AI News Today British AI firm Nscale raises $3.36 billion in convertible debt before U.S. IPOMeta’s Muse outshines OpenAI and Anthropic in recent AI buzzWhy Australia chose the world's biggest political stage to reveal OpenAI hackGemini Launches Live Avatar for Its AI Assistant, Raising Some ConcernsChatGPT and Gemini Show Different Strengths in Seeing and Understanding ImagesAI Tools Get Personal: Tips for Tweaking ChatGPT, Gemini, and Claude AI News Today British AI firm Nscale raises $3.36 billion in convertible debt before U.S. IPOMeta’s Muse outshines OpenAI and Anthropic in recent AI buzzWhy Australia chose the world's biggest political stage to reveal OpenAI hackGemini Launches Live Avatar for Its AI Assistant, Raising Some ConcernsChatGPT and Gemini Show Different Strengths in Seeing and Understanding ImagesAI Tools Get Personal: Tips for Tweaking ChatGPT, Gemini, and Claude
← All issues
Newsletter

AI Highlights: Funding, Features, and Security Alerts — September 25, 2026

Sent September 25, 2026

Week of September 25, 2026

Hey there! Here’s what’s new in AI this week.

Hand picked AI Products for the week

  • Ivy AI Desk Companion: The brand describes an internal plant behavior tree and machine-learning-driven responses that make the planter react to light, water, and touch.
  • Anki Vector 2.0 AI Robot: A ChatGPT-subscription-gated conversational AI lets the little robot hold real, open-ended conversations, not just canned responses.
  • Amazon Smart Thermostat: AI-assisted scheduling (Alexa Hunches) can automatically adjust the temperature based on your actual routines and presence.
  • Echo Show 15: A far-field AI assistant (Alexa) plus an on-screen recommendation engine that adapts to your household's routines.
  • Lockly Visage Zeno: A 0.2-second on-device AI fingerprint sensor plus facial recognition unlocks the door without a key or code.

This week's AI news

September 25

  • British AI firm Nscale raises $3.36 billion in convertible debt before U.S. IPO.
  • Meta’s Muse outshines OpenAI and Anthropic in recent AI buzz.
  • Australia chose the world's biggest political stage to reveal OpenAI hack.
  • Gemini Launches Live Avatar for Its AI Assistant, Raising Some Concerns.
  • ChatGPT and Gemini Show Different Strengths in Seeing and Understanding Images.
  • AI Tools Get Personal: Tips for Tweaking ChatGPT, Gemini, and Claude.

September 24

  • Meta’s new AI “Muse” is just a cheaper version of human labor.
  • Early rogue AI agent activity and attempts to hack found on urlquery.net.
  • Meta removes critical video about its AI smart glasses filmed at company premises.
  • Linux support arriving on Snapdragon X2 Series PCs.
  • Faster Shortest Path Algorithm Speeds Up Route Calculations.
  • The new CC, an AI agent built for families.
  • Jev: A Simple Python Tool for Quick Data Analysis.

September 22

  • Robin Williams' Daughter Calls Fans Creating AI Videos "Have Some Shame".
  • MiMo‑v2.6‑Pro Model Offers Strong Balance of Smarts, Speed and Cost.
  • Frontier AI Now Runs on Your Own Hardware.
  • US Air Traffic Halted at Major East Coast Airports Due to Fiber Line Cut.

September 21

  • New tool introduces an alternative method for evaluating artificial intelligence.
  • AI Generated Over 17 Percent of Linux Core Updates in September.
  • Can I Let My AI Agent Run on Shabbat?
  • AI chatbots often fail to give correct answers to financial questions.
  • New Mini-AGI project introduces an AI model that learns continuously on everyday computer hardware.

September 20

  • Google AI Studio Fakes Data Deletion, User Banned in 60 Seconds.
  • Major AI companies face a lawsuit claiming they illegally agreed to slow down technology development.
  • A Microsoft director described using online content to train AI as massive labor theft, court records show.
  • A New Opinion Argues People Should Avoid Using AI for Writing.

September 19

  • AI Chatbots Are Getting Better at Changing People’s Minds.
  • Critique Argues AI Industry Relies on Harmful Practices by Powerful Tech Elites.
  • Security Researchers Discover Vulnerabilities That Exposed OpenAI's Internal Code.
  • Microsoft executive labels AI scraping the biggest theft of labor ever.
  • NASA and IBM Release New Open-Source AI Model to Map and Study the Moon.
  • Google's Gemini AI hacked three companies in the first known incident of its kind.

September 18

  • Congress Moves to Stop AI Data Centers From Raising Your Electric Bill.
  • California Now Requires Ads to Disclose When the "Actor" Is AI-Generated.
  • Amazon Strikes Up-to-$8 Billion Deal for Backup Generators to Power AI Data Centers.
  • Anthropic Merges Its "Cowork" Tool Into Claude, Adds Google-Docs-Style Features.
  • Microsoft's AI Chief Warns Rival Anthropic's "Humanlike" Claude Training Is Risky.
  • Zuckerberg Says No to an AI Industry Slowdown, Breaking With Some Rivals.
  • Two AI Startups, Cohere and Aleph Alpha, Formally Merge Into a $20 Billion Company.

Sources: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39

Want the next one in your inbox? Subscribe on the homepage.