AI News
Short, sourced, and free of hype. Page 11 of 20.
A Free Command Line Tool Now Shows You What a Model Is Thinking, and Talks to Almost Any Model
Simon Willison released LLM 0.32, the biggest update to his open source AI tool yet: visible reasoning traces, server-side tools and a one-liner for any OpenAI-compatible endpoint.
8 Aug 2026
OpenAI Says Its Next Model Might Hit the Highest Cyber Risk Level, and Paused Part of the Work
Internal tests of the upcoming Astra model showed cybersecurity skills strong enough that OpenAI can no longer rule out the Critical level in its own safety framework.
8 Aug 2026
Suno Tightens Download Limits After a German Court Ruled Against It
The AI music generator is limiting bulk downloads and adding AI labelling, after a Munich court found it trained on copyrighted songs and an investor conceded it competes with human artists.
8 Aug 2026
Same model, four different tools, three times the price: a test of what the wrapper costs you
Composio ran DeepSeek V4 Flash through Claude Code, Codex, OpenCode and Oh My Pi on 30 real tasks. Success rates were similar. Cost varied by nearly 3x and speed by 2.2x, depending only on which tool ran the model.
7 Aug 2026
ChatGPT drops its daily limit for free users and swaps in a more accurate default model
OpenAI is removing the cap on text chats for free ChatGPT accounts, making GPT-5.6 Luna the default for everyone, and giving Plus and Pro subscribers a slider that controls how hard the model thinks.
7 Aug 2026
Anthropic loosens Claude Fable 5's biology filter, cutting blocked questions by 85 percent
Fable 5 launched with almost all biology questions blocked. A retrained safety classifier now lets through everyday health and study questions, reducing biology-related fallbacks by about 85 percent. Professional virology and drug design stay blocked.
7 Aug 2026
Google Assistant switches off on 4 September, and there is no way back
Google is retiring Assistant on Android, Wear OS, Android Auto and headphones starting 4 September 2026. Gemini takes over, the rollout is one way, and cars with Google built-in keep Assistant for now.
7 Aug 2026
A startup just raised 30.5 million dollars to put ads inside chatbots, and its next product advertises to the bots
Gravity closed a 30.5 million dollar Series A to run an ad network across ChatGPT and other AI assistants. Its agent-to-agent product feeds product catalogues to AI shopping agents, with no human ever seeing the ad.
7 Aug 2026
Alibaba's newest model scores higher and guesses more: hallucination rate jumps from 23 to 40 percent
Qwen3.8 Max gained 10 points on the Artificial Analysis Intelligence Index, matching Claude Opus 4.8. It also got slower, more expensive per task, and far more willing to invent an answer rather than admit it does not know.
7 Aug 2026
OpenAI's first gadget is reportedly a moving, camera-equipped speaker the size of a hockey puck
Bloomberg reports OpenAI's long-rumoured device is a ring-shaped, battery-powered smart speaker with no screen, a camera, sensors and parts that move, priced at 300 to 400 dollars for a 2027 launch.
7 Aug 2026
When AI writes a kids' story about animals, the female characters almost disappear
A University of Washington study ran nearly 24,000 story completions through six leading models. Two percent of the animal characters came out female, 41 percent male, and 57 percent neutral or ungendered. The researchers argue the guardrails meant to prevent bias caused it.
6 Aug 2026
Cloudflare wants to give your AI agent a wallet, and you the spending limit
Cloudflare Wallets will let AI agents pay for APIs and content directly, with allowances, allowlists and hard caps set by the human owner. Handles are claimable now, stablecoin funding and x402 micropayments are planned.
6 Aug 2026
Since 2 August, AI in Europe has to say it is AI
The EU AI Act's transparency rules are now in force and the Commission's AI Office can enforce them. Chatbots must identify themselves, deepfakes must be labelled, and AI generated content needs machine readable marks. Fines reach 15 million euros or 3 percent of global turnover.
6 Aug 2026
Google reshuffles its AI leadership, and four of its most famous engineers walk out the door
Demis Hassabis moves from Google DeepMind CEO to Alphabet chief scientist, while Jeff Dean, Sanjay Ghemawat, Oriol Vinyals and Quoc Le leave to start a company called Discovery Loop. The flagship Gemini model is still not out.
6 Aug 2026
Meta ships Muse Code, a coding agent that keeps working after it crashes
Meta released Muse Code in beta for macOS and Linux, powered by a new Muse Spark 1.2 model. Its two interesting ideas are background agents that stay alive for a whole session and a local event log that lets an interrupted job resume exactly where it stopped.
6 Aug 2026
Most of Microsoft's AI revenue is just OpenAI paying Microsoft
A new Microsoft filing shows 24.1 billion dollars of revenue from OpenAI in the fiscal year ended June, which analysts read as roughly 70 percent of Microsoft's AI business. Outstanding receivables from OpenAI stood at 6 billion dollars.
6 Aug 2026
An open model is now months, not years, behind the frontier. Its safety testing is not.
A SaferAI report finds Z.ai's open-weight GLM-5.2 sits only a few months behind the leading closed models on cyber and biology capabilities, and refused none of the offensive tasks it was given. Claude Opus 4.7 refused so consistently the test could not be completed.
6 Aug 2026
OpenAI ships three education plugins, and quietly admits most students barely use the tools they have
New plugins for K-12 teachers, college educators and college students land in ChatGPT Edu and ChatGPT for Teachers. Buried in the announcement: even advanced student users tap ChatGPT's capabilities 90 to 99 percent less than power users.
6 Aug 2026
Anthropic just signed a 10 billion dollar compute deal with a company that did not exist six months ago
Anthropic booked six years of computing capacity from Volta, a cloud startup founded in early 2026, at a Norwegian hydropower data centre. On the same day, reporting showed Google shifting billions in Anthropic related chip risk off its own balance sheet.
5 Aug 2026
Someone got DeepSeek V4 Flash running in production on a single AMD card and published every patch it took
A public repository documents the configuration, patches and tuning needed to serve DeepSeek V4 Flash on one AMD MI300X, unquantised. It is a useful record of how much work running an open model on non Nvidia hardware still is.
5 Aug 2026
Researchers asked why chatbots are bad at spreadsheets, and the answer is stranger than expected
A new paper tests five explanations for why large language models lose to decades old methods on tabular prediction. In low dimensions the model behaves like a simple distance based classifier. In higher dimensions nothing classical explains it.
5 Aug 2026
Mistral released a free safety filter that you describe in plain English, and it fits on one graphics card
Shieldstral is a 3 billion parameter open-weights model that checks text and images against a policy you write yourself, in ordinary sentences, without any retraining. It runs on a single 16 GB GPU under an Apache 2.0 licence.
5 Aug 2026
A record eight Pulitzer winners and finalists disclosed using AI, and what they used it for is the interesting part
Eight Pulitzer awardees declared AI use this year, the most since disclosure became mandatory in 2024. In almost every case the tool was used to search or translate large document piles, not to write.
5 Aug 2026
An 80 billion parameter model now runs on a normal Mac in 4.3 GB of memory, and a 35B one runs on an iPhone
Swiftlet is an open runtime that streams a model's expert weights from the SSD instead of loading them into memory. A Qwen model with 80 billion parameters peaks at 4.3 GB of RAM on a Mac. The trade off is speed and a large download.
5 Aug 2026