You’re not behind (yet): How to learn AI in 18 minutes — Transcript

Full transcript

1. 0:00Believe it or not, you're not behind on
2. 0:02AI, yet. Because the truth is, most
3. 0:04people still don't understand how to use
4. 0:07it well. By just watching this video,
5. 0:09you'll know more about AI than 90% of
6. 0:11the people out there. And if you make it
7. 0:13all the way till the end, you'll be in
8. 0:15the top 1% of all AI users. I've
9. 0:17personally tested thousands of AI
10. 0:19prompts across [music] every major
11. 0:21platform, built dozen of custom AI tools
12. 0:23for my own companies, and helped
13. 0:25hundreds of founders turn AI into their
14. 0:27most profitable team member. So, today,
15. 0:29I'm going to show you how AI works, the
16. 0:32fundamentals you need to use it, how to
17. 0:34master it, and how to future-proof
18. 0:36yourself so AI makes you a [music]
19. 0:38weapon instead of replacing you.
20. 0:39Starting with how it works. Essentially,
21. 0:42you've got all the world's information.
22. 0:44I'm talking videos, podcasts, documents,
23. 0:47Reddit threads. And then AI analyze it,
24. 0:49and then it did this mathematical
25. 0:51equation. It broke all that text into
26. 0:53[music] tokens, and then it tries to
27. 0:55predict the most logical next token
28. 0:57based on what you gave it. It's just
29. 0:59like if you say, "Twinkle, twinkle,
30. 1:01little," it's going to say, "Star."
31. 1:03That's why most of the answers that come
32. 1:05out of any chat AI have a lot of em
33. 1:08dashes. It's not because AI likes em
34. 1:10dashes, it's because humans like em
35. 1:12dashes, and it's trying to replicate
36. 1:14what it's seen us create. And this is
37. 1:16why if you [music] give it really good
38. 1:18information, the output is really good.
39. 1:20Because it can use all that input to
40. 1:23build context. [music]
41. 1:24It's kind of like if somebody tells you
42. 1:26all the details of a story and then ask
43. 1:28you a specific [music]
44. 1:28question, you're more likely going to
45. 1:30get it right because you have all the
46. 1:31details of the story. That's what AI
47. 1:33needs to give you the best outputs, and
48. 1:35that's where people fall short. The
49. 1:37truth is, AI isn't magic. It's just
50. 1:40pattern recognition with a fancy suit
51. 1:42on. So, now that we've covered how AI
52. 1:44works behind the scenes, we need to talk
53. 1:46about the fundamentals and how you need
54. 1:48to use it the right way. Starting with
55. 1:50the right way to prompt. The more
56. 1:51information you give it, aka the prompt,
57. 1:54the better the output will look. Did you
58. 1:56know you can give it like two, three
59. 1:58books worth of information about your
60. 2:00question before it even gives you an
61. 2:02answer? Most people are giving it a text
62. 2:05message. It's giving it nothing. And
63. 2:07then you're expecting it to do magic. A
64. 2:09few months ago I was working with a
65. 2:10founder that built this really cool AI
66. 2:12technology and he's working on the
67. 2:13marketing. And he was having a really
68. 2:15hard time getting the marketing to
69. 2:17actually communicate what his product
70. 2:19did. He spent like three weeks on it,
71. 2:21iterating, going back and forth, trying
72. 2:23to get the messaging right, trying to
73. 2:25get the headline right, trying to get
74. 2:26the copy right. All I did was sit down
75. 2:28and I gave it the format I'm about to
76. 2:30teach you and I hit enter. And within 12
77. 2:33seconds it outputted the perfect plan,
78. 2:36structure, and strategy for him to get
79. 2:38his customers [music] to buy. It took
80. 2:40less than 15 minutes to implement it on
81. 2:41his website and all of a sudden he's
82. 2:43converting that traffic into customers.
