#14: What's Left for Us if AI Can Do It All?

Hey, guys, welcome back to the Junction.

I've got Mel here and we have a hot topic today.

A lot of the news is doom and gloom, but

we want to talk about the positive side and today's

title, we're going to be talking about what's left for

us if AI can do it all, you know?

Think that's kind of depressing inherently the title.

Well, when I think about that title, I take the

optimistic route and I'm like, if AI can do everything

for me, then the one thing that it can't do

is sit on the beach and drink a beer, right?

Bet you a billion bucks that AI

ain't going to be ever drinking beer.

What about you?

That's not the route my head went, but I like it.

Cheers to that.

What's left for us?

So that's where you envision yourself if

all of your work is automated.

Yeah, I mean, for the most part, right.

But I got to sleep.

Is it making money for you?

Is the AI, have you developed some workforce?

Well, that's what everybody's talking about, right?

Is this idea of an assistant, right.

Not somebody that does it for you, but

that can assist you in doing these things.

But I tell you what, I don't see

it ever assisting me brush my teeth.

There's a lot of real world kind of physical.

I have to use my hands to make this happen.

So there's going to be a lot

of that that we continue to do.

Like all the trades, right?

Like I'm still going to be replacing

the toilet parts with my own hands.

We are on a bathroom theme right now.

Tell you what, let's do this.

Let's go back and forth and see how many different

let's take it to like a profession or something that

we don't think can be automated by AI.

I'll start.

All right, you go.

A personal trainer. That's not fair.

You talked about that earlier today.

What about a car mechanic?

I mean, that one's really easy.

What about a river rafting guide?

Okay.

Airline pilot.

Oh, no, airplanes are going to fly themselves.

Do you think so?

Yeah, I mean, they do.

Your car drives itself.

Yeah, but will we ever replace the airline pilot?

I mean, I hope not.

I love when Jim gets on the

intercom and says, hello, welcome, everybody.

So great to have you today.

Blue skies.

People are going to write in about that voice

and be like, mel, bring that voice back.

It's so good.

Okay, so yours is airline pilot let me go.

I was going to say yoga instructor, but that's

in the same vein as a personal trainer.

What about a fly fishing know?

I think if that person is showing me how to move

my hands and in the river you're going to bring an

AI bot into the beautiful remote rivers of Montana.

I can have my AirPods in and it can

be coaching me based on my watch movement.

That it's interpreting.

Just like they have the golf

swing, they're ruining all the fun.

All right, is that where this game falls apart?

You tell me that all of these things.

So I will say, yeah, people can use these tools to

go write them a workout program or to ask how give

me a program for doing yoga or fixing my vehicle.

But replacing the human.

There's some sort of accountability there when you know that

every day Kyle expects you to be at the gym

at 530 in the morning, and when you don't show

up the next time you do, he's going to kind

of be razzing you a little bit.

That's if you go back to the gym, bell.

I just never go back. I can't fix that. I don't know.

The automation, I guess you could create one

of those automated reminders that just pings you

and you forget you tell it.

You don't know how to jailbreak it, and it

just pings until you finally do the thing.

But there's so many jobs and people

that this is not going to replace.

Well, you're hitting on accountability, right?

Like, the AI is going to say,

did you go to the gym today?

And you're going to be like, yep, but not really.

And you're just going to keep

like but your wearable doesn't lie. Okay.

Are you one of those do you like, wave your arm

up in the air to get your steps when I walk

by your office and I see you fist pumping?

Is that what you're doing?

You're not that's what I'm jamming to

our intro music because it's so good.

It is so good.

But I used to do that when the Apple Watch

first came out, you could trick it into thinking that

you were cycling, but you were actually in your car.

Yeah, if you just kind of, like, moved your

fists in a circular motion and you kind of

went like 5 miles an hour in your car.

But when you go to sleep at night,

you know, I just don't think about it.

So I think that I've been thinking about this a lot.

Humans with AI do have the

potential to replace humans without AI.

So we've talked a lot about it's not going away.

That's our take.

I don't think it's going to go anywhere.

I mean, everybody keeps asking, is

this a flash in the pan?

We had some discussion about

this internally the other day.

Actually, our founder posted about it on LinkedIn.

What was the last AI craze and sort of

solicited some ideas around what was the last technology

Craze that you remember and did it flop?

So I brought up the QR code when it first came out.

Total flop. I hated it.

Total flop.

As a marketer was putting it on all sorts

of we were doing conferences and go download the

agenda, but you needed like, a specific app.

And if you were like on an iPhone versus

an Android, you could not just point your phone

at it and take it to a link. Right.

The greatest comeback ever. Totally. Yeah.

So that was one of those technologies.

Now, the QR code I'm not suggesting

is as impactful as AI, and certainly

QR, that's just a more specific technology.

