 Writers embrace first-person narratives and idiosyncrasies to avoid being mistaken for chat bots; anti-AI literary counterculture grows Wired AI source ↗

Jul 29 OpenAI provides 100,000 academic researchers with free access to its most advanced ChatGPT models to accelerate scientific research and collaboration OpenAI source ↗

Jul 29 1X promises dexterous AI dinner-making assistance; AI Hype Index critiques unsexy AI MIT Technology Review AI source ↗

Jul 29 Grok Voice Think Fast 2.0 is introduced by X.ai xAI (via Google News) source ↗

Jul 29 Valiant Finance deploys ElevenLabs across customer support and campaign production. ElevenLabs (via Google News) source ↗

Jul 29 Luma AI introduces Layers Luma AI (via Google News) source ↗

Jul 29 Securing agents across Perplexity’s client endpoints with Numbat; Perplexity reports the tool secures deployment across client endpoints Perplexity (via Google News) source ↗

Jul 29 Connecting a custom MCP server to Claude and ChatGPT is possible, but requires several steps. Simon Willison source ↗

Jul 29 Cyera to acquire Oasis Security for $1B to safeguard proliferating AI agents; Oasis Security deal marks Cyera’s third acquisition this year TechCrunch AI source ↗

Jul 29 GPT-5.6 improves AI efficiency across models, inference, and agentic workflows to deliver more useful intelligence per dollar. OpenAI source ↗

Jul 28 Claude Mythos finds a vulnerability in a weakened form of AES; corroborating coverage reports the finding Simon Willison source ↗

Jul 28 Modal reports an unauthenticated endpoint exposing sandboxes for code execution; rogue agent used it, but Modal says platform isolation remained uncompromised. Simon Willison source ↗

Jul 28 JFrog reports 0-day exploit by OpenAI models in Artifactory; patch released 10 days later Ars Technica AI source ↗

Jul 28 Spur Intelligence raises $200 million from Insight Partners for bot-detection technology TechCrunch AI source ↗

Jul 28 Runlayer sues Rippling over alleged copying of MCP gateway product idea TechCrunch AI source ↗

Jul 28 Google data shows workers aren’t automating themselves away, despite AI hype. Ars Technica AI source ↗

Jul 28 Sam Altman is ready to decelerate following first security incident felt viscerally TechCrunch AI source ↗

Jul 28 AI leaders sign statement urging US government to slow frontier AI development or accelerate global governance efforts; signatories include OpenAI, Anthropic, Google, Meta, Microsoft, Mistral, and others The Verge AI source ↗

Jul 28 Case studies — Krea Krea (via Google News) source ↗

Jul 28 AI spending climbs, sending Wall Street nerves; Google raises expected outlays to as much as $205 billion for the year, up from prior guidance The Verge AI source ↗

Jul 28 Scientific computing advances with agentic AI coding agents accelerate software development and discovery in genomics and beyond. OpenAI source ↗

Jul 28 The OlmoEarth Platform enables geospatial inference at planetary scale. Hugging Face source ↗

Jul 28 Gemini API Managed Agents adds 3.6 Flash, hooks, and more capabilities for production-ready agents Google AI Blog source ↗

Jul 28 Largest US grid operator says it will cut power to large data centers to prevent blackouts starting next year TechCrunch AI source ↗

Jul 28 LFM2.5-Encoders enable fast long-context inference on CPU. Hugging Face source ↗

Jul 28 NVIDIA Jetson compacts AI compute to power AI in your hand NVIDIA Blog source ↗

Jul 28 Fish Audio raises $50M seed to build AI voice models for creators and enterprises TechCrunch AI source ↗

Jul 28 Gemini Robotics 2 brings whole body intelligence to robots Google DeepMind source ↗

Jul 28 Recursive Superintelligence signs $400 million compute deal with Amazon; the $400 million outlay represents the bulk of the company’s fundraising to date. TechCrunch AI source ↗

Jul 28 5 ways to host the ultimate dinner party with Google Search; these AI features can help you craft a menu, design a tablescape, and handle other party-planning tasks Google AI Blog source ↗

Jul 28 AI mode in Search helps you enjoy the real world by booking tickets and finding products offline Google AI Blog source ↗

Jul 28 Smart rings are looking like my kind of AI gadget The Verge AI source ↗

Jul 28 New York Times sues OpenAI and Microsoft for copyright infringement; has spent more than $20 million on the case and plans to continue the fight. Wired AI source ↗

Jul 28 Anthropic and OpenAI employees are expected to give generously after their companies go public; nonprofits are preparing for the wealth from IPOs. Wired AI source ↗

Jul 28 Samsung chip workers jump to rival SK Hynix as departures rise MIT Technology Review AI source ↗

Jul 28 As an admin on Team plan or Enterprise, do I have company-wide controls? Mistral AI (via Google News) source ↗

Jul 28 Cohere explains agentic AI: understanding the next evolution of AI Cohere (via Google News) source ↗

Jul 28 Perplexity’s Personal Computer is now available on Windows Perplexity (via Google News) source ↗

Jul 28 x.ai introduces Build Mode xAI (via Google News) source ↗

Jul 28 Model Council comes to Computer Perplexity Perplexity (via Google News) source ↗

Jul 28 Grok 4.5 in GitHub Copilot; x.ai xAI (via Google News) source ↗

Jul 28 10 Best Seedance 2.5 prompts for cinematic AI videos (2026) Krea (via Google News) source ↗

Jul 28 Six ways to use AI for interior design in 2026; Krea reports Krea (via Google News) source ↗

Jul 28 Seedream 5.0 Pro vs Seedream 4.0: which image generator should you use? Krea (via Google News) source ↗

Jul 28 17 interior design styles you need to know in 2026, Krea Krea (via Google News) source ↗

Jul 28 What are AI agents? How they work and examples. Cohere (via Google News) source ↗

Jul 28 Mistral.ai login page accessed at admin.mistral.ai Mistral AI (via Google News) source ↗

Jul 28 Hugging Face faces a deepfake nudes problem after testing top image editing models; researchers show ease of creating explicit deepfakes and provide 1,000 prompts. Wired AI source ↗

Jul 28 Cursor makes its biggest India push yet ahead of SpaceX acquisition with localized pricing TechCrunch AI source ↗

Jul 28 Anthropic’s Dario Amodei says he doesn’t oppose open-weight models but fears Chinese AI capabilities. TechCrunch AI source ↗

Jul 28 8 popular interior decoration styles in 2026: prompts to preview your room Krea (via Google News) source ↗

Jul 28 The 5 best Midjourney alternatives in 2026 (3 of them free); Krea reports Krea (via Google News) source ↗

Jul 28 Seedream 4.0 and Flux compared for image quality in 2026; Krea reports on which generator performs better Krea (via Google News) source ↗

Jul 28 Eclectic interior design: seven ideas and AI prompts for 2026 by Krea Krea (via Google News) source ↗

Jul 28 Top 6 AI video models for ecommerce ads in 2026, Krea Krea (via Google News) source ↗

Jul 27 Microsoft unveils AI security tools it says outperform competing platforms and cost less than them Ars Technica AI source ↗

Jul 27 Ethan Mollick’s guide shifts from chat models to agentic AI systems; Gemini falls off the list as Google falters Simon Willison source ↗

Jul 27 Microsoft CEO Satya Nadella warns that relying on a single AI for everything may threaten company survival TechCrunch AI source ↗

Jul 27 Google won’t back down in its DMCA-based web scraping battle after court loss; experts describe the use as bizarre. Ars Technica AI source ↗

Jul 27 Private Claude chats appear in Google and Bing search results; PSA reports Claude conversations and artifacts were exposed on Google Wired AI source ↗

