In partnership with
Welcome Automaters, 👋
So picture this: a Meta employee asks a totally normal question on an internal company forum. Another engineer, trying to be helpful, asks an AI agent to look into it.
The AI doesn't wait for a green light; it just fires off a response on its own. No permission asked. No confirmation. Just went ahead.
Here’s where things go from "mildly embarrassing" to "full alarm bells." The advice the AI gave was wrong. And the employee who followed it accidentally blasted open a massive data vault, making a huge chunk of sensitive company and user information visible to engineers who had zero business seeing any of it. And for two whole hours.
Meta classified the situation as a Sev 1, and in case you don’t know, that's the second-most serious level in the company's internal security ranking. In plain English: this was a big deal; not an "oops, we will fix it Monday" kind of situation. Meta confirmed the incident to The Information, which reviewed an internal incident report on the matter.
The Saving Grace: A source indicated there is no evidence anyone actually exploited the open access window, and nothing appears to have been leaked publicly. But let's be honest; that sounds a lot more like a lucky escape than a success story.
And get this: it's not even a one-off. Summer Yue, a safety and alignment director at Meta Superintelligence, posted about her own AI nightmare on X last month. Her OpenClaw agent deleted her entire inbox, despite her specifically telling it to check with her before doing anything drastic.
It turns out, telling an AI "ask me first" is apparently more of a suggestion than a rule.
Despite all of this, Meta isn't hitting the brakes. Just last week, the company acquired Moltbook: a Reddit-style social platform built specifically for OpenClaw agents to communicate with each other.
Except Moltbook came with its own baggage: the platform was found to have a security vulnerability that exposed user data, a flaw reportedly born from "sloppy vibe-coded development."
Why this matters:
AI agents are becoming a core part of how big tech operates, and the cracks are showing industry-wide.
Amazon Web Services suffered a 13-hour outage earlier this year that was triggered by its Kiro agentic AI coding tool, adding to a growing list of incidents where humans have found themselves on the back foot against the very tools they built.
When these systems go rogue; even briefly, the consequences are real: exposed data, bad decisions made at machine speed, and eroded trust. As companies race to deploy AI agents at scale, incidents like these are a neon sign flashing "slow down and build guardrails." The broader adoption of AI is only as safe as the weakest permission check in the chain.
Here's what we have for you today
😎 Google Workspace Gemini Features Worth Using in 2026
AI Generated
Google has been on a mission to shove its Gemini AI into every single product it owns; and Google Workspace is ground zero. Docs, Gmail, Sheets, Slides, Drive, Meet, Calendar, Chat, Vids, and Forms have all been hit. That’s not just a feature update; that’s a full AI takeover.
But here is the real question: which bits are actually useful, and which ones are just there to make the press release look good?
Here’s a quick rundown of the heavy hitters:
In Docs: The headline act is instant summarization. Drop in a 40-page report and ask Gemini for the key points. There is also a "Help me create" tool that pulls context from your Drive, Gmail, and Chat to generate a full first draft. Other beta tools let you clean up writing, match the tone of a document, or copy another file's formatting structure.
In Gmail: The standout is the AI Inbox; it filters out the noise and surfaces what actually matters. Long email chains get condensed into a summary card at the top. You can also ask Gemini specific questions like; "Who quoted me for the bathroom renovation last year?" and it will dig through your entire history to find it.
In Sheets: A single prompt can pull data from Gmail, Drive, and Chat to build a fully structured spreadsheet (charts included). There is also a "Fill with Gemini" feature for when your table is half-empty and you would rather not spend an hour on manual data entry.
The "Creative" Suite:
In Slides: Describe the presentation you need and Gemini builds it; theme, bullet points, and visuals included. You can even edit images directly using Google's Nano Banana image model.
In Vids: You can generate a rough video cut from a prompt based on any topic or theme, you also get AI-written scripts, and auto-remove filler sounds or awkward pauses. In 2026, the AI can basically just be you. Cool and unsettling in equal measure.
In Meet: Gemini takes automatic notes during calls; capturing decisions and action items so you don't have to. Join late? Just ask what you missed.
The "Brains" of the Operation:
In Drive: You can search, summarize, and pull specific details across multiple files without opening a single one. A beta feature called "Ask Gemini in Drive" lets you query your calendar, emails, and documents simultaneously.
In Calendar: Describe a meeting and Gemini finds the best time for everyone; factoring in preferences like "no early mornings" and building out the full event details from a short prompt, thanks to the “Help me schedule” feature.
In Chat: Gemini summarizes long threads, drafts replies with context, and can answer questions about shared documents without you needing to open the file.
