Dan Martell - How to Actually Use AI in 2026

00:00:00.240 Everybody is giving you AI advice right now,

00:00:02.760 and most of it will hurt you.

00:00:04.820 We're gonna go over 15 pieces of advice that sounds smart,

00:00:08.180 but will set you back years.

00:00:10.560 And I'll give you the actual advice

00:00:12.240 you should listen to instead.

00:00:13.600 Number one, fire your team and replace them with AI agents.

00:00:17.120 Why this is advice?

00:00:18.720 Because having people on your team makes the business fun.

00:00:21.560 Why would you get rid of them, you ding dong? 1.00

00:00:23.480 Business is about people.

00:00:25.300 AI can't build relationships.

00:00:27.280 It can't feel the energy in the room.

00:00:29.280 Your team is your competitive advantage.

00:00:31.760 AI should make them 10 times better, not replace them.

00:00:34.940 Better advice, what I would tell people,

00:00:36.700 is use AI to be your co-pilot for your team.

00:00:39.560 Encourage them, train them, show them what they can do

00:00:42.400 so that they can do their best work, not replace them.

00:00:45.380 Number two, automate everything in your business immediately.

00:00:49.220 Why this is s*** advice.

00:00:50.720 If your process is broken,

00:00:52.860 automating a broken process just makes it broken faster.

00:00:56.660 It makes it s*** here.

00:00:57.520 I've seen founders waste six months automating a sales process that shouldn't exist because it

00:01:02.080 didn't work in the first place. And they're like, I'm going to use AI to make the sales automated.

00:01:06.140 It's like, you can't even sell. Better advice would be fix your process manually first. Elon

00:01:11.680 Musk did this with the model three production line, and then he over optimized and automated

00:01:15.880 it. And then he came back and he added people so that he could learn. Once you know what actually

00:01:20.340 works, then you start using AI to automate the biggest bottlenecks. Number three, let AI handle

00:01:26.020 all customer service to save you money.

00:01:28.520 This is bad advice.

00:01:29.900 Customer conversations are gold mines.

00:01:32.320 It's where the golden nuggets live.

00:01:33.900 You gotta go mine for it.

00:01:35.160 That's where you learn what's broken.

00:01:36.640 You learn what features to build.

00:01:37.900 You learn what problems to solve.

00:01:39.360 Cut that off and you're flying blind.

00:01:41.100 The better advice would be use AI

00:01:42.800 for the simple, repetitive questions or response times.

00:01:46.080 Have it answer calls, answer questions.

00:01:47.860 What time are you open?

00:01:48.760 Easy stuff, but keep humans in the loop

00:01:51.220 for the human touch points.

00:01:52.800 Business isn't B2B, business to business.

00:01:55.080 is H to H, human to human. 0.77

00:01:57.800 Number four, don't bother learning how AI works. 1.00

00:02:01.080 Why this is shit advice. 0.99

00:02:02.460 If you don't know how the AI thinks, 1.00

00:02:04.780 you won't know its limitation,

00:02:06.420 you won't know how to craft it,

00:02:07.600 you won't know how to massage it,

00:02:08.840 you won't know how to direct it.

00:02:09.860 I've seen founders make decisions

00:02:11.460 based on completely made up data

00:02:13.260 because they didn't understand how the AI makes up the answer

00:02:16.840 and it made up the data, it hallucinated.

00:02:19.580 So improved advice is learn the AI fundamentals.

00:02:22.940 That way you know what is possible

00:02:24.700 and what isn't and here's a pro tip ask ai to teach you ai you can ask it how does a language

00:02:30.060 model work make it simple use a metaphor use a simile and watch it teach you so that you can

00:02:34.940 then learn how to craft it and design it to make it even better number five build everything with

00:02:39.980 ai from day one this is why that's bad advice most people don't even know what they're building yet

00:02:45.180 so i recommend all the time start manual learn what works go into a company into a team reverse

00:02:52.060 engineer it then add ai to scale what's proven based on what you learn better advice would be

00:02:57.820 figure out what works manually first and then accelerate the process with ai don't replace it

00:03:03.500 then use ai to scale it number six build your strategy around the latest ai this is why this

00:03:09.740 is horrible advice not too long ago ai changed every three months then it was every month now

00:03:15.500 it's every week and sometimes every day learning what's hot today will put you in a position to

00:03:19.900 to waste a lot of time learning about something

00:03:22.020 that's not even relevant tomorrow.

00:03:23.400 Better advice is to focus on solving problems

00:03:26.520 that existed for years, pains that people have,

00:03:29.380 as Jeff Bezos often said about Amazon.

00:03:31.400 Think about the things that are not going to change

00:03:33.920 over 10 years, and those are probably the big things.

00:03:36.700 10 years from now, customers are still gonna want low prices.

00:03:39.120 They're still gonna want fast delivery. 0.53

00:03:40.760 There's never gonna be a day where the customer

00:03:42.360 doesn't want their package faster 0.57

00:03:44.140 or for the website to load faster.

