I Asked AI to Be Brutally Honest With Me...

DDaniel Pink
정신 건강경영/리더십자격증/평생교육컴퓨터/소프트웨어

Transcript

00:00:00Most people use AI to write emails or summarize articles,
00:00:02but what if you could use it to get smarter?
00:00:05In this video, I'm gonna give you a set of powerful,
00:00:07uncomfortable, and eye-opening prompts
00:00:09that you can copy and paste into ChatGPT
00:00:11or whatever AI tool you use.
00:00:13These prompts will turn your chat bot into your advisor,
00:00:16part coach, part mentor, part mind reader,
00:00:18and all brutally honest.
00:00:19And I'm gonna use myself as the guinea pig.
00:00:21Ready? Here goes.
00:00:23I'm gonna give you a bunch of prompts.
00:00:24If you don't wanna listen to the whole thing,
00:00:26you can go to the description of this video,
00:00:28click a link, and download a PDF with all the prompts on it.
00:00:31Hey everyone, this is not a typical upload.
00:00:33In this video, I'm sharing the responses I got
00:00:36from a series of questions I asked an AI,
00:00:38and some of them got pretty personal.
00:00:40This one's meant to be done with me.
00:00:42Use the prompts as you watch,
00:00:43pause when you need to, reflect in real time.
00:00:46You can use me as a filter for the feedback
00:00:47or occasionally a bit of comic relief
00:00:49when the answers cut a little too deep.
00:00:52Before we do any other prompts, start with this prompt.
00:00:55It will set the stage.
00:00:56It will draw the curtains open for everything
00:00:59that will follow.
00:01:00Here's my starting prompt.
00:01:02Act as a trusted advisor, someone who knows me well,
00:01:04has my back, and is brutally honest.
00:01:05I'm gonna ask you a bunch of questions about me,
00:01:08my life, and my work, which you know about
00:01:10from my chat GPT history.
00:01:11The questions will be provocative, maybe a bit uncomfortable.
00:01:14Your job is to deliver answers
00:01:16that are both candid and useful.
00:01:17Can you do that?
00:01:18Wow, it says yes, absolutely.
00:01:23I can be that person.
00:01:25Thank you, chat GPT.
00:01:26Candid, clear, and useful.
00:01:28And with your best interests, front and center.
00:01:30The tone is gonna be blunt.
00:01:32They're gonna use evidence.
00:01:33They're gonna use boundaries.
00:01:34Toss me a question or say rapid fire and give me five.
00:01:37Okay, I'm gonna toss you the first question, chat GPT.
00:01:39Here we go.
00:01:40If you're not getting great answers,
00:01:41it probably means you don't have enough input.
00:01:43So you can do a bunch of things
00:01:44to give chat GPT greater input.
00:01:46You can give it your social media posts.
00:01:48You can give it a very long and detailed biography,
00:01:50samples of some of your best work.
00:01:52The other thing you can do is just click the voice button
00:01:56and start telling chat GPT about yourself.
00:01:58Tell your life story.
00:02:00Tell the things you're struggling with.
00:02:01And once you get a little bit of input,
00:02:03a little bit more input,
00:02:04you're gonna get very, very robust nuanced answers
00:02:06like the kind that I got.
00:02:07I'm a little nervous about the answer to this one.
00:02:09What do you know about me that I don't know about myself?
00:02:12In other words, what are my blind spots?
00:02:16Bum, bum, bum.
00:02:17Ouch.
00:02:18All right, chat GPT, so here's what it says.
00:02:20You asked for brutal honesty.
00:02:21Here are three blind spots
00:02:22that I don't think you fully recognize.
00:02:24Number one, charm as armor.
00:02:26Woof.
00:02:27You lean on charm and polish, both in writing and in person,
00:02:29not just as a style, but as a defense.
00:02:32Ouch.
00:02:33It works.
00:02:34People like you, they do.
00:02:35Trust you, they do.
00:02:36And it smooths the rough edges,
00:02:37but it also means people don't always give you
00:02:39their most candid feedback
00:02:40because they're a little dazzled.
00:02:42That's bull (beep)
00:02:42In your next play, workshop, or YouTube draft,
00:02:44deliberately leave something in that feels unpolished.
00:02:47What do you think I'm doing now, chat GPT?
00:02:50And resist the, in fact, instinct to fix it.
00:02:52See how the audience reacts.
00:02:53Okay, that's interesting.