83. 2:45What did I do different? Here's how it
84. 2:47works. The first thing is the role. You
85. 2:50got to tell it who to be like. It's kind
86. 2:52of like act like a professional dancer,
87. 2:54act like a killer musician. So all I did
88. 2:57when I was helping my buddy to fix his
89. 2:58prompt is said act like a world-class
90. 3:01marketing strategist that focuses on
91. 3:03conversion for software. What that does
92. 3:06is it takes all the freaking information
93. 3:09in the world that's all been cataloged
94. 3:10and structured and it says, "Okay, out
95. 3:12of all of this, only focus on this
96. 3:14piece." And then with that it can then
97. 3:17give you an answer that's very specific.
98. 3:19If not, it just looks watered down. The
99. 3:21second part to your prompt is the
100. 3:22context. [music]
101. 3:23I want you to give it all the background
102. 3:25information. If you have marketing
103. 3:28documents, you have transcripts from a
104. 3:30call, you have any kind of standard
105. 3:32specification, the more you can give it,
106. 3:35it can use that in its little brain for
107. 3:37that prompt to guide it in its output.
108. 3:40The third is the command. You have to be
109. 3:42very specific. Whatever you want it to
110. 3:45do, tell it. Be very explicit. I always
111. 3:48say make the implicit explicit when
112. 3:50we're prompting. And the last the fourth
113. 3:53is the format. Often times for me, I
114. 3:55don't want these long, lengthy outputs.
115. 3:58I want bulleted lists. I want it short
116. 4:00and concise. I want it in a table
117. 4:02format. I want it in a PDF. The format
118. 4:05is how you take that output and then use
119. 4:07it in other tools. You might want
120. 4:08[music] to put it in a CSV so that you
121. 4:10can import that into another tool, but
122. 4:12if you don't tell it that, it's not
123. 4:14going to know what format you want. If
124. 4:16you do those things,
125. 4:17>> [music]
126. 4:17>> you will notice a complete difference in
127. 4:20the output. I'll even go as nerdy in the
128. 4:22format where I'll say, "Here's the
129. 4:25format." Then give it a template [music]
130. 4:27to follow, then hit enter, and it can't
131. 4:30go outside of those parameters. The
132. 4:32truth is is the quality of your output
133. 4:34will never exceed the quality of the
134. 4:36input to your [music] prompt. So now you
135. 4:38know how to structure your prompt, but
136. 4:40there's this other crucial step that if
137. 4:42you don't know, you can't learn AI
138. 4:44properly, which is you got to pick one
139. 4:46tool and don't switch. The amount of
140. 4:48people that come to me and they're like,
141. 4:50"Hey, I'm using ChatGPT and Claude and
142. 4:52Gemini and Grok." And I'm like, "Show me
143. 4:54your prompts." And it's Crayola. It's
144. 4:57simple. They haven't mastered the tool.
145. 4:59Learning AI is like learning an
146. 5:01instrument. If you tried to learn a
147. 5:03little bit about the piano and the
148. 5:04guitar and the drums all at once, you
149. 5:05wouldn't be really good. But if you
150. 5:07decided to pick one and you decided to
151. 5:09be a master at it, the others become way
152. 5:11easy after it because you understand
153. 5:13music. You understand tone. You
154. 5:15understand how to read sheet music. Same
155. 5:17thing with AI tools. Pick one [music]
156. 5:19and master it before you switch. Once
157. 5:21you understand how one works, the others
158. 5:23are way easier to pick up. And I will be
159. 5:25honest, the people that keep switching
160. 5:27tools all the time, I just call it
161. 5:29procrastination. Masters go deep before
162. 5:32they go wide. So know the tools, know
163. 5:34which one's right for you, and go deep
164. 5:35on it. So I've already mentioned the top
165. 5:37three. Okay, we've got Claude, Gemini,
166. 5:39and ChatGPT. I'm going to walk you
167. 5:41through each one and tell you exactly
168. 5:43which one does what best so that you can
169. 5:45pick the right one for your current use
170. 5:47case. So, for example, if you do a lot
171. 5:49of writing, creative work, deep
172. 5:51thinking, code, well, Claude's got
173. 5:54everybody beat there. And it could
174. 5:55change in a few months, but for right
175. 5:57now, I'm telling you Claude's the one.
