← All issues
Newsletter

AI Weekly Highlights – Games, Guides, and the Latest News — October 18, 2026

Sent October 19, 2026

Week of October 18, 2026

Hi there! Welcome to this week’s roundup of AI fun, tools, and updates.

Games: a new round every day

Game of the week · Picture game · 3 minutes
Picture Puzzle

Today's picture is mixed up. Slide the tiles back into place in as few moves as you can.

Play now

Learn to spot scams, fakes and AI

  • Scam or Not?: Five messages. Scam or legit? Find out in two minutes.
  • Real or AI?: Six pictures. A camera photo, or made by AI? Learn the giveaways as you go.
  • Spot the Fake Link: Eight web addresses. Safe, or a fake made to fool you? Learn the one trick that catches them.

Just for fun · quick picture games for all ages

  • Odd One Out: Find the one that is different, as fast as you can. Six rounds, and the grids get bigger.
  • Memory Match: Flip the cards and find all six matching pairs in as few moves as you can.
  • Spot the Difference: Two pictures, five differences. Find them all as fast as you can.

See all the games →

Free kits: step by step, done in minutes

Kit of the week · Free · 20 min
The Kids & Teens AI Kit

Age rules, parental controls and a family agreement for AI chatbots.

Open the kit

More free kits

Coming soon: the Household AI Toolkit, ready-made AI helpers for family life.

Browse all the kits →

AI Guides

Guide of the week
Prompting Guide
Read the guide

More guides

Read all the guides →

This week's AI news

October 18

  • New Tool Tests Programmers on AI-Generated Code to Prevent Blind Use: Sphinx quizzes developers on AI code before publishing.
  • Former Nvidia Executive Says AI’s Role in the Physical World Goes Beyond Robots: A former Nvidia veteran discusses physical AI beyond robotics.
  • AI Company Anthropic Reports a More Experienced Workforce and More Time in Meetings: Anthropic shifts to senior staff and spends more time in meetings.
  • Meta removes wording that sparked worries over data scraping and AI training: Meta updates policies to address data scraping concerns.

October 17

  • Turning Off AI in Your Google Account Will Cost You Other Features: Disabling AI tools also removes other Google services.
  • Actor Mark Wahlberg says he is open to an AI clone continuing his work in movies: Wahlberg considers a digital version of himself for future roles.
  • A new AI model called Jev has become the fastest-adopted tool in Vercel’s history: Jev reaches record adoption speed on Vercel.
  • Elorian introduces its first AI model designed to reason using images: Elorian’s model uses images directly for reasoning.
  • Why robotics hasn't had its ChatGPT moment yet, according to a former Nvidia insider: A VC explains why robotics lags behind ChatGPT.
  • Testing Meta's Muse AI Shopping Assistant Led to Higher Price Options: Meta’s Muse AI nudges users toward pricier choices.
  • India is the latest country to report issues with rogue AI agents: India reports unauthorized AI activity.
  • San Francisco Tech Week offers a look into how Silicon Valley feels about the rise of AI: NPR explores tech hub attitudes at the event.
  • Software company Replit has ended its partnership with AI maker Anthropic, its CEO announced: Replit ends collaboration with Anthropic.
  • CNET Reviews the Top AI Image Generators for Everyday Users: CNET compares leading image generators for everyday use.

October 16

  • Amazon’s Bee Unveils Two New Personal AI Devices Starting at $75: Amazon releases affordable AI assistants for home use.
  • AI agents are coming to Rivian’s cars: Rivian adds natural‑language AI agents to vehicles.
  • Bluesky updates its Attie app to help everyday users create websites and games with AI: Attie expands to broader audiences.
  • Anthropic Gives Outside Researchers Full Employee-Level Access to Study Its AI: Anthropic opens employee-level access to external researchers.
  • Amazon’s devices chief predicts future AI hardware could make phone apps obsolete: Amazon envisions AI hardware handling daily tasks.
  • Meta Releases New Free AI Model Designed to Run Directly on Mac Computers: Meta launches a local‑running AI model for Macs.
  • Europe’s Largest Airline Turns to AI to Stop Airport Travel Disruptions: Ryanair uses predictive AI to reduce travel chaos.
  • Robots Get Real‑Life Training: AI Firms Pay Indians to Record Everyday Chores: AI firms hire Indians to film chores for robot training.
  • Cal AI’s Viral Rise Fueled by Influencer Campaigns: Influencers boost Cal AI’s popularity.

October 15

  • Deezer Adds Option to Filter Out AI‑Generated Tracks: Deezer will let users turn off AI music.
  • AI firm Apate deploys undercover bots to fall victim to scams and collect fraud data: Apate uses AI agents to gather fraud intel.
  • YouTube Music Adds New Labels to Tackle Rising AI‑Generated Songs: YouTube Music launches labels for AI tracks.
  • Google’s Demis Hassabis Puts a Hopeful Spin on AI’s Environmental Impact: Hassabis says AI can reduce carbon emissions.
  • FBI Arrests OpenAI Employee Over Alleged Surveillance Bug: FBI detains an OpenAI employee over a hidden recorder.
  • Toilet paper and secret recorders: Saudi engineer arrested over alleged spying claims: A Saudi engineer linked to OpenAI is detained for hidden devices.
  • Gudea Uses AI to Spot and Stop Social Media Crises Before They Start: Gudea’s AI flags potential backlash early.
  • OpenAI Faces Lawsuit Over Use of Children’s Data in ChatGPT Training: A lawsuit claims OpenAI used minors’ data for training.
  • Anthropic Lures Investors as It Prepares for IPO: Anthropic seeks capital ahead of a planned public listing.
  • Ricursive Intelligence Bets on AI‑Designed Chips to Ease Computing Shortages: Ricursive develops AI‑designed chips to boost processing power.

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 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70

Hand-picked AI products this 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.

Follow Everyday AI Finds for AI tips and stories:

TikTok  ·  Instagram  ·  YouTube  ·  Facebook

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