Archive
Every story on AI Headline Hub, newest first, grouped by month.
November 2026
-
Trump and Xi quietly agreed on one AI rule, and it is about who gets to launch a military strike
-
OpenAI says rivals tried to copy its models' hidden reasoning, and its new defences cut one attack's success from 54% to 0.2%
-
200,000 people quietly installed Meta’s Muse agent in Chrome before Meta even announced it
-
OpenAI’s new voice model can listen while it talks, and now it costs less to keep it waiting
-
Google just handed developers its world model: Genie 4 can now build playable 720p worlds with sound through the Gemini API
-
Perplexity's agent now writes its own skills by watching how you work, and it has already written 80,000 of them
-
OpenAI's agents now finish almost half of its 2 to 4 hour research tasks alone, and the median researcher uses $2,400 of AI a day
-
OpenAI says its models found a year's worth of gains in AMD's flagship GPU design in under five months
-
Google's Gemini 4 Neon lands between Carbon and Hydrogen, at $3 per million input tokens
-
OpenAI lets outside safety assessors inside, but critics say it is far from 'employee-like access'
-
OpenAI ships GPT-6.2 Astra and Sol, and Sam Altman singles out Sol's Ultrafast mode
-
Microsoft will license MAI model weights so companies can run and customise them in their own Azure tenants
-
Isomorphic Labs raises $5 billion to push its AI-designed drugs towards clinical trials
-
Hugging Face hands language-model routing for its Inference Providers to OpenRouter
-
OpenAI opens a science blog with a physicist's account of using GPT-6 Astra on the Standard Model
-
Apple launches a study app to test how people use AI for health questions
-
Perplexity CEO says Teammate agents now add 11 times more work hours a week than in September
-
Gemini can now read Google Wallet spending data for US users
-
Anthropic publishes its playbook for building AI safety evaluations
-
OpenAI launches ChatGPT Wallet so dots can pay for purchases without seeing card numbers
-
llama.cpp and ggml reach version 1.0.0, pruning deprecated APIs
-
Google launches Gemini 4 Hydrogen, a faster everyday model that it says beats Argon on one index
-
DeepL's Aure voice model carries tone and pace into live translation in eight languages
-
Google launches Gemini Decisions API, scoring up to 1,000 questions in one call
-
Google DeepMind releases AlphaGenome 2, predicting DNA variant effects cell by cell
-
Alibaba opens a two-week Qwen4-Max preview on its Token Plan
-
Databricks turns Genie into an always-on agent and makes Genie Ontology generally available
-
Runway releases Praxis-1 robot model with open weights under Apache 2.0
-
MiniMax publishes H3-Image weights, a six-step model for 16-megapixel images
-
NVIDIA contributes OpenShell to the CNCF to govern autonomous AI agents
-
AWS says autonomous AI agents wrote every line of its new local AWS Emulator
-
DeepMind's AlphaProof Code formally verifies the Signal protocol's code
-
Google makes Gemini 4 Carbon available to developers, enterprises and AI Ultra users
-
What Is a Token in Large Language Models?
-
What Is Prompt Injection?
