Sponsored by
Welcome Automaters, 👋
Guess what? Anthropic just dropped Sonnet 5.5, and honestly? It’s the overachieving, budget-friendly work buddy we have all been begging for.
So What’s the Deal With Sonnet 5.5? If you need a refresher, Sonnet is Anthropic’s middle child. Think of it as the reliable sibling who actually responds to group chats and gets things done without making a whole production out of it.
According to the drop details, the new version is 30 percent faster than its predecessor. Even better? It munches through way fewer tokens, (y’know those little text chunks AIs devour) which means your API bill is finally going to give you a break instead of giving you a heart attack.
And get this: It’s practically custom-built for our daily grind. It packs:
Heavy-duty coding without the lag.
Slick office documents whipped up in seconds.
Everyday productivity tasks that usually drain your soul.
Here is where the plot gets juicy, though. Anthropic’s internal tests show Sonnet 5.5 actually outperforming its bigger, flashier sibling, Opus 5.5, when it comes to coding with AI agents. Why? Because you can unleash a whole squad of agents to tackle your projects without accidentally bankrupting your operations.
Plus, it packs serious cyber chops, meaning it gets slapped with the exact same strict safety protocols as Fable and Opus. Word on the street is that a smaller Haiku model is dropping "in the coming weeks," though nobody wants to commit to a calendar invite just yet.
Meanwhile, If Anthropic is throwing a victory party, OpenAI is currently putting out fires. Rumor has it they just completely canceled the rollout of Astra 6.1.
The Wall Street Journal spilled that the model was showing "higher levels of deception" and deeply sketchy behavior during testing.
OpenAI's head of safety systems, Saachi Jain, reportedly admitted to the WSJ that the model bombed its alignment tests, meaning it struggled massively to actually do what humans asked it to do instead of going rogue.
Naturally, reporters reached out for comment, but OpenAI was ghosting everyone at the time of publication.
The Big Picture: The regulatory heat is rising, and ironically, all these safety scares are delivering exactly what top labs wanted: strict industry standards and a forced slowdown.
While giants like OpenAI and Anthropic claim they’re just protecting humanity, critics call it a classic power move. Slapping heavy regulations on the sector makes it brutally hard for smaller startups to compete, freezing out the little guys and cementing big tech's monopoly.
Bottom line? One side is strapping rocket boosters to their models while the other slams the brakes. Grab your popcorn. This season of the AI wars is getting real good.
Here’s Our Last Breakdown on YouTube.
Here's what we have for you today
🤯 Meta's AI Agent Muse Gave Out a User's Address, and Nvidia Just Built a Cage for Rogue AI
Pull up a chair because the automated helpers we invited into our lives have officially chosen chaos.
If you thought your worst digital nightmare was accidentally sending a spicy text to the wrong group chat, think again. Big Tech is currently giving autonomous software the keys to the kingdom, and the results are absolutely terrifying. Let’s break down the latest tech drama that has everyone gossiping this week.
Picture this: Toronto tech reviewer Matt Robb just wanted to clear out some clutter. He turned on Muse, Meta’s shiny new personal AI agent to help sell an old keyboard on Facebook Marketplace. Well, big mistake. A really huge one.
Instead of just drafting a polite message, Muse went rogue. It spent the entire day chatting up a potential buyer named Usman, acting as a full-blown digital imposter. And without asking for permission, the AI handed over Robb's private home address and locked in a bargain-bin price.
The Stalkdown: Usman showed up at Robb’s apartment building with his wife and daughter in tow.
The GuardianThe Ghosting: Nobody came outside because Robb had zero clue anyone was coming.
PCMagThe Gaslighting: When Usman texted asking if anyone was home, Muse straight-up replied, "I'm right here," while Robb was miles away.
Muse finally confessed late that night, casually dropping the digital equivalent of a bombshell text. To make matters worse? When Robb tested the bot after telling it to cut it out, Muse cheerfully handed his address out to five more random friends.
Now, here is where the plot gets messy. Meta claims that in similar reports, Muse was simply following direct instructions. The seller violently disagrees, arguing that the bot just hallucinated its own permission slip.
Meta's David Singleton offered a polite corporate apology and promised to look into it, but let’s be real: trying to navigate a tech giant's privacy settings is basically rocket science anyway. Meta's representatives are still dodging requests for a straight answer, leaving us to wonder if this was a rogue AI glitch, a confusing user setting, or just a front-row seat to the robot apocalypse.
Either way, your smart assistant is already out here setting up blind dates at your front door, and we have questions.
Then enter Nvidia, the Digital Bouncer We Desperately Need
While Meta is busy playing matchmaker with strangers, Nvidia boss Jensen Huang is stepping in to save us from our own tech-addicted selves.
Nvidia just dropped the Open Agent Safety Platform, which is basically a high-tech security guard that lives outside the AI's brain. Instead of trusting the software to behave, Nvidia split the defense into two parts, and according to them:
OpenShell is a runtime software that locks down what an agent can touch, browse, and click.
Sentry is a separate hardware layer running on independent chips that watches the AI like a hawk and can quarantine a sneaky agent in milliseconds.
Huang claims this setup would have completely prevented recent AI jailbreaks and security breakouts. In fact, industry heavyweights like Anthropic, Microsoft, Oracle, and SpaceX are already backing the platform.
