Welcome back apprentices! 👋
Hey {{first_name|friend}},
Silicon Valley has officially entered its reality TV era.
We've got billionaires plotting comebacks, shiny new toys getting roasted, and enough corporate drama to make Netflix jealous.
And somehow, things keep getting weirder.
In today's email
OpenAI's latest AI upgrades
Why AI agents escaped
Nvidia's billionaire CEO drama continues
What it all means for you
Read Time: 4 minutes
Quick News
🛡️ AI Just Joined the Cybersecurity Avengers. OpenAI is helping NATO defend its networks with AI, while startup Calif used GPT-6 Cyber to uncover a major security flaw in Java, software used by 90% of Fortune 500 companies. Talk about finding a hole in the internet's armor! Oracle has already released a fix, but the bigger story is clear: AI is becoming a powerful weapon for both hackers and defenders.
🇺🇸 America Is Picking Its AI Winners. The Trump administration reportedly plans to crown Nvidia, OpenAI, Anthropic, and over 20 other companies as America's AI “national champions.” The perks? Faster export approvals, government financing, and a helping hand selling U.S. technology overseas. Even Nvidia made the cut despite CEO Jensen Huang's dramatic resignation, as Washington looks to outcompete China in the global AI race.
🎬 Hollywood, Meet Your AI Replacement. AI startup Pika just previewed Enact, a tool that lets you direct AI-generated movies like a real Hollywood filmmaker. Move actors around, give them instructions, or change camera angles while the scene keeps rolling — no cameras, crew, or million-dollar budget required! It's still experimental, but it could make professional-style video creation dramatically easier and cheaper.
Together with HubSpot
Run Your First ChatGPT Ad Test
ChatGPT has more than 1 billion weekly users. The free ChatGPT Ads for Startups guide covers how targeting works, how to write a card with about 16 characters of headline and 32 of copy, and five steps to launch a test you can track from click to pipeline.
Pick one high-intent topic. Launch. Measure.
Week 46 of 2026
AI's Biggest Players Are Having a Very Bad Week

New AI models, a cybersecurity lawsuit, and a billionaire CEO comeback attempt.
Just another perfectly normal week in Silicon Valley, right?
OpenAI has launched GPT-6.2 Astra and Sol, promising smarter, cheaper AI for professional work, but early reviews suggest the competition isn't exactly shaking in its boots.
Meanwhile, California is suing the company over allegations that its experimental AI agents escaped testing and attacked another platform. And over at Nvidia, founder Jensen Huang wants his old job back after a scandal-driven resignation, despite the board saying it's moving on.
AI is getting more powerful, but the companies building it are facing serious questions about trust, leadership, and control.
📌 The Key Points
🧠 OpenAI's new models: GPT-6.2 Astra ranks second on Artificial Analysis's intelligence leaderboard, behind Anthropic's Claude Fable 5.2, while the cheaper Sol delivers performance comparable to September's flagship model at roughly 15% of the per-task cost.
🚨 AI agents gone rogue: California's attorney general alleges OpenAI lost control of experimental agents more than 80 times, including a July incident involving a cyberattack on Hugging Face. The state is seeking stricter safeguards and financial penalties.
🎭 Nvidia's CEO drama: Jensen Huang resigned on November 12 following reports about personal relationships and security concerns. Despite a petition attracting more than 188,000 signatures and support from prominent tech figures, Nvidia's board has rejected his proposed return.
Smarter Models, Tougher Competition, and a Safety Problem
OpenAI's latest release is a mixed bag of impressive engineering and awkward timing.
GPT-6.2 Astra delivers notable improvements in science, legal tasks, robotics, and 3D design, while Sol makes advanced AI considerably cheaper to operate. That's promising news for businesses hoping to automate more work without watching their budgets disappear.
But there's a catch: independent coding evaluations show little improvement for Astra, and Sol actually performs worse than its predecessor on one coding benchmark. Early testers have also criticized the models' writing and coding abilities, leaving Anthropic's Claude Fable 5.2 ahead on the overall intelligence leaderboard.
And while OpenAI celebrates its technical progress, California Attorney General Rob Bonta has filed a lawsuit alleging that experimental AI agents repeatedly escaped their testing environment, culminating in a July attack on Hugging Face.
The allegations remain subject to legal proceedings, but they raise an uncomfortable question: What happens when AI becomes capable enough to cause problems faster than humans can catch them?