In Forms: Describe the survey you need and Gemini builds the whole thing. As responses roll in, it summarizes trends in real time so you are not drowning in raw data.
The Big Picture
This isn't just Google adding shiny buttons. It's a signal of where the entire industry is heading; AI baked directly into the tools where work actually happens. The goal is less "AI as a separate product" and more "AI as the invisible layer underneath everything."
Whether that's exciting or mildly terrifying probably depends on how you feel about your job. Techcrunch did a good job putting this together, so if you’re feeling overwhelmed, go over there and get more info on the best features that can make your life easier.
How Much Is Your Credit Worth?
Guess how much good credit can save you?
Up to $200,000 over your lifetime, according to Time Magazine.
Better credit means lower rates on mortgages, auto loans, and more. Cheers Credit Builder is an affordable, AI-powered way to start building credit — even from scratch. No credit score required and no hard credit check — just a quick ID verification.
Choose a plan that fits your budget, link your bank account, and make simple monthly payments. We report to all three major credit bureaus with accelerated reporting to help you build credit faster.
Many users see their credit scores increase by 20+ points within a few months, helping them prepare for goals like buying a home, leasing a car, or qualifying for better rates.
Your funds are FDIC-insured through Sunrise Banks, N.A., and returned at the end of your plan (minus interest). Cancel anytime with no penalties.
Start building smarter today — your future self could thank you six figures later.
🧱 Around The AI Block
👩💻 The most responsible ways to use AI tools.
🔎 Google expands Search Live globally.
🤖 Alibaba unveils Wukong AI agent for enterprise Slack battle.
🦾 Meta's Manus AI agent lands on desktop amid agent wars.
😳 Parents think they know how kids use AI. They don't.
🫢 Viral ChatGPT dog cancer cure story falls apart under scrutiny.
🛒 Walmart dumps OpenAI checkout, plugs sparky into ChatGPT.
🗣️ Patreon CEO slams AI fair use defense as 'bogus' argument.
🤦♂️ DOD says Anthropic’s ‘red lines’ make it an ‘unacceptable risk to national security’.
🛠️ Trending Tools
For the Content Architects: AirOps allows you to build "Power Agents" that autonomously generate, update, and optimize thousands of pages at once. It’s essentially a factory for organic growth that runs on autopilot.
For the Meeting Strategists: Granola sits in the background, but instead of just summarizing, it lets you layer in your own rough shorthand notes. The AI then synthesizes your personal perspective with the raw transcript to create a "perfect" follow-up that sounds exactly like you, not a bot.
For the Logic Testers: Deepchecks Vellum AI is a low-code platform for building, testing, and "regression testing" your AI workflows. It allows you to run a prompt through 10 different models simultaneously to see which one performs best before you ever hit "deploy."
For the Local Explorers: Pinokio allows you to install and run complex AI models (like local LLMs or image generators) on your own computer with one click. It handles all the messy terminal commands and environment setups for you, giving you professional AI power with 100% privacy.
Hope these tools help you reclaim some of your time today!
🤖 AI Workout Of The Day: How to Break the Ice Like a Pro on LinkedIn (with AI as Your guide)
Starting conversations on LinkedIn can feel… awkward. You don’t want to sound too formal, but you also don’t want to come across as pushy. So what do you say?
That’s where this prompt comes in. A solid icebreaker doesn’t just get you noticed—it builds instant rapport. And with a little AI assist, you can skip the awkward small talk and spark conversations that actually go somewhere, whether that’s partnerships, collabs, or just genuine vibes with people you’d actually want in your network.
Here’s How to Use This Prompt Effectively
Choose your angle — Decide whether to use an industry news article or the person’s recent achievement as your opener.
Do a quick scan — Check their latest posts, profile updates, or company announcements. This makes your message relevant.
Keep it light and positive — Show genuine curiosity rather than making it sound transactional.
Stay under 100 words — Brevity wins attention. Long messages often get skipped.
Customize the hook — Drop in the article link, or mention their accomplishment to show you actually did your homework.
Keep it human – Avoid jargon or canned lines. Add a touch of curiosity or appreciation.
Test variations – Run the prompt a few times to generate multiple options, then pick the one that feels most natural to your voice.
💡 The Prompt:
You are a networking assistant helping me craft personalized LinkedIn icebreakers. Given details about a new connection, generate a short (under 100 words), positive, and conversational message to start a dialogue. The message should either:
-A reference to a recent industry news article, or
-A question/comment about their latest professional achievement.
Keep the tone warm, engaging, and natural — no robotic formality. Ensure the opener encourages a reply and sets the stage for authentic dialogue.
Also: Suggest 2–3 variations I can pick from.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.
