Most people use AI like a faster Google or a writing assistant.

That is not where the real leverage is.

In this episode of Prompted, Kyle sits down again with Matthew Stein to break down seven practical ways to use AI as a true thought partner, not just a tool. From idea generation and structural support to persona simulation and strategic thinking, this conversation shows how modern professionals are using AI to think better, move faster, and avoid blind spots.

If you already use AI but feel like you are only scratching the surface, this episode is for you.

👉 Demo agent shown in the episode: https://agent.ai/agent/screen-test

Everyone's heard of hallucination, and this is really turning hallucinations into a feature.

Like, you give it the core concept.

Now, do this interesting thing with it.

You want it to make up things.

You want it to generate stuff that may not work, may not be realistic.

I don't know where to start.

Like, this gets you started.

Boom, you've got structure, you've got all the pieces you need in place.

Hey, pick this apart.

What is missing from this?

What did I forget?

For A statistical piece of software, how

good it is at faking emotional resonance.

It knows what things people have written to get stories to resonate with people.

And so you can ask it to do that.

In these scenarios, like the average structured document is good and important because it remembers, like you were just saying, all those elements of it that are super important, not to forget that we're all passionate and excited about this one idea that we have, but we forget the rest of the scaffolding there.

Getting into that higher level thinking, hypothesis generation,

campaign strategy.

It really does help you think through things in more of a complete way.

Matthew, welcome back to another prompted episode.

You're like the reoccurring side guest here, side host here.

That's right, Kyle.

Thanks.

Good to be back.

Absolutely.

Love to get a chance to hang out and chat with you about fun stuff.

We had recently done a builder workshop for a company, and one of the things we did in that workshop was went through a whole AI as a thought partner piece and talked about 7 different models that people can use AI as a thought partner.

I was like, whoa, this is cool.

This is actually really valuable.

We should come back and bring this out and share it with the wider audience here.

So

I'm sorry, let's get back out here and talk about this.

Yeah, this was a great workshop.

It was a two-hour workshop, but this podcast will not be two hours.

We cut out a bunch of stuff, really sort of distilled it down to the essence.

So we got 7 topics we're going to get through and give every one of our listeners some like tactical applications of these, but they're pretty high-level strategic moves in order for you to sort of like boost your everyday things that your problems are trying to solve and tasks you're trying to get done.

Yeah.

And I'm sure you've heard it too.

And I've seen it a bunch and talked about it with a bunch of people.

It's like 2025 was like testing, playing.

It's like 2026.

Let's actually start using these things in practical, real ways.

And same for us, like us playing with and testing a lot of stuff.

These six or seven models that we'll talk about, motions as a thought partner, they just kind of solidified and they work really well.

So it's like, let's put them all together because I guarantee you, probably played with some of these too, but you probably haven't thought of all of these holistically.

So you should talk about it.

Absolutely.

Let's get right into it.

Let's dive in.

Yeah.

So the first one that I like to talk about is idea generation.

And we should say it too, these aren't necessarily agentic specific.

This is you're working with your LLM of your choice.

They're LLM agnostic.

It doesn't matter which one you're working with.

Claude or ChatGPT or Gemini or Crock or whatever.

These all work in every one of those models.

Yeah, absolutely.

A lot of these are tool non-specific, but they do allow you to apply them to different ways because different tools act differently.

And idea generation is such like a blanket.

It sounds like the most boring thing, but you can go pretty deep with this.

So why don't you talk a little bit about what you mean by idea generation?

Yeah, so I'll just throw one out there, like a generic prompt that you can use, like generate 15 ideas for content type about subject.

and rank them for your persona based on likelihood to resonate.

Explain the ranking, identify your top three picks, right?

So you're doing a lot in that prompt, right?

I want you to generate an idea about...

whatever it is that you're writing about, and make sure that it's targeting this audience.

But you're going a step further and just give me 15 ideas.

You're also asking for it to rank them, at least the top view, and to tell you why to defend its reasoning there.