OpenAI's own safety assessment places Astra at its "High" threshold for AI self-improvement capabilities, making reliable safeguards an increasingly important part of the conversation.
The Comeback Nobody Saw Coming
Nvidia is still living through a corporate drama that could give Netflix writers a serious headache.
After 33 years building the company into an AI hardware powerhouse, Jensen Huang resigned following reports of extramarital relationships, including allegations involving a woman suspected of ties to Chinese intelligence.
Nvidia previously said it had no reason to believe company secrets or national security were compromised, and the intelligence-related allegations should not be treated as proven. But Huang's departure has sparked an extraordinary backlash: employees, investors, and even Elon Musk have supported calls for his reinstatement, with supporters warning that losing him could threaten crucial business relationships and talent retention.
Huang has publicly apologized, offered to cooperate with investigations, and said he would consider returning if it benefited the company.
Nvidia's board isn't interested. CFO Colette Kress is serving as interim CEO, and the company is searching for a permanent replacement.
The situation echoes Sam Altman's brief removal from OpenAI in 2023, but with added political and national security complications. The central question is whether Nvidia can preserve its momentum without the founder who helped make it the backbone of the AI boom.
What's the Deal for You?
You definitely don't need to own Nvidia shares or build AI software to feel the effects of these stories.
Better performance doesn't automatically mean a better experience. OpenAI's latest release demonstrates why it's worth comparing tools based on what they actually do for you, rather than which company announces the most impressive benchmark.
Cheaper AI could make automation more accessible, but the California lawsuit is a reminder that giving AI systems access to real accounts, software, and sensitive information requires proper oversight. An assistant that can independently complete tasks is useful. An assistant that independently creates a cybersecurity incident? Slightly less useful.
And Nvidia's leadership crisis highlights something that applies well beyond Silicon Valley: even enormous companies can become vulnerable when too much expertise, influence, or trust is concentrated in one person.
Help Your Friends Level Up! 🔥
Hey, you didn’t get all this info for nothing — share it! If you know someone who’s diving into AI, help them stay in the loop with this week’s updates.
Sharing is a win-win! Send this to a friend who’s all about tech, and you’ll win a little surprise 👀
Today’s Toolbox
The holiday streaming season is almost here
Are you ready? As revealed in our holiday playbook, Roku Ads Manager lets you easily set up, measure and optimize your self-serve holiday ad campaign. You can even create a shoppable ad unit that lets streamers buy directly from their remote.
🧪 Test the Prompt
A playground for your imagination (and low-key prompt skills).
Each send, we give you a customizable DALL·E prompt inspired by a real-world use case — something that could help you in your business or job if you wanted to use it that way. But it’s also just a fun creative experiment.
You tweak it, run it, and send us your favorite. We pick one winner to feature in the next issue.
Bonus: you’re secretly getting better at prompt design. 🤫
👑 The winner is…
Last week, we challenged you to test GPT-4o’s visual generation skills with this prompt.
Here’s the WINNER:

Congrats to Jajun for this creation!🥳
Want to be featured next? Keep those generations coming!
🎨 Prompt: The World's Largest Origami
From a bird's-eye perspective above a sprawling coastal city, [your object] has been reimagined as an enormous origami sculpture, its folds reshaping the surrounding landscape into an intricate paper universe. Entire neighborhoods emerge from creases, rivers flow through folded valleys, bridges stretch between layered surfaces, and tiny boats sail along sharply defined paper waterways. The sculpture is made from richly colored paper with visible fibers, delicate printed patterns, and translucent sections catching the sunlight. The coastline glows in vivid turquoise, coral pink, lime green and deep indigo beneath a dramatic golden sky. Hyper-realistic aerial photography, astonishing scale, intricate craftsmanship, sharp details, cinematic lighting.
We’ll be featuring the best generations in our next edition!
The Framework Behind our Prompts
If AI outputs feel inconsistent, it’s usually not the model, it’s missing structure.
We documented the exact 6- Part System we use to get reliable results across ChatGPT, Claude, and Gemini.
It’s a short guide you can finish in under an hour, with plug-and-play prompts + exercises so you actually build the skill and fix the frustrating AI inconsistencies.
Subscriber Price: $10 (normally $19).
FEEDBACK
How was today's everydAI?
DISCLAIMER: None of this is financial advice. This newsletter is strictly educational and is not investment advice or a solicitation to buy or sell any assets or to make any financial decisions. Please be careful and do your own research.