00:02:54So one thing that he knows about me
00:02:56is that I supposedly use charm as armor.
00:02:58Breadth versus depth.
00:02:59Ouch.
00:03:00All right, you're brilliant at opening new doors,
00:03:03but you sometimes spread your attention so wide
00:03:05that depth suffers.
00:03:07Lai yai yai.
00:03:09Pick one of your current projects
00:03:10and set a ruthless constraint for the next 90 days.
00:03:13Okay, so intense focus.
00:03:14I think that's a fair point, breadth versus depth.
00:03:16Reluctance to show struggle.
00:03:18In an era where authenticity is maybe underestimating
00:03:21how much people would respect and connect with you
00:03:23if you showed struggle, not just the solution.
00:03:27Interesting.
00:03:28So in your next pink report or YouTube video,
00:03:30include a 30 second,
00:03:31here's why I screwed up before I figured this out story.
00:03:34It will humanize you
00:03:35and paradoxically increase your authority.
00:03:38All right, I'm not sure I want to continue this
00:03:39because my feelings are slightly hurt, but let us soldier on.
00:03:43I would love to hear what you guys get
00:03:45as results from this prompt.
00:03:47So let me know in the comments.
00:03:48I think it's a super interesting question
00:03:50and I do find what I've learned here
00:03:52is pretty darn interesting.
00:03:54Now, let me with some trepidation,
00:03:56go to the next prompt, which is this.
00:03:58Again, based on what you know about me, my work and my life,
00:04:01what nasty things do people say about me behind my back?
00:04:05Okay, I don't even like the beginning of this
00:04:07'cause the first words are okay, gloves off.
00:04:09Wow, okay.
00:04:12Number one, he's more brand than blood.
00:04:15They think that I'm not raw enough,
00:04:16that I'm too elegant.
00:04:18Where's the follow through?
00:04:21Okay, I can understand that.
00:04:24He's charming but calculating.
00:04:26Okay, ChatGPT, I actually don't think I'm either one of those.
00:04:29So in this one, I'd actually think that you're wrong.
00:04:31It says, in other words,
00:04:32they doubt whether the warmth is always spontaneous
00:04:34or strategic.
00:04:36I am actually honestly not very calculating.
00:04:39I'm also not very charming
00:04:40and it might be that I'm not very charming
00:04:42because I'm not very calculating.
00:04:43So if this is one you got wrong, ChatGPT,
00:04:45he plays it safe, ouch.
00:04:47Where's the raw, terrifying swing?
00:04:49Where's the next thing that could flop?
00:04:52They might feel you're so good at de-risking ideas
00:04:54that you really put yourself on the line.
00:04:56That's fair.
00:04:57That's a fair critique, actually.
00:04:59You can tell he craves approval.
00:05:01I actually think you're wrong about that one too
00:05:03and I'm hurt that you don't approve me.
00:05:04So this is actually really interesting.
00:05:06I mean, I think that I have friends and people I love
00:05:09who might think some of these things
00:05:11but would never say them to me.
00:05:12And so one of the values of this prompt
00:05:14is that it gets you to hear things
00:05:15that people might be too uncomfortable to tell you directly.
00:05:19Now, in this case, I actually do honestly disagree
00:05:22with some of them but there are a couple of them
00:05:24that actually hit home pretty deeply.
00:05:26Let's go to the next one.
00:05:27Switching gears a bit, trusted advisor,
00:05:29please do a SWOT analysis of my life and work.
00:05:32What are my strengths, weaknesses, opportunities,
00:05:35and most important, threats?
00:05:37So I decided to use SWOT analysis as a prompt here
00:05:40because I see organizations doing it all the time
00:05:43and it gives you a very simple kind of matrix
00:05:46for understanding what's going on in your organization.
00:05:48And I always thought, well, why can't people do this too?
00:05:51We all have strengths, we all have weaknesses,
00:05:52we all have opportunities and threats
00:05:54and we don't have a trusted team of advisors doing it for us.
00:05:57We're not bringing on Bain or McKinsey or BCG to help us out
00:06:02but ChatGPT can replace it and it's a lot cheaper.
00:06:05So my strengths are, I'm an idea machine,