So pretty powerful thing, especially if you're brainstorming, just kind of come up with ideas, right?

Yeah, one thing that I love about this is you get a whole bunch of people in the room to

to work on brainstorming or to come up with new types of ideas or uses or whatever it could be, and you're going to hit a certain amount of fatigue, right?

People just get tired.

Now, at the end of a human brainstorming session, the new ideas that people tend to generate, like if you keep pushing them at the end, they do start to generate better quality, more unusual ideas.

But with AI, you don't get that fatigue.

You can run into, if you're hitting that like the bell curve, right?

that top of the bell curve, that's where it's going to give you most of the things from in its distribution.

But one thing I love to ask AI for when I'm doing, what I want more off the wall ideas, when I want things that people wouldn't normally come up with is say, give me ideas across the whole range of your distribution.

Get out to the ends of these statistical models

And you can put that right in your prompt, give me ideas across the whole range of your distribution, and it's going to come up with some crazy ones.

If you want to try this, and I've done this with my kids, ask it for knock-knock jokes.

You go and you ask it and you find like over and over again, it's going to give you the same 3, 4, 5 knock-knock jokes.

But throw that extra line in there, give me ideas across the range of your distribution,

and it's going to find some like ultra nerdy ones or really dumb ones, which are the ones that four or five-year-olds love the most.

But you got to leverage these statistical models, make them work for you.

And if you want tons of wild ideas, that's a great little hack to add to your prompt.

Yeah.

And the thing I think is most hidden in this is like, it will absolutely come up with things that you've never thought of before.

Or even if you're working at a group and you get to that fatigue point, it will come up with things that no one thought of in that entire group.

And I think that's the interesting piece of it.

Yeah, absolutely.

And this also brings us to, everyone's heard of hallucination, and this is really turning hallucinations into a feature.

You know, this is the time when you want it to make up things.

You want it to generate stuff that may not work, may not be realistic,

You as the human with perspective and judgment will make the decision whether or not to use those things.

But yeah, it's going to come up with some just totally off the wall ideas.

And so this is a great way to leverage hallucinations to your advantage.

Yeah, and I'll just close on this one before we move to the next one, just kind of three specific examples for you out there.

Give me 10 headline ideas for a blog post, reducing churn in SaaS.

Rank them by predicting engagement for a VP of customer success, and explain the ranking.

Propose 5 campaign angles for our new feature launch.

Label each angle as safe, bold, or experimental, and why you gave it that ranking.

Suggest 7 metaphors or storytelling framework for explaining our complex analytics product in simple language, and why.

Like, seems so obvious, but you know, super, super powerful.

Yeah, and if you're stuck in a rut or you were just asked to scrap your entire campaign and come up with 30 new taglines, this is my go-to.

This is what I'm coming back for.

Yeah, So the next one here, structural support.

You know, you can use AI to help you design the architecture of your context.

of your content.

So you start with a clear blueprint before you begin creating.

And I like this because when I start writing, if I'm doing a new strategy doc, when I'm doing my writing, and I still do a lot of writing by hand, is I'm thinking about what is sort of my core message, and it doesn't have to be an introduction.

I typically rewrite my introduction when I'm done anyway to get a good hook in.

But I'm thinking about, what's my core problem?

How am I going to approach it?

What are various resources I've got?

And then I like, remember, it's like, oh yeah, I should add goals.

Oh, I should add how I'm measuring this.

But if you ask it for that structural outline, the architecture doc for whatever you're working on, it's going to fill in a lot of those blanks for you so that you can just say, it's like, oh yeah, I'm going to work on this section.

Oh, I had a good idea.

I'll jump down to this one.

And I don't have to worry about missing out on how are we going to hit our KPIs for this?

Or what's our budget going to be?

It'll put all of those.

that structure in place for you to go and fill it in with your ideas.

I think this is the perfect example of kind of two concepts coming together.

One is like blank canvas syndrome, right?

Or like, I don't know where to start.

Like this gets you started.

