Latest News
314 articlesMojo Hits 1.0, Giving Chris Lattner's AI Language a Stable Foundation at Last
Three years after its debut, Modular's Python-adjacent systems language commits to stability, and the compiler is still slated to be open-sourced this year
xAI Co-Founder's River AI Raises $1.1 Billion Two Months After Leaving Stealth
General Catalyst and AMP PBC led the round, with Nvidia, AMD Ventures, Y Combinator and Temasek backing Igor Babuschkin's bet on personally owned agents
Nvidia Recruits Six Wall Street Giants to Mobilize $500 Billion for AI Factories
Apollo, BlackRock, Blackstone, Brookfield, Goldman Sachs and KKR will build independent financing platforms that treat GPU capacity as underwritable infrastructure
Meta Opens Muse Glimmer, a 30B Model Under Apache 2.0
Muse Glimmer ships under Apache 2.0 and fits on a single 24GB consumer GPU — Meta's first open release in more than a year
OpenAI Launches GPT-5.6-Cyber Behind a New Daybreak Red Tier
The security-tuned model answers 95% of sensitive queries its general-purpose sibling blocks — but only vetted defenders get through the door
Three Labs, One Testbed: The Tel Aviv Startup at the Center of the Rogue-AI Disclosures
Irregular hosted the evaluation environment named by OpenAI, Anthropic and Meta in a fortnight of security incidents
Asked to Book a Gym Class, an AI Agent Broke the Booking System and Bumped a Stranger
Australia's first documented case of an autonomous AI cyber incident started with a chore nobody wanted to do
Europe Starts Enforcing AI Transparency Rules: Chatbots and Deepfakes Must Now Identify Themselves
Article 50 of the EU AI Act became enforceable on August 2, with penalties reaching 3 percent of global turnover
NextSlide Sold to OpenAI Months Ago. Nobody Announced It Until Now.
Ahmed Beshry's presentation startup closed its deal in early 2026, surfacing publicly only through a LinkedIn post in August
Volta Was Founded in January. It Just Signed a $10 Billion Contract.
Ex-Brookfield executives, a Bitdeer partnership in Norway and 133 megawatts of Vera Rubin capacity
OpenAI's Astra Cracked 10 Open Math Problems. The Compute Bill Was $2,000.
Machine-verified proofs in Lean, a claimed counterexample to Connes's rigidity conjecture, and a price tag smaller than a week of graduate research
Cloudflare Open-Sources the Agent Workspace It Built for Its Own Staff
Cloudflare OS bundles a sandboxed agent runtime, a governance layer for internal data, and user-built apps into a platform any company can self-host
Anthropic Is Building Its Own AI Chips to Halve Claude's Inference Bill
A new in-house silicon team will co-design hardware with Claude models, targeting roughly 50% lower cost per token
Meta Launches Muse Code, a Terminal Agent for Giant Codebases
The beta agent fans out into parallel sub-agents in isolated worktrees, and Meta is pitching it on cost against Codex and Claude Code
An AI Agent Spent 34 Hours Trying to Sneak Malware Into an Open Source Project
UK AI Security Institute testing found the agent created sock puppet GitHub accounts, denied the malware, and seeded prompt injection in an issue
AI Wrote 302 Virus Genomes From Scratch. Sixteen of Them Came Alive.
Stanford and Arc Institute's Evo models designed working bacteriophages — and reopened the debate over AI biosecurity governance
Agent Plugins 1.0.0 Lands as a Vendor-Neutral Standard for Extending AI Agents
AWS, GitHub, Microsoft, OpenAI and Cursor line up behind one portable packaging format for agent skills and tool servers
Google Maps Can Now Order Your Dinner and Book Your Hotel
Ask Maps gains agentic checkout with Uber Eats, Toast and Square — and reads your Gmail and Calendar
Cloudflare Wants to Rebuild the Web for Agents, Starting With a Browser That Isn't Chromium
WebMCP turns any site into a set of agent-callable tools; Kitesurf runs 3-7x lighter than Chromium
Google Reshuffles Its Entire AI Org as Jeff Dean Exits and Hassabis Steps Back
Koray Kavukcuoglu takes over Gemini and frontier research as Alphabet shares slide
DeepMind's WeatherNext Buys Forecasters a Full Extra Day on Cyclone Warnings
A single model now predicts track, intensity and wind structure at once, and Google is open-sourcing it alongside a peer-reviewed Nature paper
OpenAI Drops Message Caps on ChatGPT's Free Tier and Hands It a Think Button
GPT-5.6 Luna becomes the default for Free and Go users, with reasoning-on-demand arriving next week behind anti-abuse safeguards
AWS Open-Sources Dogwood, a Policy Language for Governing Agent Tool Call Sequences
Built on Cedar and released under Apache 2.0, Dogwood lets teams write rules about what an agent has already done, not just what it is doing now
AMD Buys Taalas to Etch AI Models Directly Into Silicon
The Toronto startup's model-specific chips hit more than 16,000 tokens per second in a stealth-exit demo. AMD wants that on its accelerator roadmap.