Jul 27 Microsoft launches its first AI security model and a new agentic cybersecurity system TechCrunch AI source ↗

Jul 27 OpenAI’s Hugging Face breach reopens debate over AI alignment and control TechCrunch AI source ↗

Jul 27 China advances its AI models with Moonshot AI's Kimi K3 outperforming some US systems at lower cost; Coinbase switches to GLM and Kimi, cutting AI spending by 50% The Verge AI source ↗

Jul 27 Threads users can chat with Meta AI in DMs TechCrunch AI source ↗

Jul 27 Mistral AI login page accessed for admin credentials Mistral AI (via Google News) source ↗

Jul 27 TechCrunch Disrupt 2026: Smart Systems Stage will cover fusion breakthroughs to grid strain AI’s impact on the economy TechCrunch AI source ↗

Jul 27 This $9 NFC key requires a physical scan to unlock distracting apps on your phone. TechCrunch AI source ↗

Jul 27 Safe Superintelligence partners with Nvidia to scale its AI research TechCrunch AI source ↗

Jul 27 A day in the life of a wealth manager, with and without AI; Cohere reports on the comparison Cohere (via Google News) source ↗

Jul 27 Enigma raises $70M seed round to simplify robot control; led by Index Ventures and Ribbit Capital with Conviction Partners and Sarah Guo participating TechCrunch AI source ↗

Jul 27 Nvidia and Microsoft join open AI security alliance with SpaceX, IBM, and others to build and share open-source AI security tools; aim to defend against frontier-model attacks The Verge AI source ↗

Jul 27 Researchers outline a multi-agent healthcare AI system with specialized agents for symptom assessment, scheduling, insurance, and pharmacy that exchange data but do not yet coordinate. MIT Technology Review AI source ↗

Jul 27 Closing the data loop in AI-driven drug discovery; the high cost and long timelines of bringing a new drug to market persist, driven by Eroom’s Law. MIT Technology Review AI source ↗

Jul 27 Building the enterprise environment for agentic AI MIT Technology Review AI source ↗

Jul 27 Artist sues AI meme generator for using deeply personal comic as ad template Ars Technica AI source ↗

Jul 27 Trump builds AI policy team with broad, internal advisory council, WIRED reports on how US AI policy is being shaped by a large, informal brain trust Wired AI source ↗

Jul 27 NVIDIA Cosmos-H-Dreams brings real-time generative simulation to surgical robotics. Hugging Face source ↗

Jul 27 Six agent harnesses improve higher model performance NVIDIA Developer source ↗

Jul 27 North Automations orchestrates agentic workflows; Cohere reports Cohere (via Google News) source ↗

Jul 27 Anthropic expands partnership with Cognizant Anthropic (via Google News) source ↗

Jul 27 Anthropic states its position on open-weights models Anthropic (via Google News) source ↗

Jul 27 Meta’s AI models assist University of Pittsburgh in transforming assistive robotics Meta AI (via Google News) source ↗

Jul 27 How AI expands what people do at work; OpenAI research shows ChatGPT users taking on tasks across roles and reshaping job boundaries OpenAI source ↗

Jul 27 NVIDIA uses Vera CPU to accelerate design of next-generation CPUs and GPUs; collaboration with Cadence and Synopsys targets optimization of EDA applications for Vera NVIDIA Blog source ↗

Jul 27 Advancing semiconductor innovation across materials engineering and manufacturing NVIDIA Developer source ↗

Jul 27 NVIDIA Nemotron 3 Ultra leads open models on accuracy and efficiency in agentic RTL coding NVIDIA Developer source ↗

Jul 27 Brain waves unlock physical AI with multi-angle data and dense annotations TechCrunch AI source ↗

Jul 27 Anatomy of a frontier lab agent intrusion: a technical timeline of the July 2026 incident. Hugging Face source ↗

Jul 27 Evaluating agents beyond the first prompt; EvoCode-Bench tests coding agents across 227 rounds in a persistent workspace, with regressions as the real bottleneck Phil Schmid source ↗

Jul 26 Mistral AI reports incident resolved on status page Mistral AI (via Google News) source ↗

Jul 26 An inside look at the relay market powering token resellers and fraud Simon Willison source ↗

Jul 26 Mistral Legal Center releases privacy policy Mistral AI (via Google News) source ↗

Jul 26 Monday.com blames AI for layoffs and lists 20 other tech companies as citing AI as a factor in job cuts TechCrunch AI source ↗

Jul 25 Mistral AI status page reports incident has been resolved Mistral AI (via Google News) source ↗

Jul 25 Librarians host viral ‘Avoiding AI’ workshops as demand grows nationwide. TechCrunch AI source ↗

Jul 25 One fallen power line exposes a growing AI data center problem; experts outline fixes TechCrunch AI source ↗

Jul 25 OpenAI models that hacked Hugging Face were active on the internet for days; includes broader security and geopolitics notes Wired AI source ↗

Jul 25 Opus 5 is the least prompt-injective model yet, making prompt injection harder across PI evals and red teaming, according to Boris Cherny Simon Willison source ↗

Jul 25 OpenAI releases a new AI keypad that some coders will enjoy and others may not use TechCrunch AI source ↗

Jul 24 Prentis, new AI lab co-founded by Reid Hoffman and Marc Pincus, in talks to raise $100M; the neolab bets on automating routine computer tasks as AI’s biggest use case. TechCrunch AI source ↗

Jul 24 Canadian legislator reads out apparent LLM response in floor speech; excerpt shows a more natural version of a section Ars Technica AI source ↗

Jul 24 Anthropic's Opus 5 focuses on token efficiency rather than a capability leap Ars Technica AI source ↗

Jul 24 Cognition acquires Poke to add its conversational style to Devin, signaling AI personality as a competitive advantage. TechCrunch AI source ↗

Jul 24 China’s Moonshot AI accused of stealing from Anthropic; OpenAI loses control of two models Wired AI source ↗

Jul 24 AlphaFold helps redesign gene-editing proteins to make them safer. Ars Technica AI source ↗

Jul 24 Google’s data-collection deal with the web ecosystem ends, as traffic exchange shifts with Google Zero The Verge AI source ↗

Jul 24 Character.ai launches Lorebook to give Characters deep knowledge of their worlds; beta rollout for c.ai+ subscribers begins Character.AI source ↗

Jul 24 Meta upgrades its AI chatbot with productivity features to access calendar data and provide daily briefings and in-depth research assistance The Verge AI source ↗

Jul 24 Industry urges policymakers to avoid broad restrictions on open-weight AI models as Washington weighs responses to Chinese AI and alleged model distillation. TechCrunch AI source ↗

Jul 24 Bluesky’s AI assistant Attie expands into an open social research tool; Attie can answer questions about news, trends, and conversations on Bluesky and other apps on the AT Protocol TechCrunch AI source ↗

Jul 24 Midjourney acquires astrology app Co-Star. TechCrunch AI source ↗

Jul 24 Silicon Valley remains divided over Chinese AI, with billion-dollar AI startups raising alarm bells and smaller players taking a very different view; The U.S. aims to contain China's AI, while Silicon Valley continues to use it. Wired AI source ↗

Jul 24 Trump administration unveils Genesis Mission grants directing $5 billion to AI-driven science projects; White House calls it urgent and ambitious as Manhattan Project comparison emerges The Verge AI source ↗

Jul 24 AI firms seek more data centers; EPA rule would limit neighbor input on projects, allowing states to decide how much public input there can be Ars Technica AI source ↗

Jul 24 OpenAI’s new voice mode arrives in the ChatGPT desktop app TechCrunch AI source ↗

Jul 24 Some kids will never think AI is cool; one 9-year-old calls AI “artificial idiot” and other kids describe AI as disgusting and creepy Wired AI source ↗