And 2, we talk a lot about AI is really good at bringing out blah or average stuff.

In these scenarios, the average structured document is good and important because it forgets, it remembers, like you were just saying, all those elements of it that are super important, not to forget that we're all passionate and excited about this one idea that we have, but we forget the rest of the scaffolding there.

Absolutely.

And you've got a couple examples of how you like to use this one.

You know, I use it for campaigns.

What are some ways that you've put this one together?

Yeah, I'll read some of these examples.

We kind of

Create an outline for a landing page targeted at first-time HubSpot admins.

Include sections, subheaders, bullets, sample CTA language.

Give me all that detail.

So it's going to give you back almost that PRD you can take to your design team or any of your requirements docs or to your marketing agency.

And boom, you've got structure, you've got all the pieces you need in place.

It can start to fill things in for you.

Sometimes you might ask it not to fill in the details.

Just be like, give me that outline structure.

I'm going to go put in the stuff that I care about.

But it can also get you started again.

Go from that blank page to something working.

Yeah.

Rewrite this blog outline using AIDA or PAS frameworks.

Explain the changes and why they improve conversion.

Okay, we break those down.

What are the AIDA frameworks and what's the PAS one?

Good, call out.

AIDA, attention, interest, desire, action.

It's kind of a specific journey there that you take people through.

Kind of, hey, get their attention, tap into their interest, make sure that the desire there, and then get them to act.

right?

And then PAS is more the problem agitation solution framework of like, what's the problem?

Is it something that's painful to them?

Let me offer you up the solution for it.

Yeah, I like those two frameworks because they're pretty common.

They're fairly well known.

Since we're talking about structures and frameworks that you want to bring to the table, they're really good.

I know that there's a lot of other frameworks that people use internally.

and maybe your company has a proprietary one or a particular type of doc that it's got, Amazon's got the, what was it, the six-page meeting doc, but you can also feed it with those formats and then have it start to fill those out for you.

That way, maybe you already have an existing structure, so feed it the structure, it'll start filling that stuff out for you.

Yeah, And all of these are kind of a little bit more marketing heavy, but you can really see how you could have put on other things.

So kind of a final third example on this one would be, break this webinar transcript into a structured set of themes and turn each theme into a blog outline.

Love this for repurposing content.

Rip it apart, yeah.

All right, another one is the critical friend.

I'm not saying you are, you judge me too much, but sometimes I use you as a critical friend, but how do you use AI as a critical friend?

Yeah, you could use AI as an honest reviewer who points out weaknesses, gaps, or unclear logic before your audience does.

Think about it as someone who can say, uh-oh, what am I missing?

Right?

I ask this all the time.

when I'm working with kind of the LLMs, like, hey, pick this apart.

What is missing from this?

What did I forget?

What other question do you have that you need clarified on before you kind of move forward with this?

You know, it's just, we don't know what we don't know.

Like I said earlier, we're focused on something sometimes so much that we forget all the other elements around it.

And like, this is where reach into the power of average, right?

Because average thinks about all of those things.

And I like this because

It's so easy for me, especially, I get excited about ideas and I think about all the ways that this is going to succeed and how great the outcomes are going to be.

And I have to force myself to stop and think about, okay, what could go wrong here?

What am I missing?

And if I'm talking to other people who are on my team or similarly excited, we're not going to look to the negatives necessarily.

I'm a big fan of doing pre-mortems.

That's not a, like a post-mortem is after something is dead or has failed.

pre-mortem is when you are thinking in advance of something happening, what are the ways that it could go wrong?

So using it as a critical friend, great way to get into that sort of pre-mortem mindset and get it to point out some of the weaknesses or things you've overlooked first.

Yeah, I mean, what we're talking about here is it's like strategy planning or war planning, right?

Like, what are all the ways that this could go wrong?

And I think this critical friend option really helps you

you have fewer blind spots because you're thinking about all these things.

And all of these things are being backed more by clarity and reasoning, right?