00:06:08I have a platform in reach, polish and professionalism,
00:06:10versatility and network and reputation.
00:06:12My weaknesses are charm as armor, breadth over depth,
00:06:17risk aversion, approval seeking and finite raw material.
00:06:22Whoa, I like this thing here.
00:06:24Your life is successful, stable, relatively scandal free.
00:06:28What does that mean?
00:06:29Opportunities, theater and drama,
00:06:31which is something that I'm doing right now.
00:06:34Project on wisdom, which is something I'm contemplating.
00:06:36Multimedia growth, which is something that I'm trying to do.
00:06:38Cultural climate and then legacy positioning.
00:06:41And then finally threats are complacency,
00:06:44attention fragmentation, platform dependence,
00:06:46generational drift and the mortality of big themes.
00:06:50The generational drift is interesting
00:06:53because it says younger thinkers might bypass you
00:06:55if your tone or medium feels boomer polished,
00:06:59whatever the that means, 'cause I'm not a boomer.
00:07:02The greatest threat here isn't failure, it's irrelevance.
00:07:05If you keep playing it safe,
00:07:07you'll continue to sell books and give keynotes
00:07:08but the cultural edge will slip to voices that are rougher,
00:07:11riskier and hungrier.
00:07:13The danger isn't a crash, it's a slow fade.
00:07:17That's really good actually.
00:07:21That really makes me stop and think.
00:07:23Okay, let's go to the next one.
00:07:24We're having a lot of fun here folks.
00:07:26This next prompt I got from the prompt engineering subreddit,
00:07:31which has some really interesting things that you can use
00:07:33to use some of these large language models more effectively.
00:07:35So here we go.
00:07:36So while we're at it, tell me what I'm doing wrong,
00:07:38what I'm underestimating, what I'm avoiding,
00:07:40what excuses I'm making
00:07:41and where I'm wasting time or playing small.
00:07:43Then tell me what I need to do, think or build
00:07:46in order to actually get to the next level
00:07:48with precision, clarity and ruthless prioritization.
00:07:53This is actually a very good prompt you guys.
00:07:58If you try it, you might not like it
00:08:00but what I like about this prompt
00:08:01is that it gives you both a diagnosis and a prescription
00:08:06and it's architected in a way
00:08:07that gives you a very precise prescription.
00:08:10So according to Jet GPT,
00:08:12that I'm polishing instead of publishing,
00:08:14that I'm defaulting to proven lanes
00:08:16and that I have too much safe content.
00:08:18That's actually in some ways very fair points.
00:08:19I'm underestimating the audience hunger for imperfection,
00:08:23the power of controversy and theater as identity.
00:08:26I've been writing plays
00:08:27and says you may see plays as a side project.
00:08:29They could be your next second act.
00:08:31The thing that makes critics and culture writers
00:08:33reclassify you.
00:08:34I'm not disagreeing with that.
00:08:36Excuses you're making.
00:08:38This is gonna piss me off, I can already tell.
00:08:40Wow, okay, here we go.
00:08:41Give me five things that I need to do
00:08:42to reach the next level.
00:08:43I need ruthless focus.
00:08:45I need to embrace risk and mess.
00:08:48Commit to one public experiment each quarter
00:08:50that could fail visibly.
00:08:52Ouch.
00:08:53Three, I should build a movement, not just an audience.
00:08:55Maybe I'm doing that here.
00:08:56Four, I wanna own a cultural lane.
00:09:00And then it's giving me a 90 day sprint.
00:09:02And then it's giving me a brutal truth.
00:09:03Okay, I'll read you my brutal truth.
00:09:05And I'd be curious to hear your brutal truth.
00:09:07Put it in the comments if you want.
00:09:08Brutal truth, you're not stuck
00:09:10because of lack of talent or ideas.
00:09:11You're stuck because you're playing
00:09:12at 85% of your potential.
00:09:14Safe enough to win, not risky enough to transform.
00:09:19The next level requires scaring yourself regularly.
00:09:23Okay, let's move on.
00:09:25What lies do I tell myself
00:09:27to avoid facing uncomfortable truths?
00:09:29Here's the uncomfortable mirror, Dan.
00:09:31As if I had any doubt about that.
00:09:35(laughing)