176. 5:58The other one, if you do a lot of
177. 5:59research, you need up-to-date
178. 6:01information, you live inside of Google
179. 6:03tools, you can't go wrong with Gemini.
180. 6:05It is an incredible language model, and
181. 6:07it's the one a lot of people that are
182. 6:09like Google specific are using. GPT is a
183. 6:11great one. It's the first one to get
184. 6:13really popular, it's got a lot of
185. 6:14integrations, can't go wrong with it.
186. 6:16And before you start hammering me in the
187. 6:18comments, I understand there's a bunch
188. 6:20of others. I live my life on a site
189. 6:22called Hugging Face, which is like the
190. 6:23nerdiest AI community. So, don't tell me
191. 6:27about Grok and all these Chinese models.
192. 6:29I get that. I'm just telling it for most
193. 6:31people that are like, "I'm ready to be
194. 6:33[music] masterful at this so I can
195. 6:34become a weapon." These are the three I
196. 6:36would recommend. And by the way, if you
197. 6:38choose to go towards ChatGPT, I've got a
198. 6:40full freaking masterclass that I'll link
199. 6:42at the end of this video that goes
200. 6:44through everything to make you a pro.
201. 6:46And if you're a business owner that's
202. 6:48trying to break seven figures and you
203. 6:49want to go even deeper, I've got an AI
204. 6:51implementation playbook that is what I
205. 6:53use in my companies to implement AI, and
206. 6:56it shows you everything. Just find me on
207. 6:58Instagram and DM me AI business, I'll
208. 7:01send it over to you for free. So, here's
209. 7:03the thing. If you made it to this point
210. 7:05and you [music] feel like you've got
211. 7:06everything covered so far, the cool
212. 7:08part, you're in the top 10% of AI users.
213. 7:11But, if you stick around and implement
214. 7:13what I'm about to teach you, you'll be
215. 7:14in the top 1%. So, let's talk about
216. 7:17mastering AI starting with pull versus
217. 7:19push prompting. This is where we get a
218. 7:21little advanced, but it's fun. Come with
219. 7:24me for the ride. Don't run away. This is
220. 7:26going to be good. Most people use AI the
221. 7:29hard way. They do 80% of the work, and
222. 7:32they let AI finish the last 20%. That's
223. 7:34push prompting. You're telling AI how to
224. 7:37do the task. So, you want to go from
225. 7:39push to pull prompting. And the way that
226. 7:42it's different is that you start with
227. 7:43the outcome and then you let AI figure
228. 7:46out the best way to get there. Instead
229. 7:47of doing all the work yourself, which
230. 7:49you're doing today,
231. 7:50>> [music]
232. 7:50>> you let AI do the heavy lifting. The way
233. 7:52I like to think about it is push is kind
234. 7:55of like when you used to get directions
235. 7:56from a friend when they're like, "Come
236. 7:58to my house. Come over." And then they
237. 7:59tell you you got to turn left at the
238. 8:01house and then right at the church and
239. 8:02then you'll see a red mailbox. That's
240. 8:05how most people prompt. Pull is like GPS
241. 8:08where you let the AI direct you based on
242. 8:11the destination you've given it. So,
243. 8:12here's how you use pull prompting. It's
244. 8:14super cool. First, you give it the role
245. 8:16and the context like you normally would.
246. 8:17It needs to know kind of who it should
247. 8:19act like and what's the information
248. 8:21about the scenario. Second is you give
249. 8:22it the outcome-based objective, right?
250. 8:24This is where you tell it what you need.
251. 8:26So, you might say, "I need a sequence
252. 8:27that converts cold leads into booked
253. 8:29calls, right?" Very specific. It's like
254. 8:31the destination. Third, you say this,
255. 8:34"Ask me all the questions you need for
256. 8:37you to create this for me and then give
257. 8:39it back to me in you can say the format.
258. 8:41You might say PDF or a table. Then just
259. 8:43[music] answer all the questions. My
260. 8:45little trick there is I use the voice
261. 8:46mode where I can just hit that little
262. 8:48microphone and then answer cuz they
263. 8:49might give you seven questions and I
264. 8:51don't want to type that much. But you
265. 8:52give it all the answers and then let AI
266. 8:54do the work. Watch it cook. Now, you
267. 8:57might say, "Hey, I need to refine this.