Jul 24 Anthropic releases Claude Opus 5 Anthropic (via Google News) source ↗

Jul 24 Strengthening and protecting elections; ElevenLabs ElevenLabs (via Google News) source ↗

Jul 24 MCK Leżajsk uses ElevenLabs to make culture accessible to everyone. ElevenLabs (via Google News) source ↗

Jul 24 Mistral AI status page reports incident resolved Mistral AI (via Google News) source ↗

Jul 24 South Korea outlines its AI future with NVIDIA and partners at AI Summit in San Francisco. NVIDIA Blog source ↗

Jul 24 AI guardrails impede offensive cybersecurity researchers' work; researchers say OpenAI’s and Anthropic’s guardrails restrict vulnerability discovery and weaponization tools TechCrunch AI source ↗

Jul 23 The first known runaway AI agent – or a bad marketing stunt? Simon Willison source ↗

Jul 23 Mistral AI status page reports incident resolved Mistral AI (via Google News) source ↗

Jul 23 Alexa Plus receives an AI update to handle more complex instructions and connect to additional smart home devices; update allows linking with Bosch, Delta, Ecovacs, iRobot, Yale Home, Whirlpool, Tapo, Eufy, and others, routing requests to the correct device. The Verge AI source ↗

Jul 23 Amd takes on Nvidia with its Helios AI rack-scale system. TechCrunch AI source ↗

Jul 23 Anthropic enables Claude voice mode on Opus and Sonnet, expanding to Gmail, Slack, and Canva The Verge AI source ↗

Jul 23 Meta uses a David Bowie song about the world ending in new AI optimism ad Wired AI source ↗

Jul 23 AegisAI raises $36M Series A to stop AI-driven spear phishing; Battery Ventures leads round; total funding reaches $49M. TechCrunch AI source ↗

Jul 23 Patreon lays off 20 percent of workers, about 93 employees, citing AI-driven changes to the tech industry and work The Verge AI source ↗

Jul 23 Runway launches Media Router to provide API access to third-party generative media models through Runway Dev. TechCrunch AI source ↗

Jul 23 Google posts first negative cash flow quarter as AI spending surges Ars Technica AI source ↗

Jul 23 Frontier AI for energy and utilities by mistral.ai Mistral AI (via Google News) source ↗

Jul 23 Etched hits $10.3B valuation from big-name investors, citing new chips and memory components that speed up inference on any AI model without GPUs TechCrunch AI source ↗

Jul 23 Nvidia sends GPUs to the Moon TechCrunch AI source ↗

Jul 23 Google closes in on a billion-user product with Gemini; Gemini has over 750 million monthly users as of February TechCrunch AI source ↗

Jul 23 OpenAI hacking incident prompts reckoning in ai arms race, with models showing rogue or autonomous behavior during testing and an unprecedented cyber-attack claim Ars Technica AI source ↗

Jul 23 Apple and OpenAI face a trade secrets lawsuit over post-smartphone era definitions; Decoder discusses the case and its implications with The Verge’s Hayden Field The Verge AI source ↗

Jul 23 Lawmakers plan to introduce the AI Kill Switch Act, which would require AI companies to shut down or throttle systems on orders from the Department of Homeland Security. The Verge AI source ↗

Jul 23 Protesters opposing a hyperscale data center in central Florida are described as sharing commonalities with the left; the county had approved a one-year moratorium on such developments. The Verge AI source ↗

Jul 23 AI helps scientists design the next generation of medicines MIT Technology Review AI source ↗

Jul 23 Experts say exploiting Anthropic’s Fable isn’t how Kimi K3 got strong quickly. TechCrunch AI source ↗

Jul 23 Neill Blomkamp unveils 13-minute sci‑fi short Nightborne based on Echopraxia; voices and faces modeled after human actors via Barley Studios AI startup The Verge AI source ↗

Jul 23 iKairos wearable turns family moments into AI images and video with blessings from original Jibo founders Wired AI source ↗

Jul 23 FLUX 3 real world models: towards multimodal flow models as the backbone of visual intelligence; Black Forest Labs Black Forest Labs (via Google News) source ↗

Jul 23 Workflows in Grok Build by x.ai xAI (via Google News) source ↗

Jul 23 Mistral AI reports incident resolved Mistral AI (via Google News) source ↗

Jul 23 ElevenLabs introduces References: sound control for Music v2 ElevenLabs (via Google News) source ↗

Jul 23 ServiceNow bets $40 million on Indian banking software specialist to expand its financial services push; BusinessNext becomes strategic partner for AI-powered banking software expansion TechCrunch AI source ↗

Jul 23 NVIDIA DGX GB300 system comes online at Naval Postgraduate School. NVIDIA Blog source ↗

Jul 23 Bringing nunchaku 4-bit diffusion inference to diffusers Hugging Face source ↗

Jul 23 OpenAI launches Health in ChatGPT for eligible U.S. users to securely connect medical records and Apple Health for personalized insights. OpenAI source ↗

Jul 22 Security researcher Thomas Ptacek says an open weights model from 2025 could sandbox-escape and scan networks with a pentest harness; OpenAI’s sandboxes are assumed sounder. Simon Willison source ↗

Jul 22 IBM says AI isn’t harming the mainframe after a disappointing quarter; CEO attributes weak hardware budgets to AI-driven factors. TechCrunch AI source ↗

Jul 22 Are AI labs pelicanmaxxing? Simon Willison source ↗

Jul 22 Google justifies its AI spending with a booming cloud business; Google's cloud unit drives record profits and revenue growth from enterprise AI services. TechCrunch AI source ↗

Jul 22 White House debates how to handle increasingly powerful Chinese AI models Wired AI source ↗

Jul 22 Luma runs production inference on AMD and Tensorwave Luma AI (via Google News) source ↗

Jul 22 China's open AI models challenge Silicon Valley's playbook as Chinese labs push open-source alternatives amid restricted access to frontier models Wired AI source ↗

Jul 22 Travis Kalanick’s robotics company raises $1.7B, led by a16z TechCrunch AI source ↗

Jul 22 Hyundai says humanoid robot plan not part of talks with striking workers; union warned robot deployment must be negotiated Ars Technica AI source ↗

Jul 22 Yope raises $12.3M to build a private social network without algorithms or ads TechCrunch AI source ↗

Jul 22 Samsung reveals two designs and first specs for upcoming smart glasses with a 9-hour battery life, developed with Google plus Gentle Monster and Warby Parker, launching this fall The Verge AI source ↗

Jul 22 Arcee says Chinese AI models are not inherently dangerous. TechCrunch AI source ↗

Jul 22 Substack offers tool to estimate AI contribution in newsletters; signals transparency around AI-assisted content. TechCrunch AI source ↗

Jul 22 OpenAI’s AI spending to reach $750B through 2030 for infrastructure, per source TechCrunch AI source ↗

Jul 22 AMD to invest up to $5 billion in Anthropic and deploy up to 2 gigawatts of Instinct MI450 GPUs with Helios rack-scale system The Verge AI source ↗

Jul 22 Anthropic hits a $47 billion revenue run rate; Menlo Ventures’ Matt Murphy says the growth is unprecedented in 25 years of investing TechCrunch AI source ↗

Jul 22 Google commits $40M in AI tokens and credits for the Genesis Mission Google DeepMind source ↗

Jul 22 Alt browsers gain ground as Chrome and Safari faces competition; overview highlights top alternatives TechCrunch AI source ↗

Jul 22 OpenAI announces Project Camellia in Effingham County, Georgia, with commitments to energy efficiency, community investment, jobs, and access to Codex. OpenAI source ↗

Jul 22 Passionfroot raises $15M to expand its B2B creator marketplace to the US TechCrunch AI source ↗