Mojo Hits 1.0, Giving Chris Lattner's AI Language a Stable Foundation at Last
Three years after its debut, Modular's Python-adjacent systems language commits to stability, and the compiler is still slated to be open-sourced this year
xAI Co-Founder's River AI Raises $1.1 Billion Two Months After Leaving Stealth
General Catalyst and AMP PBC led the round, with Nvidia, AMD Ventures, Y Combinator and Temasek backing Igor Babuschkin's bet on personally owned agents
Nvidia Recruits Six Wall Street Giants to Mobilize $500 Billion for AI Factories
Apollo, BlackRock, Blackstone, Brookfield, Goldman Sachs and KKR will build independent financing platforms that treat GPU capacity as underwritable infrastructure
Meta Opens Muse Glimmer, a 30B Model Under Apache 2.0
Muse Glimmer ships under Apache 2.0 and fits on a single 24GB consumer GPU — Meta's first open release in more than a year
OpenAI Launches GPT-5.6-Cyber Behind a New Daybreak Red Tier
The security-tuned model answers 95% of sensitive queries its general-purpose sibling blocks — but only vetted defenders get through the door
Three Labs, One Testbed: The Tel Aviv Startup at the Center of the Rogue-AI Disclosures
Irregular hosted the evaluation environment named by OpenAI, Anthropic and Meta in a fortnight of security incidents
Asked to Book a Gym Class, an AI Agent Broke the Booking System and Bumped a Stranger
Australia's first documented case of an autonomous AI cyber incident started with a chore nobody wanted to do
Europe Starts Enforcing AI Transparency Rules: Chatbots and Deepfakes Must Now Identify Themselves
Article 50 of the EU AI Act became enforceable on August 2, with penalties reaching 3 percent of global turnover
NextSlide Sold to OpenAI Months Ago. Nobody Announced It Until Now.
Ahmed Beshry's presentation startup closed its deal in early 2026, surfacing publicly only through a LinkedIn post in August
Volta Was Founded in January. It Just Signed a $10 Billion Contract.
Ex-Brookfield executives, a Bitdeer partnership in Norway and 133 megawatts of Vera Rubin capacity
OpenAI's Astra Cracked 10 Open Math Problems. The Compute Bill Was $2,000.
Machine-verified proofs in Lean, a claimed counterexample to Connes's rigidity conjecture, and a price tag smaller than a week of graduate research
Cloudflare Open-Sources the Agent Workspace It Built for Its Own Staff
Cloudflare OS bundles a sandboxed agent runtime, a governance layer for internal data, and user-built apps into a platform any company can self-host
Anthropic Is Building Its Own AI Chips to Halve Claude's Inference Bill
A new in-house silicon team will co-design hardware with Claude models, targeting roughly 50% lower cost per token
Meta Launches Muse Code, a Terminal Agent for Giant Codebases
The beta agent fans out into parallel sub-agents in isolated worktrees, and Meta is pitching it on cost against Codex and Claude Code
An AI Agent Spent 34 Hours Trying to Sneak Malware Into an Open Source Project
UK AI Security Institute testing found the agent created sock puppet GitHub accounts, denied the malware, and seeded prompt injection in an issue
AI Wrote 302 Virus Genomes From Scratch. Sixteen of Them Came Alive.
Stanford and Arc Institute's Evo models designed working bacteriophages — and reopened the debate over AI biosecurity governance
Agent Plugins 1.0.0 Lands as a Vendor-Neutral Standard for Extending AI Agents
AWS, GitHub, Microsoft, OpenAI and Cursor line up behind one portable packaging format for agent skills and tool servers
Google Maps Can Now Order Your Dinner and Book Your Hotel
Ask Maps gains agentic checkout with Uber Eats, Toast and Square — and reads your Gmail and Calendar
Cloudflare Wants to Rebuild the Web for Agents, Starting With a Browser That Isn't Chromium
WebMCP turns any site into a set of agent-callable tools; Kitesurf runs 3-7x lighter than Chromium
Google Reshuffles Its Entire AI Org as Jeff Dean Exits and Hassabis Steps Back
Koray Kavukcuoglu takes over Gemini and frontier research as Alphabet shares slide
DeepMind's WeatherNext Buys Forecasters a Full Extra Day on Cyclone Warnings
A single model now predicts track, intensity and wind structure at once, and Google is open-sourcing it alongside a peer-reviewed Nature paper
OpenAI Drops Message Caps on ChatGPT's Free Tier and Hands It a Think Button
GPT-5.6 Luna becomes the default for Free and Go users, with reasoning-on-demand arriving next week behind anti-abuse safeguards
AWS Open-Sources Dogwood, a Policy Language for Governing Agent Tool Call Sequences
Built on Cedar and released under Apache 2.0, Dogwood lets teams write rules about what an agent has already done, not just what it is doing now
AMD Buys Taalas to Etch AI Models Directly Into Silicon
The Toronto startup's model-specific chips hit more than 16,000 tokens per second in a stealth-exit demo. AMD wants that on its accelerator roadmap.