268. 8:59Ask me more questions to make it more
269. 9:00relevant." And it'll keep refining. So,
270. 9:02you keep essentially using, that's why
271. 9:04it's called the pull method, you keep
272. 9:06having it pull from you exactly what you
273. 9:08need and let it go find the solutions.
274. 9:11This is how software programmers have
275. 9:12been doing it for a while, okay? Most
276. 9:14coders are sitting there and we're
277. 9:16coding through outcomes. It's called
278. 9:17spec-driven development. That's now
279. 9:19becoming the mainstream for how you can
280. 9:22prompt. Now, pull prompting is crazy
281. 9:24powerful. But if you really want to
282. 9:26master AI, you need to make it personal
283. 9:28to you. So, number five, master prompts
284. 9:31for every role in your life. The way to
285. 9:34think about a master prompt is kind of
286. 9:36like a manual about a part of your life.
287. 9:39So, if there was like a manual called
288. 9:40Dan CEO, what would be in that manual?
289. 9:43If there was one called Dan father, what
290. 9:45would be in that manual? Well, I would
291. 9:46assume in they would have everything
292. 9:47about me, my life, the people in my
293. 9:49life, how things are structured, what we
294. 9:51sell. When you have these master prompts
295. 9:54and you use it to give it to the AI cuz
296. 9:56you might switch to different ones, then
297. 9:58it'll give you better answers cuz right
298. 10:01now if you don't, it's not personalized
299. 10:03to you. It's just giving you generic
300. 10:05auto completion instead of something
301. 10:07that is laser focused to your exact
302. 10:09situation. And remember earlier when I
303. 10:11talked about giving it context, this is
304. 10:13the context. And if you do it once, you
305. 10:16never have to do it again and it makes
306. 10:18prompting so much easier and faster.
307. 10:20Every team member in all my companies
308. 10:22has to have a master prompt. They have
309. 10:24to show it to me. I got to make sure
310. 10:26it's complete. I just can't imagine that
311. 10:28they're efficient chatting with AI all
312. 10:30day long and it giving great answers if
313. 10:33it doesn't have that. So, now let's
314. 10:34create one for yourself. The first thing
315. 10:36is we're going to practice pull
316. 10:37prompting. You can say, "I want to
317. 10:39create a master prompt for my role as
318. 10:42whatever your role is in life, you know,
319. 10:43father, husband, VP of marketing and ask
320. 10:46me all the questions you need to create
321. 10:48it for me." See how we're going to have
322. 10:50it pull from me? Now, it's going to give
323. 10:51you all these questions so they can
324. 10:53create this document and that's where
325. 10:55you use the voice-to-text feature
326. 10:56[music] and just answer each one
327. 10:58conversationally. And third, once AI
328. 11:01gives you your master prompt, go through
329. 11:03it and just refine anything that doesn't
330. 11:05feel right. And you can literally just
331. 11:06talk to it and say like, "Update this.
332. 11:08Add this. You're missing that. You
333. 11:09didn't understand this properly." And
334. 11:11it'll literally update the whole thing
335. 11:13for you. The last part is to save it as
336. 11:15a PDF and I like to just call the master
337. 11:17prompt and then the role name so I have
338. 11:18them on my hard drive and that way when
339. 11:20I'm using different AI tools, I can
340. 11:23always upload it so that it has perfect
341. 11:25context on my situation in life. The
342. 11:27truth is if AI sounds like a stranger
343. 11:29and anybody you have got that output,
344. 11:31it's because you haven't introduced
345. 11:33yourself. The master prompt is how you
346. 11:34say, "Hello AI, this is who I am." The
347. 11:37cool part is like I said, you can use it
348. 11:39in other tools. I personally spend a lot
349. 11:41of time in a tool called Manifold, which
350. 11:43lets you create a genetic workflows.