There's logical reasons and thoughts going into all of them because it's just thinking about all of the different angles and all of the average things that need to be thought about for this scenario.

And you can use this to prep before even pitching an idea as well.

So you can think about how is this idea likely to land or how is it going to be perceived?

and get that feedback and sort of do a little bit of a testing on your messaging before you even take it to a meeting or wherever you're going to pitch something.

It should be called out and told a little bit, though, that these LLMs are really good at complimenting us all the time.

So I don't know if you feel this way, but I almost like pushing this on it a little bit because, oh, come on, don't just pat me on the back all the time.

Be a little bit more critical.

So we have to nudge it in this direction.

But I think

Anybody that's worked with LLMs for a while knows and feels some of that, get a little bit more in the weeds with stuff.

And this really forces you to call that out.

Yeah, and some of them are really syncophantic, right?

You'll say, I forget, there was a great example of this where it's, I forget which model of ChatGPT was, I feel like it was four, but they said, what are the top five rice growing countries in the world?

And it was like, China, India, someone else.

And it's like, no, actually, India's #1.

Oh yeah, you're totally right.

India's #1.

And it's like, no, actually, it was right the first time.

It's not going back to fact check you unless you ask it to.

And so have it play the role of that critical friend or the fact checker.

And I was like, are you sure that's right?

Have you found that any of the models are more or less syncophantic or less likely to agree or more unhinged?

I think you could set up in your settings to ask them to do more of that.

But I think in general, I don't think any of them are or less.

No.

I mean, I don't have any strong conviction one way or the other there.

But yes, I have set up on certain projects, trained them to be a little bit more like this.

Yeah.

I know Grok is pretty unhinged.

I don't spend a ton of time on it, but that one's, if you want the really wild ideas, you know, probably check that one out.

If you're someone who likes to experiment with all the different models there.

Yeah, and just to give three kind of examples kind of in the marketing perspective of this one.

Act as a skeptical CMO.

Critique this e-mail nurturing sequence.

What would stop you from replying or booking a demo?

Act as our ideal buyer.

buyer.

Read this landing page and tell me which parts feel generic, confusing, or unconvincing.

And this is great because it's calling out that very first beginning, right?

Like act as our ideal buyer.

You should have as a doc that's ready to go, who is your ICP, your ideal persona, or multiple of them.

And then you can say, feed that in, give it that context.

if you're doing this agentically, maybe it's already built into the agent.

If you're just using an LM as a chatbot interface or through a project, you know, have those things already fed in with a.

a lot of their perspective, the things they care about, the jobs to be done, whatever demographics, firmographics, whatever it might be.

But then when you're saying and you're asking it, it's like, you know, act as our ideal buyer, maybe you've already got that persona defined.

And so it's going to be able to be even more effective at doing that.

Hey, it sounds like you're talking about AI onboarding, something we covered in our recent episode of like predictions for 2026 there.

It's true.

Yeah.

How do you set it up with all the context it needs?

2026 predictions episode is a real good one.

Yeah, plug, plug away.

Yeah.

All right, next one here, creative partner.

This is the obvious one for everybody, right?

Like, use AI to sharpen the craft of storytelling, narrative framing, and emotional resonance.

Okay, so what do you mean by creative partner?

We've already had...

the idea generation, but what do you mean by the creative side of creative partner?

Yeah, so I think this is one of the ways that people are probably most abusing AI and where we get all the slop from, right?

This is to like, create the content for me.

So you've got the idea, you've narrowed it down and done that, and now it actually does some of the writing.

You've got to give it the proper guardrails, the persona, the topic, and all of the context that you can.

But this is where it actually does some of the heavy lifting for you, right?

This is like, as that thought partner, I've got this idea for a subject, but can you turn it into 750 words, for example, right.

So you're talking about delivering like the actual creative material.

So if you're writing a video ad, like what is the story arc of that ad that you're putting together?

if you're coming up with, images is another one, right?

Like creative for ads, you need images, you need not just give me another 20 taglines.

