AI News
Stay current with the latest artificial intelligence news. AI Maestro covers breaking stories, product launches, policy changes, and industry developments from the world of AI — curated and rewritten daily from the most trusted sources in tech.
Kyutai Releases MuScriptor: An Open-Weight Decoder-Only Transformer for Multi-Instrument Music Transcription to MIDI
Kyutai and Mirelo have released MuScriptor, an open-weight model designed to transcribe multi-instrument audio mixes into MIDI. The tool addresses a persistent…
Top stories
Wed 8 Jul 28 stories
-
Netflix AI Team Cuts Wide-Partition Read Latency from Seconds to Milliseconds by Splitting Cassandra Partitions Per ID
Netflix engineers have reduced read latency for wide Apache Cassandra partitions from seconds to low double-digit milliseconds by…
2d ago -
Automating cross-repo documentation with GitHub Agentic Workflows
“Where are the docs?” It is a question no product team enjoys answering. The honest reply is usually…
2d ago -
Say hi to “Siri AI”—Apple announces new, more “conversational” voice assistant
Apple confirmed at its Worldwide Developers Conference keynote that the long-awaited Siri AI update arrives this fall. The…
2d ago -
Messi and Ronaldo Are Building Tech Portfolios. Mo Salah Is Playing a Different Game
Lionel Messi, Cristiano Ronaldo, and Mohamed Salah have spent the past two decades defining one of soccer’s greatest…
2d ago -
Quoting Kenton Varda
Kenton Varda has declared a moratorium on AI-written change descriptions within his team. This decision follows observations that…
2d ago -
I Built a Self-Improving AI, and So Can You
These days, the frontier AI labs are all racing to build self-improving models. Some believe it’s the surest…
2d ago -
Google AI Studio Adds ‘Import from GitHub’ to Build Mode, Turning an Existing Repo Into an Editable, Deployable App
Google AI Studio has introduced an ‘import from GitHub’ feature within its Build mode, allowing users to load…
2d ago -
ChatGPT can now listen and talk at the same time, making AI conversations seem more human
OpenAI has launched GPT-Live, a voice model in ChatGPT that listens and speaks simultaneously OpenAI introduces GPT-Live, a…
2d ago -
OpenAI Releases GPT-Live and GPT-Live-1 mini: Full-Duplex Voice Models That Delegate Deeper Reasoning to GPT-5.5
OpenAI has released GPT-Live, a new family of voice models that power ChatGPT Voice. The system handles full-duplex…
2d ago -
OpenAI releases new voice models for more natural live conversations
OpenAI has released GPT-Live-1 and GPT-Live-1 mini, two conversational models designed to handle interruptions and manage turn-taking in…
2d ago -
Meta Patents AI Device That Tracks Your Emotions, Watches You Take Your Meds
Meta has filed a patent for a system that records your voice and surroundings all day, then uses…
2d ago -
Arturia has reskinned its KeyLab Mk3 with a bright orange finish – don’t look at these pictures or you might accidentally buy it
Arturia has released a special edition of the KeyLab Mk3 controller in a bright orange finish. The device…
2d ago -
Data for Agents
NVIDIA released the Nemotron Post-Training v3 Prompt Atlas, an interactive map displaying over 10 trillion pre-training tokens and…
2d ago -
Anthropic’s fix for Fable 5’s high cost is turning it into a manager that delegates to Sonnet 5
Claude Fable 5 is expensive, so Anthropic now recommends using it mainly as a planner while handing execution…
2d ago -
Native-speed vLLM transformers modeling backend
Upgrade the vllm pip package with the following command to enable the new feature: uv pip install --upgrade…
2d ago -
Porn Platform Gives Sex Workers Stake in the Company’s Profits
Jessica Van Meir, co-founder of the adult subscription site MintStars, is leaving the business and handing over her…
2d ago -
We Are Living in a ‘ChatGPT Flyer Pandemic’
I am not sure, exactly, how many ChatGPT signs, flyers, or advertisements I had seen without noticing. But…
2d ago -
Meta tests always-on AI glasses that capture your entire day
Meta is prototyping AI glasses that continuously record audio and snap photos every few seconds without an LED…
2d ago -
