In partnership with
Hey future-proof friends 💜
Question for you:
If AI disappeared tomorrow, could you still think for yourself?
Because of course I LOVE using AI. And of course I use it every single day.
But I've started noticing this creeping fear that it could definitely replace my ability to think critically and be creative if I’m not careful and let myself be lazy with it.
And I know I'm not the only one feeling it.
In fact, I get these trolly comments all the time on social media:
A selection of the most common get comments I get on social media
But joke’s on them because I've been experimenting with using Claude completely differently.
Not to just produce more stuff - because honestly, what's the point of producing more if it's all crappy and soulless?
But to actually make me SHARPER… I want it to argue with me… challenge my ideas… catch the blind spots I never thought of.
So in this newsletter I’m sharing 10 prompts you can use to do exactly that.
But before we dive in:
200+ AI Side Hustles to Start Right Now
AI isn't just changing business—it's creating entirely new income opportunities. The Hustle's guide features 200+ ways to make money with AI, from beginner-friendly gigs to advanced ventures. Each comes with realistic income projections and resource requirements. Join 1.5M professionals getting daily insights on emerging tech and business opportunities.
TL;DR - today's lineup:
One Serious Deep Dive 💡: How to use Claude (or AI in general) without getting dumber
Piping Hot AI Tea 🫖: Caveman token hack, 15-year-old founder, Claude Mythos leak
💌 Please share your feedback through the one-click feedback poll at the end 💜
One Serious Deep Dive 💡
Most people use AI to produce. These 10 prompts help you use it to think + produce.
The difference between these 2 approaches is everything.
One makes you faster, the other makes you faster + better:
Most people are using AI like a vending machine because it’s so damn good and so damn easy.
Topic in, final version out. Question in, final answer out.
Zero friction, zero challenge, zero thinking required.
If that's your whole relationship with AI, those “AI is Making us Dumber” articles are about you (I’m sorry, just giving you some tough love here 💜).
My Unbreakable Rules For Using AI
So I’ve set myself some intellectual guardrails for how I use AI - it’s a framework I’ve called Human-AI-Human:
I originate the idea (usually a voice note using WisprFlow).
OR sometimes AI originates the idea and I voice note in my own stories, POV and experiences.
But either way, I'm the first one in and the last one to touch it. I rip things apart, rewrite, rephrase and delete whatever doesn't sound like me.
10 Prompts to Turn Claude Into Your Thinking Partner
So here are 10 prompts you can use in Claude (my go-to, but they work just as well in ChatGPT or Gemini) to turn AI as my thinking partner.
None of them ask AI to write or create anything for you and that's the whole point).
1. The Stress-Tester
"Here's what I'm planning: [insert plan]. Act as a critical thinker. Question my assumptions, logic, or blind spots, but don't rewrite anything. I want to stress-test my thinking, not get new ideas."2. The Reframer
"Here's the core idea I'm working with: [insert idea]. Help me reframe it through a different lens, like a new audience POV, emotional trigger, or brand positioning angle."3. The Real Question Finder
"Here's the situation I'm thinking through: [insert challenge]. Help me uncover the real strategic question underneath this. What should I actually be asking myself?"4. The Gut Check
"Something about this feels off, but I can't explain why: [describe it]. Help me put words to the tension I'm sensing. What might be misaligned?"5. The Consequence Mapper
"Here's the decision I'm considering: [insert]. Help me think through the second and third-order consequences. What might happen after the obvious outcomes?"6. The Hidden Variable Hunter
"Here's the plan I'm analysing: [insert]. What critical factor might I be overlooking? The hidden variable that could completely change the outcome?"7. The Principle Extractor
"Here's something that worked (or failed): [describe it]. Help me extract the underlying principle so I can apply it elsewhere."8. The Intuition Decoder
"Here's my idea, and it feels right to me: [insert]. Help me articulate why this might make sense, even if I can't fully explain it yet."9. The Pattern Spotter
"Here are a few examples I've noticed: [describe them]. Help me identify the hidden pattern or principle connecting them."10. The Reverse Engineer
"Here's my goal: [insert]. Instead of moving forward, walk me backward from the desired result. What would need to be true at each step for this to succeed?"Want Claude to challenge you on EVERYTHING you ask? Add this one line.
If the prompts above are the occasional sparring session, this is like hiring a full-time debating partner.
Go to your Claude profile, open General Settings, and find Personal Preferences. Paste this in:
I like to be challenged in my thinking. Never just agree with me or tell me what you think I want to hear. I need you to be my thought sparring partner so that I can do my best thinking and my best work, but also don't do this unnecessarily.Here’s what mine looks like:
Sneak peek inside my Claude Personal Preferences
Fair warning: this can potentially annoy you sometimes.
It will question things you thought were a given and poke holes in ideas you were excited about.
But that's the point.
Because the default behaviour of AI is to agree with you. And agreeing with everything you say is exactly how AI makes people dumber. It just makes your bad ideas sound more polished!
I've had this in my settings for months now and it's one of the single best things I've done.
If you don't love it, just delete it. But maybe just try it for a week first.
And finally, next time someone tells you AI is making people dumber, ask them what they're actually asking it to do.
Want to see how I use AI to 10x my content and audience growth?
I’ve created a done-for-you content ideation, creation, management and publishing system to help you leverage AI to 10x your content creation (without losing your authenticity in the process).
It’s called the AI Content Multiplier and the waitlist is now open - join now and I’ll send you a 30% off early bird discount as soon as it launches (which is soon).
Piping Hot AI Tea 🫖
1. The 75% Token Hack (Caveman Mode)
A Reddit user found that forcing Claude to respond in broken "caveman-style" language (short sentences, no filler, "Me fix code" instead of polished paragraphs) slashed token usage by up to 75%. If you're running high-volume AI workflows, your biggest cost isn't the task. It's the fluff.
2. The 15-Year-Old Founder Hiring a 38-Year-Old
Thomas Guthrie, founder of AI platform Runwise, just went viral for hiring his first employee who's more than twice his age. The traditional career ladder is officially broken. Skills and vision over years of experience.
3. The "Claude Mythos" Leak
Anthropic accidentally exposed details of its next-gen model due to a CMS configuration error. Nearly 3,000 internal files were briefly visible, suggesting Mythos is a major leap in coding, reasoning, and cybersecurity. They've acknowledged the slip-up. The speed of development right now is wild.
If you enjoyed today's newsletter AND got to the end of it, I’d love a quick click on the poll below to let me know what you think 💜.
See you next week,
Jess xx