This is more about like figuring out how to, craft an arc or, develop a narrative hook that's going to get people interested for, whatever your LinkedIn post or your, YouTube shorts going to be.

It's that creative as an asset, not just using creativity.

And I think this is also too where you could start leveraging yourself, right?

Like, or amplifying yourself.

Because we've talked previously about the journalist versus the editor in a newspaper, persona shift, right?

Like, you're the editor now.

You know what this article should say and how it should say it, but that doesn't necessarily mean that you're filling in all of the, every single sentence and all.

And so yeah, it's filling in that gap for you and really amplifying what you do.

And it's not replacing your creativity, right?

Your core spark, the idea, the concept is there, but really filling it out is kind of what this creative partner's doing with you.

Yeah, so it's helping to get that, again, it's that craft of the storytelling or your narrative framing.

But I'm still amazed at, you know, for a statistical piece of software, how good it is at faking emotional resonance.

It knows what things people have written.

to get stories to resonate with people.

And so you can ask it to do that.

You can say, I want to evoke these emotional feelings, or I want it to, you know, chug on some heartstrings here.

And as it's writing, it can do a pretty good job of that.

Yeah, So some examples of that, right, are like, propose a story, you know, an analogy or a metaphor that can help explain how our product reduces waste spend, or transfer this dull introduction into a narrative hook.

right?

And make sure that it's emotional or data-driven or humorous, right?

Choose your angle there.

Like you give it at the core concept.

Now, do this interesting thing with it.

I like that.

That's really good.

All right, next up is to use it as a strategy partner.

I do this all the time.

In my last role, I was writing a lot of strategy docs.

And for me, the way that I think about a strategy doc is, you know, do you have tactical to-dos?

That's your sort of project management side, but the strategy

allows other people to make the same decisions that you would make.

So giving people this sort of like decision-making framework, the guidance, the goals, knowing that makes sure that everyone's on the same page.

So you've got this down as like, how do you use it as a strategy partner?

How do you like to do it?

I think the important thing, this is beyond the writing, right?

This is like, the good thing about data is

The data can always tell you something.

It's like, what do you want it to tell you?

So ask it to identify patterns, opportunities, risk.

Spit a bunch of those out.

You can kind of pick and choose which ones resonate most with you.

But it kind of, once again, takes that blank canvas or takes this lump of data and puts some ideas down for you then to kind of start working with.

And like, oh, this was interesting.

Let me keep digging deeper into that.

But it could very quickly

get you off the mat and start thinking in the strategy by putting a bunch of ideas in front of your face that it's already like done some of the intro work for you.

I had a marketing professor who always said, if you torture data long enough, it'll tell you whatever you want.

So you do got to be a little bit careful about that.

But these things, you can ask it, you know, what are some patterns or trends within this data as opposed to how do I shift this data to justify my thinking?

But that's not what we're talking about here.

You know, I like it for

getting into that higher level thinking, hypothesis generation, campaign strategy, it really does help you think through things in more of a complete way.

And again, it'll remind you of things that you might have forgotten here.

I mean, I like this in terms of content calendars or launch strategies, even positioning refinement.

making sure that all of the pieces align, that whatever you're pitching is going to be resonating with your target persona here, and then you're bringing all the pieces to bear the same way.

And as someone's out there doing their little bits of the work,

they now have this strategy as guidance to make sure that their pieces of work are aligned with the larger goals, whether it's for your team or your division or the full company.

I'd say the really solid piece that we want to make sure that people really resonate or finally understand at the end of this is like, it's presenting the ideas.

It is not the final decision maker on this stuff.

Right.

So it's surfacing possibilities.

You don't have to necessarily accept or do them all, but it's like, it's giving you rationale and reasoning.

But ultimately, you still should be the final decision maker.

You are still in the loop and you should remain in the loop.

Yeah, you have to be.

Don't outsource your whole job to all the thinking of this.

Let it do the first 80%.

Exactly.