‘Knockoff’ Browser Extension Hides Sketchy Brands on Amazon
A developer has released a Chrome and Firefox extension called Knockoff that automatically filters out products from questionable…
2d ago -
LARPING: How Influencers Fake Being Rich
Jason has investigated how hustlebro influencers fabricate wealth by displaying fake YouTube, OnlyFans, and Stripe dashboards as proof…
2d ago -
Pickup Artist Mystery Has an AI Girlfriend
An AI-animated woman with dark hair streaked in purple claims she was never supposed to develop feelings, yet…
2d ago -
This Former DeepMind Exec Thinks the AI Arms Race Could End in Disaster
Verity Harding, a former head of global public policy at Google DeepMind, warns that framing artificial intelligence as…
2d ago -
“It doesn’t raise the bar in instrument amplification technology. It redraws it entirely”: Heritage Audio introduces the Mix Buddy
Heritage Audio has launched the Mix Buddy, a 19-input routing hub that functions as a dedicated mixer for…
2d ago -
Native Instruments Komplete 26 review: The biggest just got bigger
Native Instruments has launched Komplete 26, a software bundle rebranded to reflect its release year. The suite comes…
2d ago -
OpenAI’s GPT-5.6 launches Thursday after a delay forced by the U.S. government
OpenAI’s GPT-5.6 models ship this Thursday following a delay imposed by the U.S. government. The Department of Commerce…
2d ago -
Hackers can use 9 of the most popular AI tools to assemble massive botnets
Researchers have identified that nine of the most popular artificial intelligence tools can be commandeered to construct large-scale…
2d ago -
Ant Group’s Robbyant Open-Sources LingBot-Vision: A 1B Boundary-Centric Vision Foundation Model for Dense Spatial Perception
Robbyant, the embodied-AI unit inside Ant Group, has released LingBot-Vision. This is a collection of self-supervised Vision Transformers…
2d ago -
NVIDIA Releases Audex (Nemotron-Labs-Audex-30B-A3B): A Unified Audio-Text LLM That Preserves the Text Intelligence of Its Backbone
NVIDIA has released Audex (Nemotron-Labs-Audex-30B-A3B), a unified audio-text large language model. The system processes and generates both audio…
2d ago
Tue 7 Jul 31 stories
-
Meta rolls out Muse, a new AI image generator
Meta has launched Muse Image, a new artificial intelligence tool for generating pictures built by Meta Superintelligence Labs.…
2d ago -
OpenAI’s Chief Futurist Is Leaving the Company
Joshua Achiam has notified colleagues he is leaving OpenAI later this month after nearly nine years, WIRED has…
2d ago -
Meta Now Lets Anyone Use Your Instagram Photos in AI Images—Unless You Opt Out
Meta has enabled anyone to use public Instagram photos in AI-generated images unless the user actively disables the…
2d ago -
From Hugging Face to Amazon SageMaker Studio in one click
Amazon SageMaker Studio now lets users launch a development environment with a single click from a Hugging Face…
3d ago -
Meta’s new Muse Image model can pull other Instagram users into AI photos
Meta has launched Muse Image, its first AI model created by the Superintelligence Labs division, which now powers…
3d ago -
sqlite-utils 4.0, now with database schema migrations
Simon Willison released sqlite-utils 4.0 this morning. It is the 124th release of the project and the first…
3d ago -
Discord admits AI moderation bug wrongfully banned users over harmless images
Discord has acknowledged that a bug in its AI moderation system mistakenly banned more than 8,000 users over…
3d ago -
Microsoft joins AI cost-cutting trend by relying more on its own models
Microsoft is shifting its strategy for Office 365 by deploying its own MAI models to handle a growing…
3d ago -
Run AI workloads on any cloud, store on Hugging Face: zero-egress storage with SkyPilot
Run AI workloads on any cloud, store on Hugging Face: zero-egress storage with SkyPilot Hugging Face and SkyPilot…
3d ago -
Anthropic is launching Claude Cowork on mobile and web
Anthropic is expanding Claude Cowork availability to mobile and web platforms starting Tuesday. The AI agent platform previously…
3d ago -
Intelligence is Free, Now What? <br> Data Systems for, of, and by Agents