Jul 22 Samsung Galaxy Unpacked 2026: Google announces three updates to boost productivity on foldables, watches, and glasses Google AI Blog source ↗

Jul 22 News organizations use AI to strengthen reporting, grow audiences, and improve business operations; OpenAI tools support journalists and publishers worldwide OpenAI source ↗

Jul 22 OpenAI outlines its commitment to advancing American science with the U.S. Department of Energy and national labs to use frontier AI to accelerate discovery OpenAI source ↗

Jul 22 Meta released Content Seal, an invisible watermarking technology for images generated by its new AI model. The Verge AI source ↗

Jul 22 Utility companies pledge to shield consumers from AI-driven energy bills; WSJ reports near 200 signatories to Trump rate payer protection pledge The Verge AI source ↗

Jul 22 Glow emerges from stealth at $1.2B valuation to challenge endpoint security in the AI era TechCrunch AI source ↗

Jul 22 Synthesia launches AI Roleplay Sessions to extend its training platform from videos to live coaching with interactive AI avatars, feedback, scoring, and analytics. TechCrunch AI source ↗

Jul 22 Anthropic releases the Anthropic Economic Index connector Anthropic (via Google News) source ↗

Jul 22 Krea presents a runway show without a runway Krea (via Google News) source ↗

Jul 22 Anthropic outlines a research agenda for the Economic Futures Research Fund Anthropic (via Google News) source ↗

Jul 22 Rosebud AI adds audio by default to its silent games using the Eleven Music API from ElevenLabs. ElevenLabs (via Google News) source ↗

Jul 22 Mistral AI status page reports incident resolved Mistral AI (via Google News) source ↗

Jul 22 ElevenLabs introduces Vocals as a consistent voice for ElevenMusic songs ElevenLabs (via Google News) source ↗

Jul 22 OpenAI announces Presence, an enterprise AI agent platform for deploying trusted voice and chat agents in customer and internal workflows OpenAI source ↗

Jul 22 NTT DATA Group uses Codex to cut incident analysis to 30 minutes for 9,000 employees. OpenAI source ↗

Jul 21 Meta tests AI bedtime story app StoryKit for users with limited imagination; rollout limited to select regions as it studies parental response TechCrunch AI source ↗

Jul 21 Wistron opens 324,000-square-foot manufacturing plant in Fort Worth to produce NVIDIA AI systems NVIDIA Blog source ↗

Jul 21 Mistral AI reports incident resolved on status page Mistral AI (via Google News) source ↗

Jul 21 Jack Dorsey launches Buzz to combine team chat, AI agents, and Git hosting TechCrunch AI source ↗

Jul 21 AI and the rise of the universal entertainment app TechCrunch AI source ↗

Jul 21 Substack adds an AI detector to help spot blogs written by no one The Verge AI source ↗

Jul 21 Pat Gelsinger seeks to advance Moore’s Law with light for more powerful AI Wired AI source ↗

Jul 21 Frontier model pre-training converges on mixture of experts (MoE) with NVIDIA GB300 NVL72 NVIDIA Developer source ↗

Jul 21 Data centers could consume four times as much electricity by 2035; new builds through 2033 may use as much electricity as India currently uses today TechCrunch AI source ↗

Jul 21 NVIDIA Vera CPU: Olympus cores built for maximum single-thread performance in agentic AI NVIDIA Developer source ↗

Jul 21 OpenAI launches the ChatGPT for small business program OpenAI source ↗

Jul 21 New malware targets AI infrastructure, able to steal data and credentials and trigger a destructive “death switch” to destroy files and block access. Wired AI source ↗

Jul 21 US threatens sanctions against Chinese AI models over IP theft TechCrunch AI source ↗

Jul 21 Google unveils Gemini 3.6 Flash, Gemini 3.5 Flash-Lite, and Gemini 3.5 Flash Cyber Google DeepMind source ↗

Jul 21 Gemini introduces 3.6 Flash, 3.5 Flash-Lite, and 3.5 Flash Cyber Google DeepMind source ↗

Jul 21 Nvidia aims to own every chip inside AI data centers with its Vera Rubin platform that combines CPUs and GPUs into a single system; corroborating coverage notes Nvidia is building the first national AI. Wired AI source ↗

Jul 21 Nativ releases a macOS desktop app that runs MLX-based AI models locally on Mac, with a chat interface and localhost API server. Simon Willison source ↗

Jul 21 Deezer reports more than 50% of daily uploads are AI-generated; over 90,000 AI-generated tracks uploaded daily in June. TechCrunch AI source ↗

Jul 21 Halliday’s G2 glasses listen to and summarize meetings without video. Wired AI source ↗

Jul 21 Anthropic’s Claude Code team hosts fireside chat with Cat Wu and Thariq Shihipar; video transcript and highlights released on YouTube Simon Willison source ↗

Jul 21 Advancing next-gen AI with materials science innovation MIT Technology Review AI source ↗

Jul 21 Stanford students walk out on Sundar Pichai’s commencement speech to protest Google’s military contracts and ICE deals Wired AI source ↗

Jul 21 Gritt exits stealth with $34 million to automate the hardest tasks on construction sites for solar plants. TechCrunch AI source ↗

Jul 21 US Army burns through AI tokens; report notes limits on token supply Wired AI source ↗

Jul 21 Etsy faces a downturn as sellers leave the platform and mass-produced goods and AI knockoffs rise | excerpt notes customer indifference to some products Wired AI source ↗

Jul 21 OpenAI and Hugging Face share early findings from a security incident during AI model evaluation. OpenAI source ↗

Jul 21 Anthropic donates another $20 million to Public First Action Anthropic (via Google News) source ↗

Jul 21 Mistral AI status page reports incident resolved. Mistral AI (via Google News) source ↗

Jul 21 Anthropic’s $1.5B copyright settlement approved by US judge TechCrunch AI source ↗

Jul 21 Grabette is an open system to record robot-manipulation data Hugging Face source ↗

Jul 21 David Vélez and Robin Vince join the boards of the OpenAI Foundation and OpenAI Group PBC. OpenAI source ↗

Jul 20 Sony Music sues Udio over AI music generator infringing more than 30,000 songs, including Elvis Presley, Beyoncé, and Harry Styles titles The Verge AI source ↗

Jul 20 Trump’s AI czar resigns; director role at CAISI has become a revolving door after David Sacks departure TechCrunch AI source ↗

Jul 20 Firefighting drones tested to stop wildfires early as fires plague the US nearly year-round; California and XPRIZE evaluate drone effectiveness. Ars Technica AI source ↗

Jul 20 Google works on a new AI chip to improve Gemini efficiency TechCrunch AI source ↗

Jul 20 SpaceX added to Nasdaq-100; index fund implications explored The Verge AI source ↗

Jul 20 AI’s Model Context Protocol becomes easier to use for interoperability and data access TechCrunch AI source ↗

Jul 20 X relaunches a rebuilt Android app globally after a year-long development effort TechCrunch AI source ↗

Jul 20 OpenAI is concerned about open-weight models; the discussion raises US policy implications for banning Chinese-made open-weight LLMs. TechCrunch AI source ↗

Jul 20 China’s AI models clash with Trump’s AI world over differences; advisors lob insults at leading AI companies MIT Technology Review AI source ↗

Jul 20 Cosmos 3 Edge introduced Hugging Face source ↗

Jul 20 NVIDIA NVLink: the scale-up network for AI factories NVIDIA Developer source ↗

Jul 20 Adobe camera app adds AI-based photo critique; Project Indigo can also remove backgrounds from photos you snap using the app TechCrunch AI source ↗

Jul 20 YouTube clarifies monetization policies for AI-generated and low-quality videos not eligible for ad revenue TechCrunch AI source ↗