AI really encompasses a lot of things, but if you're

not understanding what it's doing in your industry, even if

you think, well, I don't know, I'm in construction, and

there's no way they're going to replace the way that

we go out and build a home.

If you're not understanding how other construction companies are

maybe utilizing AI to consistently update and enhance their

pricing models or using it for billing and materials

and time and project tracking, you could be missing

out on a competitive advantage. Right?

Oh, totally.

You reminded me of something I was looking into.

How Amazon fulfillment by Amazon.

I call it FBA works.

And it's this idea that you go and find

hopefully if anybody's watching or reading the, you can

email me and tell me how wrong I am.

But the general idea is go find a cheaper product or

a cheap product that you can sell on Amazon for more.

Right.

Basically, Arbitrage, I want to make money between

$10, I'm going to buy it for ten,

I'm going to sell it for 15.

As I was thinking through how I could speed up just,

I don't know, riffing in my head, well, then I'd use

an API to scrape some websites and get the prices.

And then once I got that figured out, I'd

probably figure out how do I rope in AI

to help me find the cheaper prices faster?

It's not a race to the bottom.

It's a race to how fast you can do it.

And you just made me, like, ding. Yeah.

If you're not using AI because

you don't want to, that's fine.

If you don't know how to use AI and you're

not trying to figure it out, you're digging yourself a

hole because your compadres, your colleagues, they're utilizing it.

They don't really know what to use it for yet,

but some point down the road, they're going to figure

it out, and then they're going to blast off.

And you're still going to be sitting on

planet Earth while people are over on Mars. Yeah.

Operating business as usual.

And you'll be asking yourself, how

did they experience so much growth?

How is Chase sitting on a beach drinking

a beer right now, and he's not working?

Well, let me tell you, friend, he has

this army of AI agents or assistants right.

You were telling me about let's let's

talk about this a little bit.

Bard, is it Bard's assistant or there's a couple

different ones out in the bard has been kind

of they've been coming out of various features.

Watson X has hit the news recently.

They've got this thing called

it's Watson x Orchestrate.

They got Watson X assistant.

But it's this idea that you're doing

these kind of natural language things, right?

Like, I just got off a phone call.

I need to send an email with a summary and

I need to send more times for us to get

back on the like the AI could totally do that.

Of course you got to build

out some of these things, right?

You have to have the technology.

Now you do, and now you do.

But with some of the new product enhancements, I'm sure we'll

get to a place where a startup or a tool, somebody

out there is probably doing this thing right now.

You're thinking through how you would build it,

because that's also just kind of fundamentally, I

think, how you're wired in the role that

you're in and have been in here.

You're thinking about how do I

solution and architect this thing?

But for me, I would no clue where to start to go.

Do some of the things that

you mentioned about scraping websites.

Well, so they have actually a really great example on

the front page of Watson X, it talks about Orchestrate.

And it's this idea that you start out

with some kind of end goal in mind.

And in this I'm just pulling up the front page.

So if you're watching this,

hopefully this is still there.

But the idea is that I want to pull up or I want to

go find a bunch of candidates for a job I need to fill.

Okay, so let's back that up.

Well, first I need to maybe email some folks

and ask some folks some questions, maybe colleagues.

What kind of skills are we looking for, Mel?

Like, can they do this?

Do they need to be able to do that?

Maybe I need to grab some documentation to grab

the job description and put it up somewhere.

Like all these things we are just naturally doing

because we're like, well, we should do that next,

and then we should do that next.

And the idea behind Orchestrate is that it

kind of figures that out on the fly.

And all you tell it is, hey, I need

twelve candidates that can fill this senior salesforce consulting

role and it thinks through and does all of

the things for you that needs to happen.

And then you just start getting candidates.

Is it doing that based off of

data that's already available to it?

Like it's reading data in your existing systems? Yes.

So that's what's kind of unique about where

Orchestrate is, is that you do have to

connect it to your data sources. Right.

In this image, it's got it connected to

imagine a Microsoft database and there's a box

logo and workdays on here as well.

But that's kind of the power of giving it access to

all the data sources that you right is that it can

then figure out, well, the job description is on google Drive.

So let me go search for

senior salesforce consultant job description.

Well, I found one, right?

And then it probably would, I imagine,

would say, hey, Mel, is this the

right job description that we're looking for?

And you're going to be like,

nope, actually it's this one.

Well, you're like conversing, right?

It's basically like this really high powered assistant that

has access to all the same data that you

do and you're just kind of guiding it along

the way so it goes know this idea of

right now, everybody's kind of playing with Chat, right?

How do I summarize the transcript?

How do I do this, how do I do that?

And Watson and Bard, they're kind of looking down the

road in terms of, well, what if I could take

the idea that I need twelve applications for this role,

and then let's break that down into all the pieces.

That we need to do.

And then let's get to that

front prompt of Mel typing in.

