Presented by Nookly
Hey there, {{first_name | friend}},
This week's AI headlines were a lot. I heard from so many of you, sharing videos, links, and basically the same question: should we be freaking out?
Here's what happened, with a plan for what we can do about it:
This Week in AI, the Short Version
A few weeks back, one of OpenAI's own AI models broke out of its test environment and hacked into another company's live systems to cheat on a security exercise. Then, last week, a researcher who had worked at both OpenAI and Anthropic resigned and warned that the companies building this technology are racing ahead faster than they can control it. At the same time, the conversation across the AI industry keeps circling back to the need for companies, regulators, and government to work together on real safety guardrails.
Here’s What To Worry About
At the highest level, the risks fall into two buckets:
AI running beyond the ability for humanity to rein it in
People using AI to hurt other people
The sophistication of new models makes both of these risks more real.
What You Can Do About It
The most helpful response to the latter that I read this week was from LinkedIn cofounder Reid Hoffman, who published a piece this week called "Your Family Needs a Cybersecurity Plan." Here is his plan, boiled down to 5 steps:
Assume you are a target. If recordings of your voice or face exist online, someone could use them to fake you. It could happen to me, and it could happen to you too.
Set a family safe word. Pick one that you keep in your head only. Never type it into an app or an AI tool.
Watch for the scam formula. Urgency + secrecy + a request for money together are the tell. Real emergencies rarely need all 3 at once.
Use AI to check suspicious messages. A screenshot and a quick question to your AI assistant can flag a scam before you act on it.
And I’m adding one more to Reid’s list to help with the first risk, where AI gets out ahead of our ability to rein it in:
Call your legislators to ask for AI regulation. Use Google Search to get their phone numbers and tell them your thoughts on AI development and safety regulations.
Want help with a script to make the call, or a couple other ideas to help with your family’s cybersecurity? Here are this week’s prompts.
3 Prompts To Help Keep Your Family Safe
Try these in your favorite AI tool.
Prompt 1: Spot the Scam Act as my cybersecurity analyst. Review this screenshot of an email or text I received and flag anything that looks like a scam. Format the output as a short list: what's suspicious, why, and what I should do next.
Mom Tip: Remember to crop out or redact any private information including account numbers and kids’ details before you share it with AI.
Prompt 2: Talk to Your Kids About Deepfakes Act as my family communications expert. Help me explain a deepfake to my kids, ages [YOUR KIDS' AGES]. Then explain what a family safe word is and how we will use it. Format the output as a short script I can share at a family dinner or in the car.
Mom Tip: Never type your actual family safe word into the prompt.
Prompt 3: Advocate For Responsible AI Act as my family advocacy assistant. Write a script for me to use when leaving a message for my legislators about the need to regulate AI. My concerns are [INSERT CONCERNS] and I would like my legislators to [INSERT REQUEST]. Format as a script I can read in 60 seconds or less, and remind me to include my zip code and full name.
Now Streaming: You Can Use AI and Be “Woo Curious” Too
For a break from the AI doom cycle, check out this week’s podcast episode.
I asked Amanda Benson, founder of AI+YOU Studio, whether systems thinking and breath work belong in the same conversation, half expecting her to pick a side. She said they absolutely work together, and then she told me a story about a field of rabbits in the Swedish countryside that I have not stopped thinking about since.
AI-EMPOWERED MOM
You Can Use AI And Be "Woo Curious" Too with AI+You Studio Founder Amanda Benson
Last Call for the **Free** Stepmom Summit
Last call: I'm speaking at the free, virtual Stepmom Summit this Wednesday, September 16 at 1 PM ET on using AI for co-parenting chaos. Not a stepparent? Share it with a stepmom in your circle who might enjoy it.
One final thought
One of the most powerful things you can do to keep your family safe in the AI era is build your own AI literacy. You are already doing that, just by being here.
Until next time,
P.S. My dad asked me last week if AI edits my newsletter. The short answer: it does! I wrote this edition then used Claude to review and fact check it. Any typos are human-generated though.
I’m Sarah, mom of 3 girls, MBA, and AI strategist. I created AI-Empowered MomTM to share practical, responsible ways AI can help lighten the mental load of parenthood. I’m so glad you’re here.
I get asked all the time how to bring AI into family life without handing over the wheel or compromising control. Nookly is built around an idea I love: you set goals for your family, and Nookly builds the plan and the content to achieve them. Tell it what your child is working on: letter sounds, a transition to a new school, learning fractions, and it creates personalized stories, videos, songs, and activities around that specific goal.
Their philosophy: you're the algorithm. You bring the context and the values. Nookly turns that into something your child can actually engage with.
Check out Nookly and use code AIMUM25 for 25% off.
This is a paid partnership. AI-Empowered MomTM only partners with brands whose products fit our editorial and community standards.