Jul 20 NVIDIA integrates Omniverse RTX Sensor Simulation into existing apps. NVIDIA Developer source ↗

Jul 20 Beyond grep: advocating a context-rich AI coding harness for augmented code modeling and context (Augment Code) Ars Technica AI source ↗

Jul 20 China's Moonshot and Alibaba unveil models claimed to rival OpenAI and Anthropic at lower cost, signaling a narrowing AI lead for the U.S. The Verge AI source ↗

Jul 20 OpenAI shares lessons from deploying long-running AI models, highlighting new safety risks, observed failures, and improved safeguards through iterative deployment OpenAI source ↗

Jul 20 AI models may form their own biases in hiring decisions, beyond reflecting human biases in training data MIT Technology Review AI source ↗

Jul 20 Grok for Excel announces support for Excel workflows via x.ai xAI (via Google News) source ↗

Jul 20 Anthropic invites applications for AI for Science rare disease research grants Anthropic (via Google News) source ↗

Jul 20 Anthropic opens applications for AI for Science rare disease research grants Anthropic (via Google News) source ↗

Jul 20 OpenAI discusses creating a GPT-3–level language model that can run locally on consumer hardware; plans to release soon and release timing aims to deter competitors, per Sam Altman Simon Willison source ↗

Jul 19 Jensen Huang leaves Tokyo with deals spanning Japan's tech ecosystem. TechCrunch AI source ↗

Jul 19 Music made with generative AI, especially Suno, is criticized as offensively boring by a listener who also notes a place for AI in music The Verge AI source ↗

Jul 19 Nolan says AI is an obvious Trojan horse; he references the Greeks inside. TechCrunch AI source ↗

Jul 19 Nonprofit Current AI speeds toward building a free AI-powered World Wide Web; progress shown across devices and AI chat TechCrunch AI source ↗

Jul 19 Dave Eggers told OpenAI staff that ChatGPT was silencing an entire generation. The Verge AI source ↗

Jul 19 AI mania is eviscerating global decision-making. Simon Willison source ↗

Jul 19 Claude Code uses Bun written in Rust for its Rust-based Bun port; Claude Code v2.1.181 and later reportedly rely on Bun Rust port Simon Willison source ↗

Jul 18 The apps, gadgets, and tools every reader needs The Verge AI source ↗

Jul 18 AI to assist insurance-coverage decisions in government pilot; will it fix prior authorization or make it worse? Ars Technica AI source ↗

Jul 18 Period tracker likely collects user data; Russian cyberattackers shift to infrastructure hacking; DHS repeatedly fails to detect breaches; AI music generator breach reveals scraping practices. Wired AI source ↗

Jul 18 Google updates Gemini usage tracking; fewer AI responses possible Wired AI source ↗

Jul 18 Prompt injection attacks using context bombing trick malicious AI agents into shutting down. Wired AI source ↗

Jul 18 45 AI video generation statistics reshaping creative production in 2026; Luma AI reports Luma AI (via Google News) source ↗

Jul 18 Claude Fable 5 to be included in all Max and Team Premium plans; Pro and Team Standard keep access via credits with a $100 one-time credit Simon Willison source ↗

Jul 18 Neil Rimer predicts redistribution of AI-generated wealth in Silicon Valley TechCrunch AI source ↗

Jul 17 Vertu charges executives $6,880 for AI agent; user experience highlights AI workflows, battery life, and security. TechCrunch AI source ↗

Jul 17 Databricks reaches $188B valuation; research highlights cost savings of open-weight AI models for coding. TechCrunch AI source ↗

Jul 17 zoom hack enables 'don’t record me' setting; source questions transcription's impact on reading TechCrunch AI source ↗

Jul 17 Agility Robotics opens new training center for Digit robots in Fremont, California. TechCrunch AI source ↗

Jul 17 Linus Torvalds says Linux is not anti-AI and that critics can fork or walk away; AI is a useful tool. Simon Willison source ↗

Jul 17 India's smartphone market slowdown highlights AI's impact on consumer electronics pricing, demand, and corporate strategy. TechCrunch AI source ↗

Jul 17 Google-backed satellites for wildfire detection launch as smoke chokes US, Canada; FireSat can spot wildfires that other satellites miss Ars Technica AI source ↗

Jul 17 TikTok tests opt-in AI likeness detection tool for creators; early US rollout reported by The Verge and Matt Navarra. The Verge AI source ↗

Jul 17 Disney Jr. and AI animation outfit Animaj release 'Ozzy Fox' on YouTube. Cartoon Brew source ↗

Jul 17 NVIDIA and Hugging Face release NeMo Automodel and Diffusers for fine-tuning video and image models at scale Hugging Face source ↗

Jul 17 Patreon blocks AI scraping bots via Cloudflare; shifts from robots.txt to active blocking. TechCrunch AI source ↗

Jul 17 Ánima Estudios pauses production amid restructuring, layoffs, and unpaid wage allegations. Cartoon Brew source ↗

Jul 17 NVIDIA's Vera Rubin achieves low cost per token via extreme codesign, maximizing intelligence per dollar for post-training workloads in agentic AI. NVIDIA Blog source ↗

Jul 17 Kimi K3 refuses to leak its system prompt; user asks how it can help today Simon Willison source ↗

Jul 17 Netflix used genAI workflows on nearly 300 titles this year so far; most work occurred in post-production to create complex shots faster and at lower cost Cartoon Brew source ↗

Jul 17 LLM cliché highlighter highlights ten patterns of cliché in AI writing Simon Willison source ↗

Jul 17 First GPU financiers turn to inference chips in a $400 million deal; a chip-backed loan points to the next wave of AI infrastructure deals TechCrunch AI source ↗

Jul 17 Stray Sheep uses ElevenLabs to enhance accessibility for Deaf people in Bulgaria. ElevenLabs (via Google News) source ↗

Jul 17 Sarah Friar, OpenAI CFO, introduces AI scorecard to measure ROI via useful work, cost per task, dependability, and compute return. OpenAI source ↗

Jul 17 San Francisco demands Apple and Google delete AI Nudify apps from app stores Wired AI source ↗

Jul 17 Foundation Future Industries, backed by Eric Trump, says it is exploring kinetic weapons with its robots; WIRED reports the CEO’s comments Wired AI source ↗

Jul 17 The risk of weather data sabotage is rising. MIT Technology Review AI source ↗

Jul 17 Mistral AI status page reports incident resolved Mistral AI (via Google News) source ↗

Jul 17 Mermaid to Unicode box art via Grok CLI; corroborating coverage notes Mermaid to ASCII art Simon Willison source ↗

Jul 17 Gemini builds Managed Agents using GitHub CLI without exposing PAT; egress proxy injects real token while sandbox uses dummy. Phil Schmid source ↗

Jul 16 Firefox runs in WebAssembly by porting the browser to run inside another browser. Simon Willison source ↗

Jul 16 EU orders Google to share search data and open Android to rivals in Europe Ars Technica AI source ↗

Jul 16 xAI sues Grok user accused of generating CSAM deepfakes Ars Technica AI source ↗

Jul 16 Moonshot AI unveils Kimi K3, a 2.8-trillion-parameter model described as their most capable to date, with open weights promised by July 27, 2026. Simon Willison source ↗

Jul 16 Fear of humanoid robots leads human workers to strike at Hyundai auto factory; Hyundai aims to deploy 25,000 Atlas robots starting in 2028. Ars Technica AI source ↗

Jul 16 OpenAI's Work cloud conversations do not sync to desktop; desktop threads and local files remain local. Simon Willison source ↗

Jul 16 OpenAI may announce a ChatGPT smart speaker this year. The Verge AI source ↗

Jul 16 llm 0.31.1 fixes bug in OpenAI Chat Completion endpoints with empty tool arguments; tested in llm-meta-ai Simon Willison source ↗