351. 11:45[music]
352. 11:46And when I first loaded up the tool, it
353. 11:48didn't know anything about me. So, I use
354. 11:49my master prompt to give it all the
355. 11:51context so that it could do outputs that
356. 11:53were very personalized. If I had to like
357. 11:55play with it and tweak it and talk to
358. 11:57it, it would have just took me a long
359. 11:58time. This gets an AI from stranger to
360. 12:01best friend. So, now you get the master
361. 12:03prompts. You have a way to give the
362. 12:04context to the AI so that you guys are
363. 12:06best friends. But, what if you're like
364. 12:07trying to get an output and you like the
365. 12:10output and you want to like repeat it
366. 12:12and keep it consistent? This is where
367. 12:14you need the next tool to master AI.
368. 12:16Number six, system prompts. Think of a
369. 12:19system prompt like a recipe and it has
370. 12:22all the structure and you have the
371. 12:23ingredients, but the sequencing matters.
372. 12:25How you do what you do in the order will
373. 12:27dictate if you have a big pile of mush
374. 12:29or you have a beautiful chocolate cake.
375. 12:30So, if you're chatting with the AI and
376. 12:32eventually it gets an output that you're
377. 12:34like, "Oh my gosh, this is great. Why
378. 12:36did it take me 17 minutes to get to that
379. 12:38output?" This is how you solve that
380. 12:40problem. Understanding how to create
381. 12:42system prompts is the future. [music]
382. 12:44So, here's how you build a system prompt
383. 12:45using what we've learned so far. First,
384. 12:47you got to start by saying,
385. 12:48>> [music]
386. 12:49>> "You're an expert AI engineer. I want
387. 12:51you to create a system prompt that does
388. 12:54X. Ask me all the questions you need to
389. 12:56create." See how I'm using the pull
390. 12:57structure? Next, once it gives you all
391. 12:59the questions, answer them using the
392. 13:01voice-to-text feature, then test the
393. 13:04response you get back. The third, if
394. 13:06something needs to be tweaked, just tell
395. 13:08it to refine it. It's literally you code
396. 13:10system prompts with words. Once it's
397. 13:12dialed in, then you copy that system
398. 13:14prompt into a custom GPT or a cloud
399. 13:17project or a Gemini gem. And then you've
400. 13:19got a system that works for you on
401. 13:22repeat without ever having to give it
402. 13:23instructions [music]
403. 13:24ever again. I use this for strategic
404. 13:27work, I use it for research work, I use
405. 13:30it to create systems for my team to
406. 13:32follow where I can give them the link
407. 13:33and say, "Look, just give it the topic
408. 13:35and it'll do something for us." That way
409. 13:37you take the complexity of AI and just
410. 13:39make it really simple for everybody else
411. 13:41to use. You're basically building little
412. 13:43machines that take care of work for you
413. 13:45so that you can focus on the fun new
414. 13:46stuff. Now, here's a pro tip. A while
415. 13:49ago, there was a leak. All the complex
416. 13:52system prompts from the top AI tools,
417. 13:54I'm talking Perplexity, Notion, Lovable,
418. 13:56and 20 plus others were actually
419. 13:58collected and put together. So, if you
420. 14:00want to go see what the world's best
421. 14:03system prompts could look like, just
422. 14:05search system prompts and models of AI
423. 14:07tools on GitHub on Google and you will
424. 14:09find the link and you can go look at
425. 14:11each one of them to understand how
426. 14:13powerful these can be created to do
427. 14:16literally anything for you, like build
428. 14:19mobile apps or reconcile financial
429. 14:21details. So, now you know how to use AI
430. 14:24better than most people ever will.
431. 14:26That's the truth. But, here's the thing.
432. 14:28AI is evolving fast. I'm talking every
433. 14:31day. And if you want to stay ahead, you
434. 14:32need to make sure AI will never replace
435. 14:35you. Which takes us to our last point,
436. 14:37but the most important one, which is
437. 14:39future-proofing yourself. Which comes
438. 14:41down to taste, vision, and care. There
439. 14:44are three things that when I sat back
440. 14:46and I asked myself like, "What are the
441. 14:47areas of human existence, of life, that
442. 14:49I feel is going to be really hard to
443. 14:51disrupt?" I feel like things like our
444. 14:53taste for what is great and our vision
445. 14:56for things that don't exist that should,
446. 14:59or caring about other people. What a
447. 15:01crazy idea. I really don't think that AI
448. 15:02is going to be able to do that well.