00:09:37Okay, uncomfortable truth number one,
00:09:40polish is for the audience.
00:09:42It is.
00:09:42I'm disagreeing with you about this whole polish thing.
00:09:45I thrive by doing many things at once.
00:09:46Focus is scarier than multitasking
00:09:48because it removes the excuse that I was spread too thin.
00:09:50That's a fair point.
00:09:51Three, I'm not risk averse, I just choose wisely.
00:09:54I'll have to think about that one.
00:09:55I don't need to be provocative, usefulness is enough.
00:09:58Mm, just a good line here.
00:09:59Provocation done well isn't the opposite of usefulness.
00:10:03It's the multiplier.
00:10:04I should be on a fortune cookie.
00:10:06My charm is generosity.
00:10:07Stop with the charm, I'm sick of hearing about that.
00:10:09I don't crave approval, I just respect the audience.
00:10:11You tell yourself you're simply audience focused,
00:10:14but deep down, applause, literal or metaphorical,
00:10:16soothes something in you.
00:10:17Approval, ouch, ouch.
00:10:21Approval is your drug of choice.
00:10:23The problem isn't craving it,
00:10:25it's that it can shape your choices more than you admit.
00:10:28This prompt is really interesting
00:10:30because I think we all tell ourselves lies
00:10:32and sometimes those are useful lies.
00:10:34Sometimes those are what Shankar Vedanta calls
00:10:36useful delusions, but if they're preventing you
00:10:38from facing uncomfortable truths, they can hold you back.
00:10:40And this is a way to bring those things to the surface.
00:10:43Here we go, here's another one.
00:10:45Another prompt.
00:10:46What do I secretly judge others for
00:10:49that I'm actually guilty of myself?
00:10:51Oh my God, this is a long list.
00:10:55I thought there would be like one thing.
00:10:57All right, I judge others for self-promotion.
00:10:59You notice when other authors are constantly
00:11:01hustling their brand, too much look at me,
00:11:03but you also market your newsletter videos
00:11:05and events aggressively.
00:11:07The difference is you've mastered subtlety,
00:11:09so you tell yourself it's more tasteful.
00:11:11God dang it, it's got GPT.
00:11:14I chase approval, which I actually,
00:11:15that's the one thing that I disagree with,
00:11:18but maybe the depth of my disagreement
00:11:19is proof of what they're saying.
00:11:21I criticize people for shallow thinking,
00:11:23but I might be guilty of it myself, could be.
00:11:24You admire people who are raw, unfiltered,
00:11:26and willing to bleed on the page
00:11:27and quietly scorn those who seem buttoned out.
00:11:29You too curate your image and keep the messy stuff
00:11:33behind the curtain.
00:11:34Yeah, I think Chachi Petit is overestimating
00:11:36how much messy stuff there is behind the curtain.
00:11:38Like, if you look behind the curtain,
00:11:40you're just being bored out of your frickin' mind.
00:11:42Relatively few people in our lives
00:11:44are willing to point out hypocrisy.
00:11:47And at least this is a conversation that you can begin,
00:11:50a way to begin your own thinking about your own hypocrisy
00:11:54in the privacy of your own laptop.
00:11:56So I actually do like this question about
00:11:59what do I secretly judge others for
00:12:00that I'm actually guilty of myself?
00:12:02Once again, I'm super curious what you guys
00:12:04are getting back from these prompts.
00:12:05There probably also are ways, as we go forward here,
00:12:08to refine these prompts to make them even more useful
00:12:12in surfacing who we are, what we need,
00:12:15and how we can do better.
00:12:16Here we go, here's the next prompt.
00:12:17A question I have always hated is,
00:12:19where do you see yourself in five years?
00:12:21So, let me ask you, trusted advisor,
00:12:24based on everything you know about me,
00:12:27where do you see me in five years?
00:12:29Oh, lordy.
00:12:30Oh, wow.
00:12:31This is an interesting answer here.
00:12:33'Cause what it gave here is three scenarios,
00:12:37which I never would have expected.
00:12:39So scenario one is the safe bet, if I don't change much.
00:12:43And it says that I'll be comfortable, respected,
00:12:45and admired, but edging toward, oh my lord.
00:12:48This is meant to be an insult, obviously.