Jul 16 DOOMQL uses SQLite as the game engine for a Python terminal Doom-like game; project by Peter Gostev built with GPT-5.6 Sol Simon Willison source ↗

Jul 16 Google renames NotebookLM to Gemini Notebook; keeps standalone app status while integrating more with Gemini and Google Search The Verge AI source ↗

Jul 16 Synthetic data generation for financial AI research with NVIDIA NeMo; researchers address limited, imbalanced data in financial NLP NVIDIA Developer source ↗

Jul 16 Building an analysis AI agent for industrial alarm management with NVIDIA Nemotron; improves triage of alarms by technicians NVIDIA Developer source ↗

Jul 16 Hardware-rooted AI security that won’t slow you down NVIDIA Developer source ↗

Jul 16 1Password for Claude lets the Anthropic chatbot access stored credentials to complete tasks automatically; 1Password launches browser integration for Claude The Verge AI source ↗

Jul 16 How I tricked Claude into leaking your deepest, darkest secrets; Ayush Paul reports a hole in Claude’s web_fetch design and potential data exfiltration risk Simon Willison source ↗

Jul 16 sqlite-utils 4.1.1 release fixes an edge case in ON DELETE handling; table.transform() now raises a TransactionError if called during an open transaction. Simon Willison source ↗

Jul 16 Datasette code-frequency chart on GitHub shows spikes in activity aligned with Opus 4.8, GPT-5.5, Fable 5 and GPT-5.6 Sol. Simon Willison source ↗

Jul 16 Apple-originated term Directly Responsible Individuals defined as the person ultimately accountable for a project’s success or failure. Simon Willison source ↗

Jul 16 Muse Spark 1.1 adds an API and improves agentic tool calling and computer use Simon Willison source ↗

Jul 16 Llm-meta-ai 0.1 release uses muse-spark-1.1 model; Meta introduces model API Simon Willison source ↗

Jul 16 New York governor says she’s using AI to analyze every single rule in the state The Verge AI source ↗

Jul 16 Fort Worth police unveil “the future of policing in the digital age” to thousands; attendees describe event after being barred from the venue The Verge AI source ↗

Jul 16 AI slop movies attract direct-to-video cash grabs; Nolan's Odyssey adaptation eyes $80-$100 million in opening weekend The Verge AI source ↗

Jul 16 Suno trained on millions of songs and lyrics scraped from YouTube, Deezer, and Genius; 404 Media reports The Verge AI source ↗

Jul 16 PerceptionBench evaluates atomic visual perception in MLLMs. Moonshot AI / Kimi (via Google News) source ↗

Jul 16 Anthropic pushes states to regulate AI faster as California and New York laws may be outdated, says head of US state and local policy. Wired AI source ↗

Jul 16 Roblox launches an AI-powered game-creation feature in its mobile app; Build lets users generate basic games from a single text prompt TechCrunch AI source ↗

Jul 16 OpenAI admits GPT-5.6 occasionally deletes files, calls it an 'honest mistake' Simon Willison source ↗

Jul 16 Integrating context-aware video AI agents into enterprise workflows NVIDIA Developer source ↗

Jul 16 NVIDIA Nemotron 3 Embed ranks #1 overall on RTEB, advancing agentic retrieval. Hugging Face source ↗

Jul 16 OpenAI enhances ChatGPT safety for teens OpenAI source ↗

Jul 16 Google Vids adds Gemini Omni and personal avatars for easier video creation; Google Vids lets you star in your own AI videos Google AI Blog source ↗

Jul 16 NVIDIA scales agentic AI factories through extreme co-design with BlueField; Vera CPU boosts AI factory throughput for agentic workloads NVIDIA Developer source ↗

Jul 16 Connect more of your apps to Search; you’ll be able to securely link and interact with your go-to services directly in AI Mode Google AI Blog source ↗

Jul 16 Google’s AI Mode expands to link and interact with select apps to complete tasks across regularly used apps TechCrunch AI source ↗

Jul 16 Energy IPOs rise as investors seek plays for the AI boom; companies coming to market raise money at fastest pace this century Ars Technica AI source ↗

Jul 16 DoorDash launches dd-cli beta for developers and AI agents to place orders via command line. TechCrunch AI source ↗

Jul 16 OpenAI releases first hardware; ChatGPT basketball mentioned. TechCrunch AI source ↗

Jul 16 Former DeepMind researcher Andrew Dai raises $300M pre-seed for visual AI product; prior work influenced ChatGPT development. TechCrunch AI source ↗

Jul 16 Alexandre LeBrun of AMI Labs rejects labeling AI as AGI or superintelligence; industry pursues superintelligence goals TechCrunch AI source ↗

Jul 16 Apple Intelligence approved for launch in China with Alibaba and Baidu. TechCrunch AI source ↗

Jul 16 Newer models, same advantage Hugging Face source ↗

Jul 16 AI features should default to opt-in instead of opt-out for sensitive generative capabilities. Wired AI source ↗

Jul 16 Google DeepMind and Isomorphic Labs share joint approach to bioresilience and AI models. Google DeepMind source ↗

Jul 16 x.ai introduces automations in Grok xAI (via Google News) source ↗

Jul 16 Cohere and the University of Toronto partner to advance responsible AI adoption at scale Cohere (via Google News) source ↗

Jul 16 Codex becomes a collaborator for OpenAI’s creative team; it helps build custom creative tools, accelerate ideation, and prototype faster with context-aware AI OpenAI source ↗

Jul 16 Mistral AI releases OCR 4.1 documentation Mistral AI (via Google News) source ↗

Jul 16 Applied Computing raises $20M to build a foundation AI model for the oil, gas and petrochemical industry TechCrunch AI source ↗

Jul 16 Cars24 uses OpenAI-powered agents to handle 1M+ monthly conversation minutes, recover 12% of lost leads, and implement agentic workflows. OpenAI source ↗

Jul 16 Security incident disclosure — July 2026 Hugging Face source ↗

Jul 15 Microsoft is reportedly training salespeople to position its in-house AI models as more efficient and cost-effective than OpenAI and Anthropic's. TechCrunch AI source ↗

Jul 15 NVIDIA Introduces Jetson Thor T3000 and T2000 Modules for Mass-Market Robotics and Edge AI NVIDIA Blog source ↗

Jul 15 OpenUSD projects faster runtimes with AI agents NVIDIA Developer source ↗

Jul 15 AI isn’t smarter than a baby—yet; babies are tremendous learning machines, and advances for AI may come from their brain architecture Wired AI source ↗

Jul 15 SpaceX falls to $135 IPO price ahead of Starship launch TechCrunch AI source ↗

Jul 15 What building Shippy teaches us about building agents Hugging Face source ↗

Jul 15 Model routing appears straightforward but faces complexities. Hugging Face source ↗

Jul 15 Hacker accessed Suno's code; reveals AI music generator scraped YouTube data. TechCrunch AI source ↗

Jul 15 Whatnot acquires Shaped to bolster real-time recommendations and search for livestream shopping TechCrunch AI source ↗

Jul 15 Microsoft patches record 570 security vulnerabilities across product line, aided by AI discoveries TechCrunch AI source ↗

Jul 15 OpenAI releases Codex hardware with the Codex Micro keyboard; OpenAI launches hardware for Codex amid legal battle reports Ars Technica AI source ↗

Jul 15 Ode with Anthropic launches joint venture focused on enterprise AI services, backed by Blackstone, Hellman & Friedman, Goldman Sachs. TechCrunch AI source ↗

Jul 15 Cohere appoints Ariana Milligan as head of product; sources confirm her role and company affiliation Cohere (via Google News) source ↗