449. 15:04Simulate it, maybe, but I'm talking
450. 15:06heartfelt, empathetic conversations.
451. 15:08That takes you. Machines can optimize
452. 15:11what already exists, but humans imagine
453. 15:14what doesn't. First thing is taste. You
454. 15:16got to immerse yourself in excellence.
455. 15:17[music] I think of it like standing
456. 15:19underneath a waterfall and allowing
457. 15:21amazing, world-class things to be poured
458. 15:24over me so that I might have the
459. 15:25possibility of picking stuff up. So, if
460. 15:28you're like massively interested in
461. 15:30music, go subscribe and use your social
462. 15:33media to learn music. If it's code,
463. 15:35software, I want you to use your feed to
464. 15:38feed your mind. Go and follow all the
465. 15:40top people. Go subscribe to the
466. 15:41newsletters. The more you consume the
467. 15:43best in the world and you get to
468. 15:45understand what greatness looks like,
469. 15:47the better you build your taste. The
470. 15:48second is vision. Schedule thinking
471. 15:51blocks so that you can sit down and
472. 15:52think about the future. Visualize
473. 15:54concepts cuz your brain doesn't think in
474. 15:56words. I don't know if you know this,
475. 15:57but like if I say pink elephant, you
476. 15:59just saw it. Okay? You didn't have the
477. 16:00words written down in your mind. So,
478. 16:02even just doodling, planning, exploring,
479. 16:05reading books that'll expand your
480. 16:07horizon, that gives you a sense of
481. 16:09what's possible or maybe how the world
482. 16:11should be shaped around you. And that is
483. 16:13your job to shape the AI to help you
484. 16:15with the vision. And the last one is
485. 16:16care. And this is why I'm so bullish
486. 16:19around AI doing all the boring and dirty
487. 16:22and dangerous work for us [music] is
488. 16:23then you get to celebrate your clients.
489. 16:25You get to celebrate your team. You get
490. 16:27to be with your family. You get to use
491. 16:29it to create more cuz you were born to
492. 16:32be a creator. Like that's what you're
493. 16:33here to do. Then you have the time to
494. 16:35collaborate with people authentically.
495. 16:37Start thinking about ways you can get
496. 16:39out in the world and show that you care
497. 16:42and contribute and just connect with
498. 16:44people. I really think that AI has
499. 16:46enabled us to have this deep desire to
500. 16:48be around other people. you see events
501. 16:50having record number of people showing
502. 16:52up and sold out in no time cuz we have a
503. 16:54deep need for connection. Do these three
504. 16:57things and AI will become your weapon
505. 16:59instead of your replacement. Now,
506. 17:01believe it or not, you're still early.
507. 17:03So, stop panicking if that's where
508. 17:05you're at. Most small businesses haven't
509. 17:06even started using AI in a meaningful
510. 17:09way. So, just by watching this video,
511. 17:11you're already ahead of the game. So,
512. 17:13here's what I need you to do. Just pick
513. 17:15one daily task [music] and have AI do
514. 17:17it. I don't care if it's move
515. 17:19information from one system to another.
516. 17:21Just 10 minutes [music] of action beats
517. 17:2310 hours of tutorials. So, just leave me
518. 17:25a comment and leave that commitment
519. 17:27below. I want to know, what are you
520. 17:28going to commit to to really upgrade
521. 17:30your AI skills? And remember,
522. 17:32>> [music]
523. 17:32>> DM me AI business on Instagram if you're
524. 17:35looking to grow your business and I'll
525. 17:36send you my AI implementation playbook
526. 17:38so you can scale [music] past seven
527. 17:39figures. Like I mentioned, if you want
528. 17:41the full ChatGPT masterclass to get
529. 17:44ahead of 99% of the people, just click
530. 17:46here and I'll see you on the other side.