Hey, I need twelve applications from great

applicants or candidates that are interested in

a senior salesforce consulting job.

That's the prompt, right?

And if you try to do that now in

Chat GPT, it's going to be like, great, here's

all the things you need to go do. Right.

But Watson X, at least what they're promising is

that it'll go connect to all your data sources,

figure that out, post the job somewhere, get applications.

I bet it would even screen them, ask some basic

questions, and then if it's really smart, it's going to

find time on your calendar to connect them with you.

And then all of a sudden you just got

meeting booked tomorrow to talk to the first.

That that is empowering.

I saw something the other day, someone

said, wow, IBM's really late to the

know with this announcing their stuff.

But we mean, they're a billion dollar company

and they have a huge existing customer base.

So to the extent that it's not like they're

going to the market to gain adoption, right?

They have an existing customer base that they can

now essentially and probably have been already have beta.

Well, I think of IBM, like Apple, they're

always kind of last to the know.

They've got plenty of compute power,

they've got plenty of smart people.

And the way that I see it is they've just kind of

been standing back a little bit and waiting for people to be

like, oh yeah, what if we could do this right?

And the folks over at IBM were like,

that's right, let's go knock that out.

And not to underplay their cards, I mean,

they've had Watson for a long time, right?

I was going to say or they've already been doing it.

Well, yeah, because Watson won Jeopardy in

a famous battle against that guy.

And I apologize if you're watching

this, I don't remember your name.

High hopes for our subscriber count.

Everybody's going to be watching this mel.

But yeah, they had a leg up

then OpenAI and all this stuff happened.

They've been doing it for a long time.

And so I think this was just the natural next

steps for them is to build out something that kind

of leapfrogs, like just the basic chat stuff.

So what is the natural next step for you?

If you have this tool, you have the capability to

leverage an assistant that's going to free up time.

To do what?

This is something that I think a lot

about, mainly because my calendar is booked.

I had eight back to back meetings

that were 30 minutes long on Monday.

And praise the Lord that two of them got canceled

because I didn't know I was going to do. You're welcome.

I canceled one of those. That was you.

Thank you.

I still got some of those emails out.

I sent it to my proverbial assistant, or

I would have if I had orchestrate.

But in my mind that's kind of the idea

of like, I have a meeting and whether I'm

in a sales role, an architect role, an HR.

No, I mean last night to like three in

the morning, I'm laying awake thinking about the couple

emails that I should have sent even last week.

Well, I did that on Saturday.

I told a prospect and one of our great partners that, hey,

I'm just going to use the rest of our time to send

out this email a summary of what we talked about and I'll

send you more times for us to connect next week.

Well, Thursday went by.

Friday I remembered that I didn't send it.

And I was like, oh shoot, got to send that out.

Well, something else pops up that's more important.

And then Saturday you don't have to tell that.

I was going to say you don't have

to keep, well, Saturday rolls around and I'm

like, I really got to send this.

But just imagine if every time you needed something

to happen, you could reach out to your assistant

and your assistant goes and does it?

I think that's why a lot of executives

are a lot more productive than maybe their

counterparts that are director level because they have

access to three or four assistants. True.

And they can send out those emails and they

can block that invite and they can do all

these extra things to give you more time back.

And I would wager, and I don't have an assistant,

but I would wager that if people knew the power

of the assistant and they had access to not one,

but hundreds of them, you're just going to now be

way more productive than you were before.

What I'm also thinking about.

And maybe there's some tools already doing this, but the

power in looking back at that data to say so

let's say you look at your calendar for the last

six months or twelve months and you had a way

of discerning how much of your time was spent on

the phone with customers versus partners versus internal meetings.

I think some of these calendaring tools already try to

do a level of this, maybe using, like, domain names.

So all the meetings that have@ventechnology.com at the

end are going to say, well, you spend

50% of your time in internal meetings.

It'd be really great to then have some

rules or automation set up through this AI

that says it's essentially running interference.

Every really wonderful executive assistant I've known

that's been really good at their job

essentially protects the executive calendar. Right.

They're protecting their executive from do they really

need to be in on these conversations and

they're keeping their calendar blocking out the time

that they need to do the things.

But that would be wonderful to have

something that essentially could field or interference

on things or flag to you.

You've spent X number of time and your threshold.

You said this year you set out your goal was to

go see spend 50% of your time in front of prospects.

Yeah.

And what if it could flag to you?

Oh, my gosh, you're like behind Target.

You're not on plan because you know that every time you

go see a customer, that leads to a new opportunity.

So now you're kind of like, maybe some of

the it might not be a leading indicator, but

it's something that can flag to not otherwise.

You're basically looking back at your calendar.

You're pulling out your American Airlines app.

You're looking at like, wait, how many trips did I

take to I mean, it's the power of the assistant

that most folks don't have access to, I would imagine.

