Chapter 09 Companion

Staying Safe
and Smart.

Chapter 9 is about protecting yourself and your family in the age of AI, not fear-mongering, but eyes-wide-open awareness and practical steps you can take today. Meet Frank Medina, whose 82-year-old mother nearly lost $5,000 to an AI voice-cloning scam.

Frank Medina, retired postal carrier from suburban Phoenix

The GUARD Framework

G

Get the Facts

Stop before you react. Urgent or emotional messages are built to make you act before you think. Pause and take a breath.

U

Understand the Source

Verify independently. Check multiple sources. Call the actual company using a number you looked up yourself. Ask a trusted friend.

A

Assess the Risk

Look for red flags: weird phrasing, poor grammar, unusual payment methods, pressure to act fast, requests for secrecy.

R

Research Independently

Take control of your privacy and security. Check your privacy settings, limit what you share, use strong passwords, and enable two-factor authentication.

D

Defend & Educate Others

Share what you learn. Tell your family, friends, and neighbors so scammers have a harder time succeeding.

Key Data Retention Policies

ChatGPT

30 days

Default setting. Conversations retained for 30 days unless you opt out.

Claude

30 days

Anthropic retains conversation data for 30 days. Check account settings for opt-out options.

Gemini

18 months

Google retains longer than competitors. Data used for product improvement. Settings available to limit.

Safety & Verification Prompts

Understand a Privacy Policy

Privacy

I'm going to paste a privacy policy from [company name]. Please explain it in plain English. Specifically: 1) What data do they collect? 2) How long do they keep it? 3) Do they share it with anyone? 4) Can I opt out? 5) What happens if I delete my account?

Verify a Claim

Fact-Check

I read: "[paste the claim]". Before I trust this: 1) What would I need to see to verify this is true? 2) What are the most reliable sources I could check? 3) Any red flags in how it's presented? 4) What's the most likely alternative if this is false?

Analyze a Suspicious Message

Scam Detection

I received this message and I'm not sure if it's real or a scam: [paste message]. Please check for: 1) Urgency tactics? 2) Unusual payment methods? 3) Requests for secrecy? 4) Grammar or phrasing issues? 5) Can the claims be verified? Don't sugarcoat it.

Evaluate a News Article

Fact-Check

Help me evaluate this article: [paste text]. 1) Does it cite specific verifiable sources? 2) Any emotional manipulation? 3) What's missing? 4) How could I verify the key claims? 5) On a scale 1-10, how trustworthy is it and why?

Check AI's Own Work

Fact-Check

After you answer, rate your confidence (high/medium/low), tell me which parts you're most and least sure about, and suggest 2-3 sources to verify independently.

Create a Family Safety Plan

Safety

Help me create a family safety plan. My family includes: [describe]. I want it to cover: 1) Code word system for verifying emergency calls 2) Social media sharing rules 3) Steps if someone suspects a scam 4) Age-appropriate AI safety conversations 5) Schedule for reviewing privacy settings together

Check for Bias in AI Output

Bias Detection

I just asked an AI to help with [describe task]. Help me think about potential biases: 1) What assumptions might the AI be making? 2) What perspectives might be underrepresented? 3) How could I cross-reference this for fairness? 4) What additional info should I seek from non-AI sources?

Social Media Privacy Audit

Privacy

I want to reduce my digital footprint on: [list platforms]. For each, give me: 1) Specific steps to make my profile private 2) How to limit who sees my posts 3) How to find and remove old posts 4) Settings that control AI training use 5) How to check for data breaches. Keep instructions simple.

Key Takeaway

AI safety is not about avoiding AI, it's about using it responsibly. The GUARD framework works for everything from evaluating news articles to spotting scams. And remember: these tools that help you stay safe can also hallucinate about safety topics. Always verify AI-generated safety advice independently.

Protect Yourself and Your Family.

Get weekly safety tips, updated scam alerts, and family-friendly resources. Stay informed without the fear.

Last verified: September 1, 2026