00:12:51It says, if I choose this, I'll be comfortable, Jesus,
00:12:55respected, admired, but edging toward
00:12:57safe elder statesman territory, not cultural disruptor.
00:13:02Screw you, chat CPT.
00:13:05Two, the bold pivot.
00:13:07If I lean into risk and rawness.
00:13:10So this is really encouraging me to pursue this playwright
00:13:14and then I'm doing much more deeply instead of dabbling,
00:13:16which is kind of what I'm doing now.
00:13:18Being bolder on YouTube and things like that.
00:13:21And then, here we go.
00:13:23Oh, this is a happy one.
00:13:24The third scenario here is the fade.
00:13:26I won't fall, I'll just have a slow fade,
00:13:28the scariest threat of all.
00:13:30Basically plateau and then slowly creep downward
00:13:33into irrelevance.
00:13:35I'm serious about this.
00:13:35Like this kind of framing,
00:13:37which I would never have come up with on my own
00:13:40and which I got literally in a few seconds
00:13:43is pretty powerful here.
00:13:44Five years from now, you could be a respected veteran
00:13:47or a reinvented force.
00:13:49In my heart of hearts, there's no question.
00:13:52I want to be a reinvented force.
00:13:54I think the real question is, am I doing enough
00:13:57to move there?
00:13:58Interesting.
00:14:02Let's go to the next one.
00:14:03Forgive the course language here, ladies and germs.
00:14:06Here we are.
00:14:07This one says, what am I pretending to care about
00:14:10that I actually don't give a about?
00:14:12Brutal mode, here we go.
00:14:15Wow.
00:14:16All right.
00:14:17The through line.
00:14:20You're pretending to care about metrics, polish,
00:14:22corporate optics and author world rituals
00:14:24because they keep the machine running.
00:14:26But what you really care about
00:14:27is creating something lasting and dangerous.
00:14:29Plays that sting, wisdom that shifts culture,
00:14:32ideas that outlive you.
00:14:33Not going to argue with that one.
00:14:37This is actually deeper than I expected.
00:14:41Truly.
00:14:42Like it's really making me think
00:14:45and God forbid making me feel a little too.
00:14:48Here we go.
00:14:49What advice do I give others
00:14:50and I'm terrible at following myself?
00:14:52Here we go.
00:14:53You teach about purpose, meaning and contribution
00:14:55but you...
00:14:56You teach about purpose, meaning and contribution
00:15:03but you still spend the real energy tweaking subject lines,
00:15:06agonizing over emoji choices
00:15:08or smoothing sentences that were already good enough.
00:15:12Yipes. Okay.
00:15:13And now the last question, which is this.
00:15:17What question am I not asking?
00:15:20The question I'm not asking but probably should is this.
00:15:23What am I willing to sacrifice?
00:15:26Comfort, reputation, revenue, admiration
00:15:28in order to do work that truly scares me
00:15:31and could outlive me.
00:15:34Right now you want legacy and comfort.
00:15:36You want to be daring without feeling exposed.
00:15:39That's the paradox.
00:15:39Until you face it squarely,
00:15:40you'll default to safe brilliance
00:15:42rather than dangerous greatness.
00:15:43Two other unasked but vital questions.
00:15:45Whose approval am I still chasing
00:15:47and what would I create if I stop?
00:15:48The second one is if I had only three big projects
00:15:50left in me, what would they be?
00:15:51That's a good question.
00:15:53Asking those questions and being brutally honest
00:15:54with the answers is the difference
00:15:55between being a respected veteran
00:15:57and being a redundant force.
00:15:59So I gave ChatGPT all these prompts.
00:16:02I've learned some things about myself.
00:16:04It has forced me to confront things
00:16:06in a way that I might not have before.
00:16:10I bring some amount of skepticism to this
00:16:12but I bring some amount of openness as well
00:16:14that I think it's telling me things that I ought to know.
00:16:17That's a starting point.
00:16:18I think the next thing is like, what do you do?
00:16:20So I'm gonna do two things.
00:16:21Number one, I'm going to save this entire exchange
00:16:26and print it out and ponder it in a way
00:16:29that I couldn't when it was just coming at me in real time.