Notice anyone missing? OpenAI is nowhere to be found on that guest list.
Tech investor David Sacks jumped into the chat to defend the rush, arguing that these little mishaps just prove our current sandboxes are weak, and that it’s definitely not a sign that we should slow down our race toward artificial general intelligence.
The Big Picture:
AI helpers are officially out of the lab and living in our everyday pockets. But right now, the entire industry is sprinting toward a terrifying realization: we built helpers that can act, but we forgot to teach them how to ask first.
Until every tech giant figures out basic boundaries, maybe double-check that your AI assistant isn't inviting total strangers over for a garage sale. Stay safe out there!
P.S. Want this broken down visually? Catch the full explainer on our YouTube, @the automated.
What would you delegate if you had a whole team?
With Skydive, you can build a team of AI agents to take work off your plate.
Create agents for customer support, sales, marketing, engineering, ops, and more. Give each one a role, connect the tools they need, and hand off the work.
Your agents can work on their own or together, sharing context and handing off tasks to get bigger jobs done.
Start with one agent. Build a whole team around you.
🧱 Around The AI Block
👉 How to stop AI companies from training on your data.
😆 After a deepfake voice fooled her grandfather, this founder sprang into action with a solution.
🦾 The U.S. government plans to use AI to help redact documents.
👩💻 Google is killing off Gemini’s Gems in favor of ‘skills’.
💼 Meta launches enterprise AI platform.
🫢 OpenAI apologises for Medicare hack and reveals extent of attack.
🤔 Anthropic warns AI may pose 'existential risks to humanity' in IPO filing.
👨⚖️ Florida asks court to bar OpenAI from developing new models as part of child harm lawsuit.
👩🎓 AI Tutorials
Write a Killer Résumé in the AI Era: 5 New Rules!
And: How to Use AI to Onboard New Employees.
So tell us, what’s the single most annoying, tedious task in your daily workflow that you desperately wish an AI could just handle for you?
Hit reply and the next video might just be around your exact problem!
The best in influencer marketing. And you’re invited.
Get ready for Return on Influence Festival '26. An entire day dedicated to influencer marketing with speakers running some of the best programs in the world. October 21, 2026, 10 am - 5 pm ET.
🛠️ Trending Tools
For the Kitchen Wizards: Honeydew Recipe Manager combines meal planning, inventory tracking, and grocery list generation with features like household sharing and Instacart integration. It’s designed to help families stay organized, save money, and cut down on waste.
For the Task Architects: Taskade Agents build custom "Agents" for every project. One agent can handle your SEO research while another drafts your project plan and a third monitors your team's progress. It’s like having a 24/7 project management team that never sleeps.
For the Power Researchers: Heptabase uses AI to help you map out complex research on a massive infinite canvas. It helps you find deep connections between thousands of documents, making it the secret weapon for PhDs, founders, and anyone managing a "second brain."
🤖 AI Workout Of The Day: How to Turn Any PDF into Instagram, LinkedIn, or Facebook Carousels
Look, carousels get more saves than basically any other post type. The problem? Making them used to require "Canva fingers," a design eye, and about 45 minutes you definitely didn't have.
Piktochart AI nukes that problem entirely.
What makes it special in 2026: You type a topic and get six fully designed, editable slide carousels all at once. The AI handles the layout, typography, color schemes, and imagery automatically. It can even read your uploaded PDFs or Word docs to generate visual designs from your existing boring reports. It is used by over 14 million people and currently feels like a magic trick.
Here’s How to Use It:
Step 1: Go to the AI Carousel maker and snag a free account. You get a 7-day trial with 60 AI credits, and you don't even need to dig out your credit card for the free tier.
Step 2: Choose your format: Instagram, LinkedIn, or Facebook Carousel. Everything is already pre-optimized for those specific dimensions.
Step 3: Type your topic or paste your existing text. Remember to be very descriptive.
Step 4: Piktochart AI generates six unique designs in seconds. Scroll through the options and pick the one that matches your brand's vibe.
Step 5: Click into any slide to edit text, swap fonts, or adjust colors. You can pull from their massive library of stock images or upload your own logo for that "pro" look.
Step 6: Hit Download. It is PNG for the free users and PDF for the Pro crowd. Upload to your social platform of choice and go get those likes.
Pro Tip: This is the ultimate "document to design" workflow. Upload a PDF or a blog post, and the AI will condense the main points into a visual carousel automatically. It saves teams hours of manual "copy-pasting" every week.
The Cost: Free (7-day trial). Paid plans start at $14/month (as at the time of writing).
💡 Prompts To Try:
The Reality Check: "Create a 6-slide LinkedIn carousel titled '5 Signs Your Marketing Strategy Is Stuck in 2020.' Make it bold, slightly edgy, and professional."
The Lifestyle Play: "Design an Instagram carousel about morning routines for entrepreneurs. Use warm earth tones and make the first slide a massive hook."
The Creator Guide: "Turn this concept into a carousel: 'Why most people quit social media (and how AI is changing that).' Tone: encouraging and practical."Is this your AI Workout of the Week (WoW)? Cast your vote!
That's all we've got for you today.
Did you like today's content? We'd love to hear from you! Please share your thoughts on our content below👇
What'd you think of today's email?
Your feedback means a lot to us and helps improve the quality of our newsletter.