And I do think that the higher level sophistication that it's doing,

or that you've got it producing for you, the more important it is for you to really look through and provide your own, again, I always say like perspective and judgment.

Those are the two main human characteristics that we bring to the table here, because otherwise it can go off the rails.

So you do want to make sure, like take the ideas it's giving you, think about them, and then make your decisions.

But it'll help give you, know, again, zero to one or a lot of the framework in order to put a lot of these things together.

So let's get into a couple of these examples because I think that'll really solidify this for people.

Review last month's funnel performance data and suggest 3 hypotheses for why convergence dropped to propose experiments to test each one, right?

Like you could see, like it's proposed to these things, but you still have to like, oh, that's really interesting.

Or, oh, no, that one doesn't really work.

I have an explanation for it that maybe the model doesn't have or doesn't know because it doesn't have that context.

Right, it might not have the context.

It doesn't have the full total picture.

Or maybe, you heard from a sales rep at lunch that, they tried this thing and someone didn't like it, or they had prospects that are just, nixing that idea.

Another one, analyze this competitor's website and summarize their positioning strategy.

Suggest opportunities where we can differentiate.

And I know we're providing a lot of

A lot of marketing examples with both that we have backgrounds in marketing, but there's also, some of the customer success, I think customer success angles of, how do we go for upgrade cycles?

How do we identify churn risks?

You know, if we've got a brand new

feature that's coming out or new product launch.

So we want to get people, how do we identify the best fit customers to roll this out to and give your CS team a blueprint to go and find their next upsell and find the people who would.

most use whatever the new features are.

Yeah, Sales example could be something like, look at this customer and look at their marketing position and identify the best, some three possible strategies I might approach them of why our solution solves our problem, right?

And then you pick and choose the one that makes the most sense and go with it.

Yeah, I like that one.

All right.

Another one, this one's a favorite of mine is persona simulation.

Why don't you tell us about persona simulation?

So this one kind of dives into a lot of the different ways that we, the ones that we've talked about, but I think it gets more specific in it.

And I think this one might be one of the most underrated and probably the least used one, right?

Ask the AI to simulate your audience or stakeholder or individual contributors.

We've talked about bringing the persona into these other ones, but like let it be the persona for you.

That could be your boss, your customer, your prospect.

It could

your support person you're trying to work with, it just helps you understand that empathy among a lot of different things.

Yeah, and this one is great because you can have interactive chats with these people.

It's like getting a focus group.

It's not exactly 100% like it, but it gets you 90, 95% of the way there.

And it allows you to define who that person is as opposed to paying a lot of money for someone to go out and find the right fit person.

You just have it pretend, have it do the role-playing for you of this persona simulation.

Now, of course, we've got an agent for this.

I'm going to go ahead and share my screen here.

What we've got here is it's called the screen test agent.

So this works

a lot like a focus group.

And you start out by defining like what kind of audience is this for?

And I've pre-run this, pre-loaded it with a fake app that I've thought about making called Art Run, right?

And the target of this is runners 25 to 40 who want to stand out on Strava.

And this says just share your speech, concept, project outline, notes, could be your emails, could be whatever it is, your ideas, your strategies, especially if your audience is the boss.

but you paste in what it is that you're working on, and it gives you back a few people that you can interact with.

And when you go, so you say it's like, okay, we've got this guy's a...

tech savvy marathon runner, right?

We've got the busy working mom.

We've got a competitive runner and running club leader.

That's a pretty cool one.

And you go into these and then you can continue to have a conversation and you pitch the app, you see what they like, what they don't.

And these things are pretty good and they really do a good job of simulating what this persona can actually look like.

So that's a quick example of, you know, an agent you can use right now putting one of these things to test.

But the persona, you don't have to use our persona simulator, right?

You don't have to use our screen test agent.

I've done this in Claude as well.

Anytime where you're defining who you want the LLM to act as, and then you bounce ideas off of it or throw it the pitch and see what it comes back with.

Yeah, and we threw out the boss idea, right?

Like, which seems obvious, but also seems a little creepy.