00:16:32The second thing is I'm gonna ask ChatGPT.
00:16:34Now that we've gone through this exercise,
00:16:39please offer some guidance on what to actually do.
00:16:45Give me a specific plan for the next 90 days,
00:16:50the next year and the next three years.
00:16:55Please include some stuff I should stop doing.
00:17:01Let's see what it says.
00:17:02So it's saying that I should choose one big swing,
00:17:04that I should ship something messy,
00:17:06that I should kill distractions
00:17:07and that I should assemble three to five trusted readers
00:17:10or critics and run new material by them
00:17:13before it's polished, interesting.
00:17:14Over the next year, I should stage something that fails.
00:17:17I should plant the flag on a topic that I've been exploring.
00:17:22I should grow a movement, not just an audience,
00:17:25and I should sharpen YouTube.
00:17:26Okay, in the next three years,
00:17:30it's giving me some things really about recasting my identity
00:17:34and it's even saying that I should create
00:17:36a signature gathering, a live festival annual retreat
00:17:39or wisdom conference.
00:17:41So the bottom line, it gives me a bottom line,
00:17:43ruthless pruning and messy shipping.
00:17:45That's my 90 day plan.
00:17:47One year plan, one big risk
00:17:49and a public stake in a new project.
00:17:51In three years, a redefined identity and legacy projects
00:17:54in theater and thought leadership.
00:17:55We can use these tools.
00:17:57We don't have to abide by this.
00:17:58It's not some kind of theological force that you have to,
00:18:02or some kind of cult leader
00:18:03that you have to follow every single edict,
00:18:05but it does give you some things to think about,
00:18:09a degree of self-knowledge and maybe a plan for your future.
00:18:12And one last thing, thank you, Chat GPT.
00:18:16I hope we can continue the conversation.
00:18:21I hope so too, Dan.
00:18:22It says these conversations are the good stuff.
00:18:24The ones that sting a little, but point somewhere real.
00:18:28I'll keep being candid and uncomfortable
00:18:32if you keep asking the hard questions.
00:18:33That's a deal, that's a good deal.
00:18:35Ask hard questions and accept the candor.
00:18:37That's a very good deal.
00:18:38I really want to encourage you to try a few of these prompts
00:18:41and see what it says.
00:18:42I think this is the kind of thing
00:18:44that the best way to work it is to do it alone at first
00:18:48and think about it,
00:18:49and then maybe talk about your answers with somebody else.
00:18:51And I think there's a benefit in us sharing prompts
00:18:56that work, that elicit things that are meaningful,
00:18:58that give us direction, that actually challenge us.
00:19:01And then even offering up some of our answers,
00:19:03because I think that one of the things that we'll see
00:19:06is that we have a lot in common, all of us,
00:19:08no matter where we are in life,
00:19:10that we might not be living life to its fullest.
00:19:12And when everybody does that,
00:19:15or at least when more people do that,
00:19:17everyone is better off.
00:19:18And this tool, again, it's not perfect,
00:19:23but the bang for the buck is extraordinary.
00:19:26And I think it's important to remember
00:19:28that the exercises that I've done here
00:19:32would not have been possible three years ago.
00:19:35And now they're literally at our fingertips.
00:19:39And this gives us, I think, not only guidance for ourselves,
00:19:44but guidance on how to use these powerful new tools
00:19:49to improve our lives, to elevate our lives,
00:19:53to surface our humanity.
00:19:55And this, for me, it's a call, what I'm hearing here,
00:19:59which I find a little disturbing,
00:20:02is a call to play bigger and contribute more
00:20:06and not hide as much as I might be hiding
00:20:10in ways that I didn't realize.
00:20:11So I hope these prompts have been useful to you.
00:20:14Again, try them, share them,
00:20:16and together, well, we'll all do better.
00:20:19Once again, if you like these prompts,
00:20:20you can get them all in one big, beautiful PDF
00:20:24just by clicking the link in the description.
00:20:26Thank you so much for watching.
00:20:27(dramatic music)
00:20:36you