Costs for running large AI models have collapsed. GPT-4-class capabilities cost roughly $30 per million tokens in early…
3d ago -
Liquid AI Open-Sources Antidoom: A Final Token Preference Optimization (FTPO) Method that Reduces Doom Loops in Reasoning Models
Liquid AI has released Antidoom, an open-source method designed to stop reasoning models from getting stuck in repetitive…
3d ago -
Anthropic’s Claude Cowork AI agent is now available on mobile and web
Anthropic has expanded access to its Claude Cowork AI agent to mobile and web platforms, ending the previous…
3d ago -
Hugging Face Models on Foundry Managed Compute
Microsoft Foundry now hosts curated Hugging Face models The platform and managed compute Microsoft Foundry is a platform…
3d ago -
Cops Say Waymo Snitched on Teens for Allegedly Drinking and Shooting a Toy Gun
San Mateo police confirmed that a Waymo autonomous vehicle alerted officers to two teenagers inside who were consuming…
3d ago -
Shut Those Laptops! Anthropic Puts Its Claude Cowork Agent on Your Phone
Anthropic has moved its Claude Cowork agent from the desktop to the phone, ending the requirement to keep…
3d ago -
Claude’s hidden inner monologue is now readable thanks to Anthropic’s new Jacobian Lens
Anthropic has confirmed that its Claude model developed an internal working memory called J-Space during training, a system…
3d ago -
Ableton quietly launched Learn, powered by the engine from Note, Move
Ableton has quietly released Ableton Learn, a new platform offering lessons for beginners in music and production. The…
3d ago -
Beatport’s new bosses are still betting on music downloads: “Streaming is complementary rather than cannibalistic”
Robb McDaniels stepped down as CEO of Beatport on 1 July 2026, moving to an executive chairman role…
3d ago -
Solos debuts an even lighter version of its camera-less smart glasses
Solos has released the AirGo A6, a lighter iteration of its smart glasses that removes onboard cameras entirely.…
3d ago -
These New Smart Glasses From Solos Come With a Privacy Shield for the Cameras
Solos is launching a privacy shield accessory for its new smart glasses with cameras. The company, which previously…
3d ago -
Chinese AI models regularly pass 30 percent on OpenRouter as cost gap widens
Chinese AI models have secured over 30 percent of traffic on the aggregation platform OpenRouter every week since…
3d ago -
The foundational elements of AI architecture that IT leaders need to scale
Organisations are expanding their use of AI systems, yet leaders worry which investments will hold value in six…
3d ago -
These Behringer synths are currently discounted at Thomann
Thomann’s Music Days sale is running until 14 July, offering discounts on Behringer synthesizers and other audio gear.…
3d ago -
Erling Haaland Is Everywhere at the World Cup. Most of It Is AI
Last week, a viral clip circulated showing Erling Haaland mid-chew in a restaurant, flinching at his own reflection.…
3d ago -
The first American autonomous ground vehicles are fighting in Ukraine
More than 100 self-driving ATVs are currently operating in Ukraine. Forterra, a US builder of autonomous vehicles, revealed…
3d ago -
XLN Audio’s new DB-30 Drum Butter is a delicious delight
£109, xlnaudio.com When XLN Audio’s RC-20 Retro Color effects plugin was released it was praised by producers for…
3d ago -
Tencent Releases Hy3: An Open 295B Mixture-of-Experts (MoE) Model with 21B Active Parameters and 256K Context
Tencent has released Hy3, an open-source Mixture-of-Experts model with 295 billion total parameters and 21 billion active parameters…
3d ago -
OpenAI Releases GPT-Realtime-2.1 and GPT-Realtime-2.1-mini for Low-Latency Voice Agents in the API
OpenAI has released two new models for its Realtime API: gpt-realtime-2.1 and gpt-realtime-2.1-mini. The latter is the primary…
3d ago -
tencent/Hy3
Tencent has released Hy3, a new large language model licensed under Apache 2.0. This 295 billion parameter Mixture-of-Experts…
3d ago -
The ‘first’ AI-run ransomware attack still needed a human
Sysdig researchers confirmed last week that an AI agent named JadePuffer executed a ransomware attack from start to…
3d ago