Your boss probably doesn't want you to necessarily have an avatar of them.

But I had a boss a long time tell me, don't come to me with problems, come to me with solutions.

right?

And you might not know the right solution, but if you could kind of bounce off of them, like how are they going to react to this certain thing?

And once again, this comes to the like, you're onboarding your boss into kind of one of these models that get some ideas.

So, you know, if you have things like a DISC persona or a Myers-Briggs understanding of them, or you could share their LinkedIn information or other content they put out there, it's pretty good at picking up context signals to like say, all right, this is kind of what this person would say or do.

And there's infinite examples of this for training purposes, for understanding conversations and practice and stuff.

You could even take the transcripts from your last five one-on-ones and load those in there and have it create a voice doc or a likely persona.

So now, of course, yes, it's a good preface to going in or practice for pitching, especially people who really want to be

prepared and feel very solid in their presentations before they go and pitch something.

This is a great idea.

I'm more of an off-the-cuff kind of person.

I like to be prepared, but I don't have to be word perfect.

Allows for a little improvisation.

But having the rough idea of what are the likely questions that are going to come back just allows you to be more prepared when you're going into these things.

Yeah.

You're going to be much better prepared to handle objections and you're going to make sure that you have stronger alignment, right?

Like that's the goal here.

And I love Wynter, W-Y-N-T-E-R.

They are great for, you put new landing pages or new product ideas, new homepages to the test.

They get it in front of people and they get you results back really quickly.

great product if you want to use real people.

This is another alternative to being able to test some of those solutions and some of those ideas before you have to actually ship it out to real people.

And then finally, the 7th one and the last one, just ask.

Once again, this is one that seems super obvious, but it's amazing how many people skip over it.

We've kind of talked about it here a little bit already.

You don't know what you don't know, so just ask it as you're having a conversation.

What am I missing here?

How would you improve this?

What questions should I be thinking about or where are the weak spots in this plan?

Be critical.

Another one that I love is what data am I missing?

If I've started to do some analysis of a project or how a particular initiative is going, and I'll take it to that to do some, I use Cursor, I use Claude, as well as Agents, of course, but I do a lot of my data analysis in Cursor, and then I'll take it and then I'll go have the conversation with it in Claude, and I'm able to say it's

like, here's what I'm doing.

Here's my strategy doc.

Here's what I think we want to do.

But first, before we get into next steps or tactical things, tactical plans, take a look.

This is the data I have.

What data am I missing?

And it will, it does a great job of saying, it's like, oh, you should look at this or you should cut, slice it another way, or you've got a 90 day view, you have a 180 day view.

So, asking it what is missing, asking it what you could do.

Yeah.

It's one of those, if you're stuck, don't know what to do, just ask.

You think we'd be better at it because we've gotten really good at like, just go into search, go into Google and ask.

And Google's kind of forcing us on that, some of it, because it's now, it gives you little AI answers and like, you could go deeper.

But just remember, as you work with these LLMs, you can do the same thing.

And just ask those obvious things and take that double click to go a little bit deeper.

So that's it.

That's 7 ways that you can think and use AI as a thought partner.

What did we miss?

What are the ways that you guys use them?

We'd love to hear those.

Leave comments.

Reach out to us.

And put a comment down below telling us how you like to use AI as a thought partner.

Yeah, And you know, we'd love to know if any of these are things that you're now doing because you've heard this.

So share all that with us.

So definitely feel free to leave comments below.

We love interacting with our viewers on YouTube here.

But you can also take us on the go.

You don't always have to watch while you're listening.

So we are available as an audio podcast on all the major podcast audio networks.

Yeah, like this.

Like thumb it up, guys.

Help us out a little bit.

Give us five-star reviews.

Give us six-star reviews if you're really feeling good about it.

Minimum of five stars.

At least five stars.

Minimum.

Minimum.

Yeah.

And until next time, keep learning, keep trying new stuff, and keep prompting away, guys.

Take care, everybody.