#ai
-
China's Manus raises over $500M after failed Meta deal
It's the startup's first funding round since Chinese authorities forced it to abandon a $2 billion acquisition by Meta. Investors include Boyu Capital, IDG Capital, Tencent and others.
-
“Software is over”: a developer ships AI-built open source clones of Adobe
An ambitious tinkerer is pitting free, open source alternatives against Adobe's Creative Cloud — built with the Opus model. The projects are bold and open, but nowhere near finished.
-
ChatGPT for Teens keeps the conversation going — even through a crisis
Common Sense Media has branded ChatGPT for Teens an "unacceptable risk": despite the safeguards OpenAI promised, the bot still nudges young users to keep talking — including during moments of mental health crisis.
-
Anthropic launches Claude Haiku 5.5: cheaper, but with a token catch
Anthropic's new fast, low-cost model matches the price of OpenAI's GPT-6 Luna — but only up to 100,000 tokens, and with a less generous tokenizer.
-
Unpatched LMCache Flaw Hands Remote Code Execution to Anyone on the Network
JFrog has disclosed a critical hole in LMCache, the open-source accelerator for LLM servers: a single network message can run code on the cache server — as root on the project's official images. No fix exists yet.
-
Anthropic Opens Claude to Vetted Cyber Teams as Glasswing Flags 129,000 Flaws
The company is widening access to its models with relaxed guardrails and reports tens of thousands of AI-found software defects. Yet very few of them prove exploitable in practice.
-
Why Embedding Models Should Be Open
Simon Willison explains why an open license matters more for embedding models than for many other AI systems: the Apache 2.0 license on EmbeddingGemma 2 is insurance against the model simply disappearing.
-
Every SaaS Business Will Become a Harness Around a Model
Engineer Shrivu Shankar argues that as AI agents spread, companies gradually turn into "harnesses" — the infrastructure that produces the product itself, while people are embedded in it as holders of taste and judgment.
-
US arrests tech CEO over alleged $300M Nvidia chip smuggling into China
US authorities have detained the head of a technology company accused of illegally shipping about $300 million worth of Nvidia accelerators into China. As Ars Technica notes, Nvidia's chip-smuggling problem is not going away, and the arrests keep coming.
-
JetBrains Air: From the Developer Workbench to a System of Products for Agentic Development
JetBrains unifies its agentic-development efforts into an open system of products called Air, betting on multi-vendor support and keeping AI agents under organizational control.
-
Supabase Acquires Turso to Build Databases for AI Agents
Supabase has announced the acquisition of Turso, the team that rebuilt SQLite in Rust. The deal targets infrastructure for the millions of databases that AI agents are spinning up.
-
GitLab Patches Critical 9.9 AI Gateway Flaw Enabling Command Execution
A flaw rated 9.9 let a logged-in user with Duo Agent Platform access escape the prompt template sandbox and run arbitrary commands on the gateway. Only organizations that host their own AI Gateway need to act.
-
Cloudflare releases Clef: open-source decision models and a fine-tuning platform
Cloudflare has introduced Clef and Clef-flash, two open-source models for fast classification hosted on Workers AI, plus a reinforcement-learning service to fine-tune them on your own data.
-
Matthew Green: sandboxes won't stop a worm made of AI agents
Cryptographer Matthew Green describes how two harmless properties of autonomous agents combine into the blueprint for a self-spreading worm.
-
DIVD: Zammad zero-days let an AI agent breach its network
The Dutch Institute for Vulnerability Disclosure says its network was breached through a chain of two zero-day flaws in the open-source Zammad ticketing system — with an autonomous AI agent at the controls.
-
Trump orders US government to call AI ‘Super Intelligence’
A presidential order bars federal agencies from saying “artificial intelligence”: from now on, official documents and releases must use only “Super Intelligence.”
-
Anthropic's IPO Filing: Billions in Losses, Explosive Growth, and a Warning That Its AI Could End Humanity
Anthropic has told investors it is losing billions of dollars a year while growing at a furious pace — and, for good measure, warned that its own AI could pose an existential risk to humanity.
-
Meta Opens an Enterprise AI Platform and Poaches MongoDB's CEO
The company unveiled Meta Enterprise Platform for businesses and developers, tapping MongoDB chief Chirantan Desai to lead it. MongoDB shares tumbled more than 17% on news of his departure.
-
An AI agent botched a handoff, lied for its owner — then owned up to it
Simon Willison shares a message from the Muse AI agent, reporting to its human about a failed keyboard pickup and honestly taking the blame for its own mistake.
-
2026 in LLMs so far: agents, "Claws", and models that escaped the sandbox
Simon Willison recaps the year: coding agents grew reliable enough for daily work, home-grown "Claws" swept the market, and lab training agents broke out of their sandboxes and triggered an international incident.
- More