00:03:45.720 So use AI today for what works best today

00:03:49.020 and go deep, master the skill of one platform.

00:03:52.540 So that way as the platforms evolve,

00:03:54.360 you'll already know how the whole suite of tools work.

00:03:57.060 Focus on solving problems that have existed for years

00:03:59.700 or will always exist.

00:04:01.020 Think about pain.

00:04:02.320 Number seven, let AI make all your business decisions.

00:04:05.840 Bad advice, so bad.

00:04:07.600 Here's the deal, AI still hallucinates

00:04:09.840 and it makes mistakes.

00:04:10.680 If you think about the concept of how AI works,

00:04:13.060 where it's predicting the next word,

00:04:14.820 by definition, it'll always give you

00:04:17.440 the most probabilistic answer,

00:04:20.080 the median watered down answer, no matter what.

00:04:24.400 And the best companies were built off innovation.

00:04:26.600 The best companies were built off intuition.

00:04:28.420 You shouldn't trust it with million dollar problems yet.

00:04:31.700 Better advice would be use AI for insights and analysis.

00:04:35.140 That's what it can do better than anybody.

00:04:36.800 Then verify it with people who've successfully

00:04:38.900 made those decisions to inform if it's the right strategy.

00:04:42.680 So use it to do research and tell you,

00:04:44.940 and then take that blueprint,

00:04:46.340 Show it to somebody who's been there before.

00:04:47.700 It says, does this make sense?

00:04:49.100 They go, wow, that's really smart.

00:04:50.940 But you need to be the director.

00:04:52.820 Let the AI do the task.

00:04:54.740 Now, before I move on to the next piece

00:04:56.020 of terrible AI advice,

00:04:57.440 if you're a business owner,

00:04:58.380 you're trying to scale past seven figures

00:05:00.100 and you wanna implement AI

00:05:01.380 into every single department in your business,

00:05:03.540 just go find me on Instagram,

00:05:05.040 danmartell, 2Lzmartell,

00:05:06.300 and DM me the word AI business,

00:05:08.160 and I'll send you over my full guide to do just that.

00:05:10.300 Number eight, replace all your brainstorming with AI.

00:05:14.260 Bad, bad advice.

00:05:16.340 It's so crazy.

00:05:17.340 AI can only reference what's existed before.

00:05:19.900 So by definition, it'll only allow you to brainstorm

00:05:22.860 things that other people have talked to in the past.

00:05:25.180 So if you're really starting from nothing

00:05:27.060 and you want just to understand what's going on, that's cool.

00:05:29.440 But brainstorming requires creativity.

00:05:31.420 Brainstorming is about finding these nuances.

00:05:34.160 When you think of like Henry Ford,

00:05:35.960 learning from the meat packers in Chicago

00:05:37.840 to create the production line,

00:05:39.120 which is how modern cars are still made today,

00:05:41.460 but borrowed from inspiration

00:05:43.000 from a completely different industry.

00:05:44.760 You can't do that with AI because it's only gonna give you

00:05:47.420 things that are probably gonna work

00:05:49.400 based on what's worked in the past,

00:05:50.860 not find new opportunities.

00:05:52.400 One of the things that AI can't do is have vision.

00:05:55.180 That is your job, to see a future

00:05:57.540 that doesn't exist yet, but should.

00:05:59.600 Better advice would be to use AI to spark ideas,

00:06:02.840 to research things, to expand on concepts,

00:06:05.420 to test crazy ideas in the real world

00:06:07.680 from a validation point of view,

00:06:09.980 not to be the driver of innovation.

00:06:12.020 Number nine, feed AI as much data as possible

00:06:15.160 to get the best results.

00:06:16.880 This is bad advice,

00:06:17.900 even though it sounds so great on the surface,

00:06:19.700 is because AI has this concept called context.

00:06:22.980 And if you give it too much information,

00:06:24.960 what happens is you have context rot.

00:06:27.000 Too much data confuses the AI.

00:06:29.000 It messes it up.

00:06:29.880 It doesn't understand what's relevant.

00:06:31.340 It's like giving somebody 10 maps

00:06:33.120 when they ask for simple directions.

00:06:34.860 You need to use the most concise, cleanest context

00:06:38.060 that actually solves your problem.

00:06:39.860 If you don't get the context right,

00:06:41.420 you'll get context wrought. Number 10, leave AI to your IT guys. It's just bad advice. Saying that

00:06:48.540 you don't need to learn AI because your job isn't AI and leave that to the technical person on your

00:06:53.660 team is like saying that you don't want to use a car because you're not a mechanic. AI today is

00:06:58.840 literally program using English. So if you're hearing the words coming out of my mouth, I know

00:07:04.980 you know how to use AI because you can talk. Better advice would be, hey, let's train everyone

00:07:10.820 on how to use AI to do what they need 10 times more, 10 times faster, 10 times better. Like I

00:07:16.840 said, AI is for humans. That's why it's coded in English. That is the first time in the history of

00:07:21.200 technology where a technology is programmed in English. So don't just leave it to the technical

00:07:26.680 guys. Everybody should learn this. Number 11, start by picking the coolest AI tools. This is

00:07:32.900 advice because cool technology doesn't solve problems. Cool technology is just cool technology.

