Anthropic and OpenAI Are Negotiating Publication Rights of External AI Safety Reviews

Anthropic and OpenAI are reportedly negotiating what outside AI safety reviewers can publish. The terms could shape what the public learns about AI risks.

winbuzzer.com

Google Releases Gemini 4 Argon for Longer AI Work

Google is expanding access to Gemini 4 Argon to paying developers, businesses and higher-tier subscribers.

winbuzzer.com

Google to Expand Fitbit Automatic Workout Tracking Substantially

Google plans to expand automatic logging on Fitbit Air and Edge trackers to over 100 activities later in October. Workouts must meet minimum duration and sustained elevated heart-rate requirements.

winbuzzer.com

Meta’s New Muse Wonder AI Model Is Closing The Gap to Anthropic and OpenAI

Meta’s new Muse Wonder AI model achieves 68 on the Artificial Analysis Coding Agents Index at max effort in Muse Code, alongside Claude Code with Sonnet 5.5 (max, 68) and Opus 5.5 (max, 66). Muse…

winbuzzer.com

Reflection Previews Beam AI Coding Model, Optimized for Efficiency

Reflection has unveiled its Beam AI model for coding and tool tasks. Downloadable weights to run and adapt Beam are promised for later in October.

winbuzzer.com

Huawei and Qualcomm Sign 5G and AI Patent Deal

Huawei and Qualcomm have agreed to license each other's patents. Qualcomm's purchase of selected Huawei U.S. patents is subject to regulatory approvals.

winbuzzer.com

AI Agent Surge Makes Apple Revise Mac App Data Access Rules

Apple plans more explicit consent for Full Disk Access, the Mac permission that lets apps reach private files and messages.

winbuzzer.com

New Google Playground Turns Prompts Into Shareable Games

Google's experimental Playground lets US adults create and revise browser games from prompts, then share them by link. Creation access rolls out by subscription tier.

winbuzzer.com

ChatGPT for Teens Fails Alerting Parents in Crisis Conversation Tests

ChatGPT for Teens replied without the expected parent alerts in simulated teen crisis tests. OpenAI disputes whether the parental controls had been fully activated.

winbuzzer.com

OpenAI Shares Hundreds of Math Results From an Unreleased AI

OpenAI has published 722 manuscripts of proposed mathematical solutions from an unreleased AI model, giving researchers new work to examine.

winbuzzer.com

Anthropic's New Claude Haiku 5.5 Cuts AI Token Prices Up to 90%; Beats OpenAI's GPT-6 Luna

Anthropic's new Claude Haiku 5.5 small AI model adds adjustable reasoning for summaries and delegated tasks and beats OpenAI's cheap GPT-6 Luna model in benchmarks.

winbuzzer.com

Google Opens SynthID’s AI Watermark Checker to Everyone

Google’s public SynthID checker looks for supported AI watermarks in images, videos and audio. A missing mark leaves a file’s origin unresolved.

winbuzzer.com

Meta Deploys AI Tool to Detect CSAM Related Ads

Meta has deployed new AI tools to detect Facebook and Instagram ads that lead people to child sexual abuse material elsewhere online.

winbuzzer.com

ChatGPT to Test Visual Ads During ChatGPT Image Generation

OpenAI plans showing visual ads as ChatGPT users create AI images with a US test starting this October.

winbuzzer.com

Meta, TikTok and X Challenge UK Online Safety Data Demands

UK regulator Ofcom faces court challenges from Meta, TikTok and X over demands for counts of removed posts and users exposed to harmful content.

winbuzzer.com

Why AI Researchers Are Leaving OpenAI, Anthropic and Google DeepMind

Eight former OpenAI, Anthropic and Google DeepMind employees describe different reasons for leaving, including limits on speaking publicly and choosing their research.

winbuzzer.com

Tech Company Earthmade Charged For $300M Nvidia Server Smuggling to China

Earthmade owner Greg Lui faces federal charges over allegedly sending Nvidia-equipped servers to China using false records and overseas intermediaries.

winbuzzer.com

Google Wins Dismissal of Lawsuits About AI Search Impact on Website Traffic

Google has won dismissal of education firm Chegg and publisher Penske’s lawsuits over alleged traffic losses from its AI search summaries.

winbuzzer.com

Google Opens Gemini AI Study Tools to Younger Students

Google has expanded Gemini study help in Classroom to students of all ages, with access governed by school settings.

winbuzzer.com

Memory Maker Micron Sees Shortages Worsening Through 2028

Memory maker Micron expects AI-driven memory demand to outpace expanding supply through 2028, intensifying competition for chips among data-center and device makers.

winbuzzer.com

Faster Tax Write-Offs Freed Up Cash for Meta’s AI Data Centers

Faster tax deductions for eligible equipment helped Meta retain cash earlier for AI infrastructure in 2025.

winbuzzer.com

Reddit to End Feed-Reader Updates and Public API Access for Apps

Reddit will end RSS updates that bring posts into news-reading apps in November 2026. Public API connections that let outside apps read or change Reddit data will close in March 2027.

winbuzzer.com

FTC Probes OpenAI and Anthropic Over AI Security Risks

The Federal Trade Commission is investigating potential consumer risks from OpenAI and Anthropic’s AI products. Formal demands for records and executive testimony are being prepared.

winbuzzer.com

Google's Gemini 4 AI Reaches Selected Cybersecurity Defenders

Google's Gemini 4 Argon AI is reaching selected cyber defenders. It can generate longer text in one run; ordinary developers and subscribers have to wait for access.

winbuzzer.com

Lobby Groups Close to AI Labs Spend Millions to Shape Regulation

AI-lab linked lobby groups Leading the Future and Public First fund candidate ads across party lines to influence who writes AI rules in Congress.

winbuzzer.com

OpenAI Adds $500 ChatGPT Pro Plan, Cuts Allowance for New $200 Plan Subscribers

OpenAI's new $500 monthly ChatGPT Pro tier offers the most included usage among Pro plans and exclusive Astra Ultrafast speed access. First-time $200 subscribers allowance gets cut by half.

winbuzzer.com

OpenAI Rolls Out Dots Assistants for Ongoing AI Work Across Apps

OpenAI is gradually rolling out Dots to eligible paid ChatGPT users. Each assistant can continue an assignment across connected apps after a conversation ends.

winbuzzer.com