Most people are using AI wrong.
They collect tools.
Save prompts.
Watch tutorials.
Test agents.
Buy courses.
And six months later?
No business.
No audience.
No system.
No leverage.
Here’s the uncomfortable truth:
AI is not a shortcut.
AI is not a business model.
AI is not a magic employee.
AI is leverage.
But leverage only works when attached to:
1. Clear thinking
2. Real workflows
3. Business context
4. Distribution
5. Taste
6. Judgment
Most people use AI like a toy.
A few use it like a team.
This account is for the second group.
I write about AI without the grift:
Systems > tools. Leverage > prompts. Execution > hype.
If you’re tired of fake AI gurus, “100 tools” lists, and prompt bro nonsense — you’re in the right place.
Stop overthinking the perfect words.
Start obsessing over how you deliver them.
The game isn’t writing better hooks.
It’s performing them better.
Save this thread.
Which hook style resonates with you the most? Drop it below 👇
One sentence.
Same words.
Millions of views vs almost zero.
This one TikTok broke my brain about hooks.
Most creators obsess over WHAT to say.
The real game is HOW you say it.
Thread 🧵
Pro tip:
Study the top 1% in your niche.
Steal their delivery style, not just the words.
Then add your own personality.
That’s how you create hooks that feel fresh but still work.
Practical takeaways for you:
• Record the same hook 5 different ways (calm, excited, serious, storytelling, urgent)
• Test which one gets the highest watch time
• Double down on the winner
AI can help you brainstorm 10 variations in seconds.
The creator in the video nailed it:
Words are only 50% of the hook.
The other 50% = Tone + Pace + Emphasis.
This is why some “basic” hooks go viral while “clever” ones die.
Money hook:
“Is it possible to make $100,000 in 90 days?”
One version feels like hype.
Another feels like a real conversation.
The second one hits harder.
Your delivery decides if people stop scrolling or keep going.
Another example:
“If I woke up 20lbs overweight tomorrow…”
One guy delivers it with calm confidence → 790K views.
Another with raw urgency → only 1K.
Same structure.
Different emotional punch.
Watch this.
Two creators say “Sell me this pen.”
One gets 4.8 MILLION views.
The other — 94K.
Same hook.
Completely different delivery.
The difference? Energy, tone, pace, and what they emphasize.
@Polymarket Nothing surprising there. Industrial espionage has always been a thing. But this could be a serious blow to OpenAI. Anthropic is probably the one who’ll benefit the most from all this, though. 😏
Nothing surprising there. Industrial espionage has always been a thing. But this could be a serious blow to OpenAI. Anthropic is probably the one who’ll benefit the most from all this, though. 😏
Apple just sued OpenAI, and the wildest part is how they got caught: one candidate screenshotted confidential Apple files on his Apple work laptop hours before his OpenAI interview. Apple reads its own server logs. The recruiting pipeline generated its own evidence trail.
The complaint says OpenAI's hardware chief Tang Tan, a 24-year Apple veteran, directed candidates still employed at Apple to bring "actual parts" (batteries, logic boards) to interviews for show and tell sessions. One candidate was surprised, saying he didn't even know you could take those out of the office.
Apple also alleges Tan circulated an internal Apple offboarding document to coach new hires on dodging exit security checks, and that a departing engineer kept his Apple laptop, found a bug that still gave him access to Apple's cloud storage, and downloaded dozens of confidential hardware files after joining OpenAI.
Then the supplier: OpenAI allegedly got one of Apple's manufacturing partners to demonstrate a proprietary metal finishing technique by letting the partner believe Apple had approved it.
Over 400 former Apple employees now work at OpenAI. Apple says it flagged all of this to OpenAI in February and never got a response. Five months later, it filed.
The ask reveals the strategy. Apple wants an injunction barring OpenAI from using the secrets, the return of every file, and full discovery into io, right as OpenAI preps its first device launch and an IPO. If a judge grants it, OpenAI may have to prove the device was built clean, component by component, before it ships.
The device was supposed to run on the world's best hardware talent. Now its bill of materials is evidence.
@elonmusk Nothing surprising there. Industrial espionage has always been a thing. But this could be a serious blow to OpenAI. Anthropic is probably the one who’ll benefit the most from all this, though. 😏
AI THOUGHT OF THE DAY!!!
When you use different models for the same project—like Grok Build writing the code and Opus reviewing it—don't tell Opus 4.8 that Grok wrote it, because it will find bugs.
And don't tell Grok that Opus is unhappy with its code, or you'll hear a whole lot of new things about Opus.
A 23 YEAR OLD KARACHI CREATOR TURNED CHATGPT + GOOGLE FLOW INTO A $9,400/MONTH AI CARTOON STUDIO WITH ZERO ANIMATORS
dawood is 23, gym selfie profile pic, ring light in a spare room, no animation degree, no Wacom tablet
feeds a reference image into ChatGPT with one prompt — break down this cartoon style in extreme detail and understand how to recreate it — gets back a full style breakdown: line weight, shading rules, color palette, character proportions
pastes that breakdown into Google’s Gemini Flow to generate consistent frame-by-frame scenes, same characters, same style, no drift between shots
11 small brands and 2 YouTube storytelling channels pay him $400-$900 per finished short for custom-style cartoon content, no studio overhead, delivery in 48 hours
$0 in tools beyond subscriptions already owned, first client covered a month of software instantly
the window is open, follow and bookmark before every studio does this in-house
@gregisenberg I'll watch it when I get a chance, maybe I need to change my approach to Grok. But it's hard to filter out actually useful information from just clickbait headlines.
Grok 4.5 might be the BEST model to run inside Hermes or OpenClaw RIGHT NOW.
I've been sleeping on Grok to be honestNot anymore.
It's more than 60% cheaper than Opus 4.8 and lands around $2.49 per task versus ~$12 for Fable in Claude Code. And it's fast.
So what happens when you give Hermes + Grok 4.5 its own email, its own phone number, its own debit card, and access to every tool you use?
You pretty much get an AI co-founder.
Everything you need to know about Grok 4.5 + Hermes below.
Full episode is available to watch at @startupideaspod ( thanks @nickvasiles for coming on)
I slept on Grok. Not sleeping on it anymore.
Watch
@gregisenberg I haven't watched the episode. I tried working in Grok build today, and it disappointed me a bit. Even though it was fast — only 5 minutes of total work — the task of creating a web app remained unfinished and incomplete