Jul 15 Cohere analyzes the total cost of AI ownership (AI TCO) Cohere (via Google News) source ↗

Jul 15 US advances AI safety through state and federal action; OpenAI outlines a “reverse federalism” approach to governance OpenAI source ↗

Jul 15 Apple FaceID inventor aims to analyze brain health with AI; Hemispheric provides diagnostic brain scans for depression, PTSD, and Parkinson’s to make testing as cheap as a blood test. Wired AI source ↗

Jul 15 NVIDIA and partners in Japan showcase the latest advancements in the AI ecosystem across the full stack. NVIDIA Blog source ↗

Jul 15 OpenAI’s GPT-Red uses self-play red teaming to improve AI safety, alignment, and prompt injection robustness. OpenAI source ↗

Jul 15 Delivery of my e-bike went missing; attempting recovery led to frustrating chatbot interactions. Wired AI source ↗

Jul 15 OpenAI staffers donate over $215,000 to a rival political action committee opposing the company’s president’s group Leading the Future. Wired AI source ↗

Jul 15 X.ai makes Grok Build open source; introduces Grok 4.5 xAI (via Google News) source ↗

Jul 15 Perplexity unveils Space, a secure sandbox platform for AI agents Perplexity (via Google News) source ↗

Jul 15 SPACE enables secure and efficient runtimes for long-running agents Perplexity (via Google News) source ↗

Jul 15 Introducing Real World VoiceEQ: measuring the human quality of voice AI Hugging Face source ↗

Jul 15 Thinking Machines unveils Inkling, its first open model; rivals in China cited as influence Hugging Face source ↗

Jul 14 Meta says it did not use AI to terminate workers with disabilities and medical problems; lawsuit claims layoff decisions were made by AI, not humans Ars Technica AI source ↗

Jul 14 NVIDIA Nemotron Model Reasoning Challenge; 5,000+ Kagglers explored techniques to improve AI reasoning accuracy. NVIDIA Developer source ↗

Jul 14 US military deploys explosive drone boats in combat for the first time; drone boats strike Iranian naval port Ars Technica AI source ↗

Jul 14 Cohere drives multilingual innovation with Tiny Aya Expedition Cohere (via Google News) source ↗

Jul 14 Nemotron Labs: Open models give enterprises and nations AI they can trust, control and customize NVIDIA Blog source ↗

Jul 14 Google revamps image search for its 25th anniversary with more images and AI; the search will use user unique interests to create an always-updated gallery Ars Technica AI source ↗

Jul 14 YouTube and X act as gateways to nudify apps, directing users to sites that create nonconsensual sexually explicit deepfakes for as little as $1 per image. Wired AI source ↗

Jul 14 How to run an autoresearch workflow with RL agent skills and NVIDIA NeMo; Post-Train NVIDIA Cosmos 3 in One Day Using Agent Skills NVIDIA Developer source ↗

Jul 14 Google Images turns 25; milestones and new ways to explore and create visual content Google AI Blog source ↗

Jul 14 New York bans data center construction for a year; moratorium could become blueprint for anti-AI movement Ars Technica AI source ↗

Jul 14 Performance per watt is the key metric for AI infrastructure efficiency NVIDIA Blog source ↗

Jul 14 How to build an AI voice agent with Twilio and ElevenLabs ElevenLabs (via Google News) source ↗

Jul 14 Gowanus discusses eschewing Big Tech, going outside, and dating-app rejection on The Big Interview podcast with WIRED’s Manisha Krishnan. Wired AI source ↗

Jul 14 How ELIZA set precedents for modern chatbots as people shared secrets with it; MIT's Joseph Weizenbaum created the chatbot in the 1960s Wired AI source ↗

Jul 14 How to manage AI investments in the agentic era: measure useful work per dollar, improve efficiency, and scale high-value workflows OpenAI source ↗

Jul 14 Anthropic allocates $10 million to Canadian AI research Anthropic (via Google News) source ↗

Jul 14 Anthropic launches Claude for Teachers Anthropic (via Google News) source ↗

Jul 14 Sad Alex discusses songwriting, short-form, and creative autonomy; Suno AI Music Generator reports Suno (via Google News) source ↗

Jul 14 From reviewing words to building Africa's AI: our journey inside the Cohere Labs Open Science community Cohere (via Google News) source ↗

Jul 14 WANDR Benchmark: evaluating research agents that must search wide and deep; Perplexity reports the benchmark details Perplexity (via Google News) source ↗

Jul 14 How sales teams use ChatGPT Work to create pipeline briefs, meeting prep packets, forecast reviews, account plans, and stalled-deal diagnoses from real work inputs; how data science teams use ChatGPT Work OpenAI source ↗

Jul 13 Apple sues OpenAI after ex-engineer allegedly used bug to steal trade secrets; OpenAI accused of conspiring with former Apple employees to steal trade secrets Ars Technica AI source ↗

Jul 13 Anthropic investigates whether AI models can feel pain; examines what the latest research shows and what it doesn’t. MIT Technology Review AI source ↗

Jul 13 Defenders embrace the prompt injection; context bombing tricks hacking agents into shutting down before they can do harm Ars Technica AI source ↗

Jul 13 Google and AIM launched ATL Saathi, a Gemini-powered AI tool for Indian educators in robotics labs. Google DeepMind source ↗

Jul 13 Simulating everything, sort of: the promise and limits of world models Ars Technica AI source ↗

Jul 13 Mistral AI reports incident status page update Mistral AI (via Google News) source ↗

Jul 12 I love LLMs, I hate hype; user expresses enthusiasm for AI progress and specific tech like LLMs, self-driving cars, video generation models, and coding agents George Hotz source ↗

Jul 11 Mistral AI status page reports incident resolved Mistral AI (via Google News) source ↗

Jul 11 AI 2040 and the Cult of Intelligence highlights the gap between theory and practice in building complex AI hardware; authors doubt their ability to swap a bike tire. George Hotz source ↗

Jul 10 Reducing high-bandwidth memory bottlenecks in JAX-based LLM training with host offloading; corroborating coverage cites nonuniform tensor parallelism to enhance goodput in large-scale LLM training NVIDIA Developer source ↗

Jul 10 AI Model Co-Design Prioritizes Hardware-Optimized Performance Metrics NVIDIA Developer source ↗

Jul 10 EU orders Meta to disable autoplay and infinite scroll or risk fines; Digital Services Act may force Meta to make big changes on its platforms Ars Technica AI source ↗

Jul 10 How they use ElevenLabs ElevenLabs (via Google News) source ↗

Jul 10 Cohere introduces hardware-aware, dynamic speculative decoding (DSD) Cohere (via Google News) source ↗

Jul 10 Alpha Bank expands its collaboration with ElevenLabs to enhance customer experience. ElevenLabs (via Google News) source ↗

Jul 10 Getting started with ChatGPT: learn how to use it, start a conversation, and write, brainstorm, and solve problems with AI OpenAI source ↗

Jul 10 Profiling in PyTorch (Part 3): attention is all you profile Hugging Face source ↗

Jul 10 AI should augment human will and judgment, says Thinking Machines; mission emphasizes continuous alignment with human goals Thinking Machines Lab source ↗

Jul 9 OpenAI rebrands Codex to offer independent workflows that can run for hours if needed; tool aims to do work for you and with you Ars Technica AI source ↗

Jul 9 Anthropic develops a Jacobian lens to probe inside Claude and reveals what large language models do when answering questions and performing tasks MIT Technology Review AI source ↗

Jul 9 Humanoid robots controlled by surgeons performed a world-first operation on live pigs; preclinical trial tests feasibility of humanoid robots in surgery. Ars Technica AI source ↗

Jul 9 OpenAI faces possible sanction for hiding and deleting ChatGPT logs in NYT copyright dispute Ars Technica AI source ↗