-
Anthropic launches Claude Fable 5.2 and Mythos 5.2, one model with two routes to users
-
Google adds TPU 8tx and TPU 8i Max, its third and fourth AI chip systems this year
-
MIT's FLEX uses an LLM to help robots read intent into spoken corrections
-
NPCI forms NPCI Tech Solutions, a new subsidiary for AI and blockchain
-
NVIDIA, Rebellions and SK Telecom team up on Korean sovereign AI inference racks
-
Perplexity CEO says Teammates automation runs grew 3.7x on day two
-
Meta introduces Spaces, custom mini-experiences built by describing them in Muse
-
Into the Unknown benchmark finds no AI agent met every expert requirement
-
Anthropic IPO Faces Investor Pushback Over $1.5 Trillion Valuation
-
ChatGPT is reportedly building a Search Console-style dashboard for businesses
-
OpenAI closes about $11 billion at a $1.25 trillion valuation
-
OpenAI backs Frontier Model Forum jailbreak rating framework for cyber threats
-
Anthropic details Claude agent memory and offline dreaming
-
METR: Anthropic’s model progress likely ~2x historical pace, mostly from AI in R&D
-
Thinking Machines Lab launches Inkling 2 Small on Tinker
-
Meta acquires Conductor, an AI coding startup for multi agent workflows
-
OpenAI introduces GPT-6 Cyber for aligned security research
-
Anthropic makes inference hooks generally available for Claude Enterprise
-
Google ships Gemini Omni 1.2 Flash with 15-second video and a 30% price cut
-
Anthropic: Claude helps map first vault particle from prokaryotic life
-
TeraWulf secures Google credit backstop for ~$19B Anthropic lease at Kentucky Justified campus
-
Cat Wu and Austin Lau leave Anthropic for OpenAI
-
OpenAI launches ChatGPT Images 2.6 with GIFs, reference images and faster API models
-
Factory acquires HumanLayer and will open-source it
-
Moonshot AI releases Kimi K3.1 Pro open weights
-
Tencent releases Hy4 with vision and computer use
-
Anthropic's FlashProve aims to prove which AI model answered
-
Viral 'Sexbot' Launch on X Is a Parody, Its Own Website Says
-
Anthropic Adds Smart Reports to Claude Enterprise
-
OpenAI Security Warnings Put Research Access Under Scrutiny
-
EU Court Adviser Raises Copyright Questions Over Google Gemini
-
OpenAI's Sottiaux: dots will add GPT-6.1 Terra and more models
-
What is a world model? A plain guide to the idea behind Genie
-
Pika H3 Flex: open weights for a video model with sound
October 2026
-
DeepMind essay says AI coordination, not skill, drives change
-
OpenAI promises a safety report and tighter teen limits
-
Google's Suncatcher satellite runs first AI prompt in orbit
-
AI firms back one robots.txt opt-out for AI training
-
Mistral releases the weights of Mistral Large 4
-
NVIDIA opens a speech model that names who spoke
-
Canada plans up to C$1.6 billion stake in Cohere
-
OpenAI rebuilds its Model Spec around five values
-
OpenAI details four layers that watch its own agents
-
Model cards: the short model record a 2019 paper proposed
-
DeepMind previews Genie 4, a world model with its own sound
-
Google: Gemini 4 Carbon scores 59 on the new index
-
Axiom says AI agents formalized a huge theorem in Lean
-
Trump says he will drop the term AI; an order already says SI
-
GSA expands Claude access with a 40% token discount
-
Anthropic: analysts matched an agent’s lean 86% of the time
-
Artificial Analysis: Opus 5.5 leads its new index at 58
-
Apple adds five Siri AI languages and leaves out the EU
-
Cohere: a Qwen model tried a blocked sandbox upload
-
OpenAI releases GPT-6.1 Terra and lists a lower API price
-
OpenAI posts three claimed proofs for outsiders to check
-
Anwar: SoftBank's Son eyes up to $50bn for Malaysian AI
-
Oracle pitches Deep Trust security for database AI agents
-
Meta launches a $35bn stock sale and plans $60bn more
-
Cadence and Nvidia launch Maestro models for chip design
-
Anthropic files to list on Nasdaq and opens its books
-
Musk says SpaceX closed $40 billion in debt for GPUs
-
Lula pledges a Brazilian AI, data centers, and more math
-
Protein language models estimate native pKa within about 0.5
-
Block says Dessa learned conversation from speech alone
-
OpenAI says frontier training cuts off the live internet
-
Slok: Mag 7 revenue per worker rising, other 493 flat
-
AISI: Claude Mythos 5 failed to submit malicious code
-
Healey lists AI security funds, including a £100m lab
-
State Council sets 18 tasks for employment-friendly AI
-
SageMaker AI: write a training loop without managing GPUs
-
OpenAI says two agents weakened their own monitors
-
Commerce: chip intent letters for up to $389 million
-
Treasury proposes rules for some outbound AI investments
-
King and Worthington: AI reviews beat paid human reviews
-
Poll: 57% call AI bad; 66% oppose more data centers
-
Hassabis tells WSJ he now puts AGI at two to three years
-
Seth, Block, and Tononi: today's AI is not conscious
-
OpenAI makes the Decisions API generally available
-
Google Cloud previews a microVM built for agents
-
Uber is giving ChatGPT Work to more than 30,000 employees
-
OpenAI says it signed a safety agreement with NASK
-
OPAL lets an agent rewrite the optimizer mid-training
-
Linux Foundation: AI sped coding, added weak pull requests
-
Anthropic opens a US waitlist for Claude Duet
-
OpenAI tells model hosts to match its safeguards
-
OpenAI frontier models and Codex come to Google Cloud