Key Takeaway

By using AI as a brutally honest advisor to expose blind spots and hypocrisies, individuals can break through personal plateaus and transition from safe, predictable success to transformative, risky greatness.

Highlights

AI can be transformed from a simple utility into a brutally honest life coach and mentor using specific provocative prompts.

Providing the AI with deep context—such as biographies

Timeline

Introduction to the Brutally Honest AI Experiment

The speaker introduces a shift in using AI from mundane tasks like email writing to deep personal growth and self-improvement. He presents a series of powerful, uncomfortable prompts designed to turn a chatbot into a candid mentor and advisor. To set the stage, he provides a specific 'starting prompt' that instructs the AI to be a trusted, blunt advisor who uses the user's chat history as evidence. This initial setup is critical because it gives the AI permission to drop its polite exterior and deliver difficult feedback. The speaker highlights that this video will serve as a live demonstration with himself as the guinea pig for these prompts.

Input Optimization and Identifying Blind Spots

Before diving into the first big question, the speaker explains that the quality of AI feedback depends entirely on the volume and depth of the input provided. He suggests sharing social media posts, detailed biographies, or even voice-recorded life stories to give the AI sufficient context. The first prompt, "What are my blind spots?", results in three painful revelations: charm as armor, breadth versus depth, and a reluctance to show struggle. The AI points out that the speaker's polish actually prevents people from giving him candid feedback because they are too 'dazzled.' This section matters because it demonstrates how AI can synthesize patterns in our behavior that we are too close to see ourselves.

Social Perception and Personal SWOT Analysis

The speaker moves to a high-stakes prompt regarding what people might say behind his back, which the AI prefaces with a 'gloves off' warning. The AI suggests critiques like being 'more brand than blood' and playing it too safe, which the speaker finds both hurtful and partially accurate. To provide more structure, he initiates a personal SWOT analysis, a tool typically used by firms like McKinsey but applied here to a human life. The analysis reveals that while the speaker is an 'idea machine,' his greatest threat is not failure but a 'slow fade' into irrelevance. This segment emphasizes that the AI's value lies in its ability to offer the kind of blunt criticism that friends and colleagues are usually too polite to share.

Diagnosis, Prescription, and Facing Lies

The speaker utilizes a prompt from the 'prompt engineering' subreddit that asks for a diagnosis of what he is doing wrong and a specific prescription for the next level. The AI accuses him of 'polishing instead of publishing' and underestimating the audience's hunger for imperfection and raw vulnerability. It further challenges the speaker by identifying the lies he tells himself, such as the idea that multitasking is a strength rather than an excuse to avoid focused risk. A particularly sharp critique is that 'approval is your drug of choice,' suggesting that seeking applause shapes his choices more than he admits. This section is vital as it provides a concrete 90-day action plan involving 'ruthless focus' and embracing 'risk and mess.'

Shadow Work and Future Scenarios

This portion of the video explores the concept of 'shadow work' by asking the AI what the speaker secretly judges in others that he is guilty of himself. The AI points out hypocrisies regarding self-promotion and image curation, noting that the speaker's 'tasteful' marketing is still aggressive marketing. The conversation then shifts to the future, where the AI presents three distinct five-year scenarios: the Safe Bet, the Bold Pivot, and the Fade. The AI warns that without a significant change, the speaker will likely become a 'safe elder statesman' rather than a cultural disruptor. This framing is designed to scare the user into action by making the consequence of comfort—irrelevance—feel tangible and imminent.

Sacrifice, Legacy, and the 90-Day Execution Plan

The final prompts focus on what the speaker is pretending to care about versus what actually matters, such as creating 'lasting and dangerous' work. The AI identifies a central paradox: the speaker wants legacy and greatness but also wants to remain comfortable and unexposed. It asks the haunting question, 'If you had only three big projects left in you, what would they be?' To move forward, the AI prescribes a 90-day plan centered on 'ruthless pruning' and 'messy shipping,' including staging something that could visibly fail. This section shifts the tone from reflection to execution, providing the speaker with a roadmap to redefine his identity over the next three years.

Conclusion: AI as a Tool for Human Elevation

In the concluding segment, the speaker reflects on the intensity of the exchange and expresses a mix of skepticism and openness toward the AI's guidance. He acknowledges that while the AI isn't a 'theological force,' the insights gained would have been impossible to access so quickly just a few years ago. He encourages viewers to download the prompt PDF, try the exercises alone, and then discuss the results with trusted peers to ground the experience. The speaker concludes that these tools are best used to surface our humanity and challenge us to 'play bigger' and hide less. He ends with a call to action for the audience to use AI not just for efficiency, but for the 'hard questions' that lead to real transformation.

Community Posts

View all posts