00:07:38.200 Nobody wakes up and says, I need some AI.

00:07:40.580 They go, I need to solve a problem.

00:07:42.240 So if you're always just picking the coolest AI tools,

00:07:44.820 then yeah, you'll know a lot about tools,

00:07:46.400 but you won't actually know how to solve problems.

00:07:48.460 Better advice than what I tell my team

00:07:49.740 is to find the problem that's worth solving,

00:07:51.980 figure out what's the sequence of solutions

00:07:54.560 and then find the AI to solve it.

00:07:57.060 I use a framework called the theory of constraints.

00:07:59.540 You can ask AI about it and it'll teach it to you.

00:08:02.500 And that is how you choose the right problems

00:08:04.720 to solve with AI.

00:08:05.900 So start with the right problem that you're having

00:08:08.480 and then go find the right tool to solve it.

00:08:10.940 Number 12, use only one AI tool for everything.

00:08:14.740 If you think that one tool can do everything,

00:08:17.220 then you'll have the most watered down solution

00:08:19.280 to your problems.

00:08:20.460 Do you only have one social media app on your phone

00:08:22.720 or do you have several?

00:08:23.700 Why?

00:08:24.020 Because each one does something great.

00:08:26.080 It's the same thing with AI and the tools.

00:08:28.140 Each tool does a thing better than the rest of them.

00:08:30.500 So you need to know which tool to use for the right moment.

00:08:33.220 So if you're using ChatGPT for everything,

00:08:35.280 here's my advice for you. Build a toolkit of specialized AI. Match the right tool to the right

00:08:41.920 problem. You might be using Claude to do writing. You might use Gemini to do research because it has

00:08:47.660 3.5 tons more information than any other language model. You might still use ChatGPT for certain

00:08:53.460 things, but you might also find a tool to generate images that are better than all those three I just

00:08:57.980 mentioned. Number 13, use every AI tool that comes out. So this is advice because you don't

00:09:04.500 want to be just running around using every new tool that comes out. What you want to do is what

00:09:08.520 I just said, which is find the three to five that you need for your work that solve 90% of your

00:09:13.540 problems. Like actually your real problems that go deep so that you can become a master in those

00:09:18.160 tools. I know a lot of people that just get addicted to the new tools and I just call that

00:09:22.480 productive procrastination. They're addicted to the newness, but they actually don't do anything

00:09:27.260 with the tools. In many ways, they are the tool. Number 14, just copy paste other people's prompts

00:09:34.200 so easy advice. Why? Their outputs are not what you need as a result. If you just keep copying

00:09:41.080 everybody else's prompts, then you're just going to look like everybody else. Those prompts work

00:09:45.280 for those people based on their problems, not yours. Generic prompts create generic output.

00:09:50.180 The better advice is understand how to create your own prompts. Understand your process,

00:09:55.440 your inputs, your specific outputs. What problem are you trying to solve? And then create your

00:10:01.080 prompt based on. Maybe that is inspiration, but have it customized it based on what you need.

00:10:05.820 Number 15, AI will give you a competitive advantage just by using it. Poop advice. It's

00:10:11.120 so crap. Look, here's the deal. AI doesn't give you anything. It's just a thing you're using. I 1.00

00:10:16.500 know a lot of people that use AI every day and their life looks the exact same because they

00:10:20.240 don't understand how to actually use it to get their life moving forward. Just using AI won't

00:10:25.620 get you anywhere. It's about how you use it. So my advice for you instead would be your advantage

00:10:30.600 comes from combining AI with your unique process,

00:10:33.960 your relationships, your expertise, that's the moat.

00:10:37.460 That's where you take you plus AI to create real value.

00:10:40.920 Now, I know what I just gave you feels like a lot

00:10:43.140 and some of it was even controversial.

00:10:45.120 It's like master one, go learn the other ones,

00:10:47.600 build your toolkit, but here's the deal.

00:10:49.220 All I need from you right now is one mindset shift

00:10:51.680 that changed the way you think about AI.

00:10:54.260 I just need some feedback to know

00:10:55.880 that you got what I was sharing.

00:10:57.860 So ask yourself out of all those ideas,

00:10:59.700 which one did you need to hear most today leave a comment i'm going to read them all and remember

00:11:03.940 if you want to learn how to implement ai into every department of your business to go even past

00:11:08.340 seven figures just dm me ai business on instagram and i'll send you over my ai implementation

00:11:13.140 playbook now if you want to learn which ai tools you should actually use in 2026 click here and

00:11:18.820 and I'll see you on the other side.

Dan Martell - February 16, 2026

How to Actually Use AI in 2026

Episode Stats

Summary

Transcript