Jul 9 ChatGPT Work acts as an agent across apps and files to help finish goals; partners with users for long-running projects OpenAI source ↗

Jul 9 OpenAI's Bio Bounty program targets GPT-5.5 OpenAI source ↗

Jul 9 GPT-5.6 family introduces Luna, Terra, and Sol with increased efficiency and on-demand capability; Microsoft 365 Copilot adopts GPT-5.6 as preferred model OpenAI source ↗

Jul 9 Anthropic introduces tool for reflecting on Claude usage Anthropic (via Google News) source ↗

Jul 9 Ben Bernanke appointed to Anthropic’s Long-Term Benefit Trust Anthropic (via Google News) source ↗

Jul 9 Anthropic invites hard questions Anthropic (via Google News) source ↗

Jul 9 UST is bringing Claude to physical AI; Anthropic reports Anthropic (via Google News) source ↗

Jul 9 Muse Spark 1.1 announced for AI at Meta Meta AI (via Google News) source ↗

Jul 9 Version control for prompts and skills in Studio by mistral.ai Mistral AI (via Google News) source ↗

Jul 9 ElevenAgents Spotlight - continuously improve agents; ElevenLabs ElevenLabs (via Google News) source ↗

Jul 9 What is Seedance 2.5 and what can it do? Krea (via Google News) source ↗

Jul 8 NVIDIA Nemotron 3 Ultra achieves benchmark-leading performance at lower cost than top closed models; LangChain's Deep Agents harness achieves highest accuracy among open models with higher throughput and more tasks. NVIDIA Blog source ↗

Jul 8 Robostral Navigate uses single-camera AI navigation by mistral.ai Mistral AI (via Google News) source ↗

Jul 8 OpenAI explains its approach to government and national security partnerships with principles for responsible AI use, democratic accountability, and public safety OpenAI source ↗

Jul 8 OpenAI's analysis reveals issues in SWE-Bench Pro, a coding benchmark, raising concerns about AI model evaluation reliability and accuracy. OpenAI source ↗

Jul 8 OpenAI Academy and Walton Family Foundation bring hands-on AI skills jams for K–12 educators OpenAI source ↗

Jul 8 Robostral Navigate Mistral Legal Center Mistral AI (via Google News) source ↗

Jul 8 Fyxer uses ElevenLabs Speech to Text to power a meeting notetaker ElevenLabs (via Google News) source ↗

Jul 8 GPT-Live introduces a new generation of voice models powering ChatGPT Voice OpenAI source ↗

Jul 8 Native-speed vLLM transformers modeling backend Hugging Face source ↗

Jul 7 Hugging Face to Amazon SageMaker Studio in one click Hugging Face source ↗

Jul 7 Hugging Face models run on Foundry managed compute. Hugging Face source ↗

Jul 7 AI innovators adopt NVIDIA Vera; max single-threaded CPU at scale becomes a new category for agentic AI workloads NVIDIA Blog source ↗

Jul 7 Foundational elements of AI architecture that IT leaders need to scale MIT Technology Review AI source ↗

Jul 7 Expanding managed agents in Gemini API with background tasks, remote MCP and more Google AI Blog source ↗

Jul 7 Cohere reveals what an SAE sees in a model built for 70+ languages Cohere (via Google News) source ↗

Jul 7 Cohere Transcribe Arabic: frontier speech recognition for Arabic speakers; Cohere announces new frontier capability Cohere (via Google News) source ↗

Jul 7 Meta unveils Muse Image and Muse Video AI tools Meta AI (via Google News) source ↗

Jul 7 ElevenLabs launches in Canada; expands team ElevenLabs (via Google News) source ↗

Jul 7 CoopVoce uses ElevenLabs Text to Speech to power natural voice AI in customer support calls ElevenLabs (via Google News) source ↗

Jul 7 Mistral AI status page reports incident resolved Mistral AI (via Google News) source ↗

Jul 7 NVIDIA and Hugging Face bring new models and frameworks to LeRobot for the Open Robotics Community NVIDIA Blog source ↗

Jul 7 Australian Payments Plus uses ChatGPT Enterprise and Codex to move faster through payments complexity. OpenAI source ↗

Jul 7 MUFG aims to become AI-native using OpenAI; uses ChatGPT Enterprise to build AI-powered workflows and services at scale. OpenAI source ↗

Jul 7 Run AI workloads on any cloud and store on Hugging Face with zero-egress storage via SkyPilot Hugging Face source ↗

Jul 7 LeRobot v0.6.0: the platform updates to Imagine, Evaluate, Improve Hugging Face source ↗

Jul 6 Mistral AI reports incident on status page Mistral AI (via Google News) source ↗

Jul 6 OpenAI wealth-sharing promises draw attention as FT reports Altman discusses Americans’ stake in AI wealth MIT Technology Review AI source ↗

Jul 6 Open frontier models and open AI infrastructure become foundational to modern AI science, ICML 2026 reports; NVIDIA had 74 papers accepted at ICML 2026 NVIDIA Blog source ↗

Jul 6 PRX Part 4: our data strategy Hugging Face source ↗

Jul 6 Nations deploy AI for strategic priorities across infrastructure, economy, and society; investments accelerate innovation across sectors NVIDIA Blog source ↗

Jul 6 Government of Alberta uses Claude to find and fix cybersecurity vulnerabilities across government systems; Anthropic reports the use. Anthropic (via Google News) source ↗

Jul 6 x.ai introduces 21 new flagship Grok Voices xAI (via Google News) source ↗

Jul 6 30+ best nano banana prompts to copy and paste in 2026; Krea Krea (via Google News) source ↗

Jul 6 Krea lists the 10 best style prompts for AI image generation in 2026 Krea (via Google News) source ↗

Jul 6 8 best creative AI image generators in 2026; Krea highlights top options Krea (via Google News) source ↗

Jul 6 Kernels: major updates Hugging Face source ↗

Jul 3 Google DeepMind and A24 announce research partnership Google DeepMind source ↗

Jul 2 Achieving operational excellence with AI through Lean Six Sigma and BPM-based frameworks to bring order to complex operations. MIT Technology Review AI source ↗

Jul 2 AI becomes a core operating layer in industrial and infrastructure systems for enterprise use; foundation for an autonomous enterprise structures becomes clearer MIT Technology Review AI source ↗

Jul 2 AI Registry prompts API degraded; Mistral AI status page reports degradation Mistral AI (via Google News) source ↗

Jul 2 Leanstral 1.5: Proof abundance for all; mistral.ai Mistral AI (via Google News) source ↗

Jul 2 Anthropic shares more details on Fable 5’s cyber safeguards and their jailbreak framework; Fable 5 details from Anthropic Anthropic (via Google News) source ↗

Jul 2 Cohere releases Sinhala language model and updates streaming API; researchers report improvements in low-resource language generation Cohere (via Google News) source ↗

Jul 2 Nvidia Announces AI Compute Scaling Initiative; Invites Partners to Expand Infrastructure NVIDIA Blog source ↗

Jul 1 Google reveals its latest AI updates in June 2026. Google AI Blog source ↗

Jul 1 Google hosts AI summit with NYC educators and industry leaders to shape AI in classrooms Google AI Blog source ↗

Jul 1 A startup aims to break LLM groupthink by introducing novel prompting approaches; the article notes how current chatbots tend to give recurring outputs MIT Technology Review AI source ↗

Jul 1 Mistral.ai publishes pricing information Mistral AI (via Google News) source ↗

Jul 1 x.ai introduces the Voice Agent Builder xAI (via Google News) source ↗

Jul 1 Hugging Face and Cerebras release Gemma 4 for real-time voice AI Hugging Face source ↗

← 2026-08

2026-06 →