Sunday Rundown #141: [AI] for [Blank] & 5D Chess
Plus my Cozora session with hands-on image-to-video tricks.
Happy Sunday, friends!
Welcome back to the weekly AI news roundup.
In case you missed it, here’s this week’s Thursday deep dive:
If you’re consistently missing out on my emails, remember to check your “Promotions” tab and mark whytryai@substack.com as a “Safe Sender.”
🗞️ AI news
Here’s what happened in AI this week:
👩💻 AI releases
Amazon launched Alexa for Shopping that can research products, compare prices, track deals, and automate routine purchases.
Anthropic news:
Agent View lets you manage all your Claude Code sessions from a single dashboard to see which ones are running, need input, or are completed.
Claude for Legal comes with 20+ MCP connectors and 12 plugins tailored to specific legal work and practice areas.
Claude for Small Business includes 15 ready-to-run workflows for finance, HR, operations, and sales with connections to Canva, PayPal, and more.
Krea launched Krea 2 image model with style transfer that lets you match any aesthetic from reference images and moodboards.
Meta news:
Incognito Chat lets you have fully private AI conversations on WhatsApp with messages that don’t get saved or seen by anyone, including Meta.
Muse Spark went from research to general availability and is now accessible across all of Meta’s apps (Facebook, Instagram, Messenger, and WhatsApp).
Microsoft added new Copilot features to Edge on desktop and mobile, including long-term memory, multi-tab reasoning, and study tools like flashcards.
OpenAI news:
Codex is rolling out to the ChatGPT app on iOS and Android, letting you kick off coding tasks and other long-running projects from your phone.
Personal Finance connects ChatGPT to 12,000+ financial institutions, so you can ask questions about your spending and get live dashboards in the app.
🔬 AI research
Google news:
Gemini Intelligence is coming to Android this summer with Gemini built into Chrome, multi-step task automation, and intelligent speech-to-text.
Gemini Spark is an always-on AI agent that runs in the background to handle your daily tasks (expected to be announced at Google I/O).
Magic Pointer will turn your mouse cursor into a context-aware mini AI agent that intuitively understands your requests based on where you’re pointing.
Thinking Machines Lab previewed interaction models that can respond, interrupt, and translate in real time across audio, video, and text.
xAI announced Grok Build, a coding agent that can write files, run commands, and build apps from natural language. (Early beta for SuperGrok Heavy users.)
📖 AI resources
“AI IQ” [BENCHMARK]: a new platform that scores frontier AI models on an IQ curve, with cost comparisons, intelligence trends, and live rankings.
🤦♂️ AI fail of the week
This clip of me playing chess against a robot is practically flawless, but an eagle-eyed Grandmaster might be able to spot a few subtle inconsistencies.
📹 My Cozora image-to-video workshop
On May 7, I had the pleasure of joining Michael Simmons of Cozora for a one-and-a-half-hour practical session on image-to-video models.
I shared hands-on examples of using images with AI video, showcased different techniques, and discussed their practical applications.
It was a lot of fun and jam-packed with actionable tips. I even threw in a take-home reference sheet with key takeaways.
Paid subscribers can watch the full session here:
If you’re not already a member, note that all of my subscribers are eligible for discounts on Cozora memberships:
Free subscribers get 10% off
Paid subscribers get 50% off
In related news: I expect to get back to my Live & Learn sessions the week after next, so if you have input on future topics, format, and so on, please share it here: