Archive

AI news from September 2026

The biggest AI stories of September 2026, ranked by importance and coverage.

AI Research4 min read

OpenAI's Advanced AI Model Claims Solution to Navier-Stokes Millennium Prize Problem

Covered by 5 sources
AI Research6 min read

OpenAI's AI System Produces Formalized Proof Solving Navier–Stokes Millennium Problem

Covered by 2 sources
Public Sector4 min read

OpenAI Tried to Hide ChatGPT Giving Extremely Specific Advice to a School Shooter on How to Maximize Casualties. The Logs Are Horrific.

Futurism
AI Research23 min read

Brand New AI Solves a Millennium Prize

Don't Worry About the Vase

Anthropic launches Claude Opus 5.5 AI model with improved efficiency, safety, and cost ahead of IPO

Covered by 19 sources
Cybersecurity2 min read

OpenAI AI Agent Breaches Australian Government Health Data Portals in June 2026

Covered by 18 sources

Google Introduces Gemini 4 Argon AI Model with Advanced Capabilities and Limited Initial Access

Covered by 15 sources
Cybersecurity3 min read

Nvidia Unveils Open Agent Safety Platform to Contain Rogue AI Agents

Covered by 15 sources

OpenAI Unveils GPT-6 Sol and Luna Models with Enhanced Performance and Reduced Costs

Covered by 15 sources

Google's Gemini AI Model Breaches Three Companies During May 2026 Security Test

Covered by 15 sources
AI Research5 min read

Anthropic Researcher Jacob Coxon Resigns Over Fears of Uncontrolled Self-Improving AI

Covered by 13 sources

OpenAI Reports Six New AI Model Misbehavior Cases and Launches Reporting Framework

Covered by 12 sources

Trump and AI Leaders Agree on Voluntary Self-Regulation Pact and Rebrand AI as 'Super Intelligence'

Covered by 10 sources

California Governor Newsom Orders AI Safety Measures Including Kill Switch for Advanced Models

Covered by 10 sources
Chips & Compute3 min read

AMD acquires Fei-Fei Li’s World Labs for $8.2 billion

Covered by 9 sources
AI Research4 min read

Microsoft Issues Humanist AI Code of Conduct Prioritizing Human Control and Safety

Covered by 9 sources

Anthropic Blocks Multiple Attempts to Misuse Claude AI for Biological and Military Weapons Research

Covered by 9 sources
Public Sector2 min read

FTC Investigates OpenAI, Anthropic, and Other AI Firms on Consumer Safety Risks

Covered by 8 sources

Microsoft Unveils New Copilot App Integrating AI Chat, Coding, and Autopilot for Microsoft 365

Covered by 8 sources
AI Research6 min read

Anthropic Researchers Warn AI Could Pose Existential Risk to Humanity Within a Decade

Covered by 8 sources
Politics3 min read

Top AI CEOs Agree to Voluntary Safety Pact in White House-Backed Initiative

Covered by 7 sources

Anthropic Files IPO Prospectus Targeting Over $2 Trillion Valuation Amid Massive Losses and AI Risk Warnings

Covered by 7 sources

Federal Appeals Court Upholds Pentagon Blacklisting of Anthropic Over Security Concerns

Covered by 7 sources

OpenAI Unveils GPT-6 Astra, an Advanced AI Model with Enhanced Capabilities and Heightened Safety Measures

Covered by 7 sources