We don't really have access to that.

But when you realize that those are some things that these

people are doing for you, and now you can do that

with orchestrate or with some of these other things that other

folks are coming out with, it's really in.

You and I were in California last week with

some of the team, and Brent and I went

to dinner with one of our customers, and they

wanted to go to a Korean barbecue place.

We got there, we didn't have reservations.

They were like, hey, by the way, it's

about 2 hours before you can sit down.

And we're like, we're kind of starving.

Well, I pulled up Google Maps, and I'm

trying to find Korean barbecue places because that's

where our customer wanted to go.

Well, none of them had open table.

None of them had, like, the quickbook, but it

did have these links to have Google call the

restaurant and talk to that receptionist and say, hey,

I need a table for five 4730.

Do you have any room for us? What's?

The computer talking to the person, and the

person doesn't know that it's computer, or at

least from what I remember from the demos,

it may or may not identify itself. I can't remember.

Sure.

But this was from 2018, and I think Google has

had this and at least in the San Francisco area

for a while, and that's really not all that new.

But it made me realize, like, well, I actually

just had kind of like an assistant reach out

and schedule this reservation for me, and I didn't

have to do anything, and I saved probably, I

don't know, 30 minutes of trying to call every

different place and try to get a reservation.

Like, it texted me back two minutes later,

hey, I got a reservation for you for

five at this place at this time.

And I was like, wow, that was awesome.

That's from 2018.

What if you could do that now with

all these other things that we're asking ourselves?

This is timely.

Today on LinkedIn, I saw a post in this marketing

AI group that I'm a part of, and somebody's soliciting

we're trying to share a feature of their product, but

they call attention to listen to our human, like, AI.

And there's a recording of basically an AI setting an

appointment with a prospect, like, booking a meeting, and it

says it can call all of your leads at the

same time and set 200% more appointments than a human

can for a fraction of the price.

So, of course I'm listening to it, like, trying to scrutinize

the AI voice, like, Nah, that sounds like a bot.

And just when I think I'm

like, no, that's not believable.

It almost does this, like, okay, yeah,

I can put you down for that.

It's even kind of like the inflection.

It's thinking you think about, like, a

prompter or system that you go to.

They don't generally have those

nuances that a human does.

And I was pretty almost, like, freaked out. Yeah.

You know what's really interesting is when

it's talking and it's, oh, yeah, those

little nuances that make it more human.

Yeah, we know what's happening in that moment.

It's the call from one system to the another.

The API calls going out, and it's the latency, which

is why they needed to add those in there. Filler.

Yeah, because it's like, hey, what do you want today?

Right?

And then the AI is, like, having

to process, what do you want today? Right?

And it's thinking, oh, I need to

respond to what do I want today?

And that time between the call out and the call

back in is enough, basically almost enough time for you

to be like, oh, yeah, this is I don't know.

You'll notice this if you see more of those that's

the call out and the call back in, because they

can't get the latency low enough, at least yet, to

where it's instantaneous, like, hey, what do you want today?

Oh, I'd love to have there's

that whatever in between middle thing.

Okay, I have to ask you, I

want to move on to headlines.

Let's talk about this reddit post that I shared

with you, chat GPT ruined me as a programmer.

Yeah, I love this post on Reddit.

It came out, I don't know, 2020 a day ago.

And the premise is like, this guy that's trying

to learn how to program got stuck on something

and I've done this already like a million times,

but he pulls up Chat GPT and basically asks,

how do I get around this problem?

And Chat GPT quickly solves it for him.

And that's not really the surprising thing anymore.

That was really cool, like six months ago.

But what's really interesting is

the next question he asked.

He says, what is the point of me

doing this and then stop learning encoding?

That's where the really interesting

questions start to come in.

Because going back to me on

the beach with the beer, right?

If we lose this aspect of learning and exploring and

being excited about life, we're just going to result in

whatever chat GPT go attend my meetings for me.

Like Google Bard. Yeah, it's depressing. We'll do.

How do you combat that?

I'm telling you right now, if you and I have

a meeting scheduled over zoom and you send your AI

assistant, I will to have a meeting with me.

Not do that, except for a couple of times and just to

try it out purely for podcast research, I will allow it.

No, I think it's a very depressing, honestly, sentiment

when you kind of think of it that way.

Like, what's the point?

I think about my domain in marketing.

I haven't actually had that feeling yet with any

of these tools I've shared that I was apprehensive

in the beginning because I just didn't really understand

the implications and the whole will it replace me?

And since getting my hands on them and

using them and just trying to understand how

other people in other industries are utilizing these

tools, I don't have that outlook.

But I could see especially if you're setting

out to learn a new skill set. Yeah.

Especially if you've been doing it for a long

time and then you go see what Chat GPT

or some other tool can spin out.

I understand the sentiment. Yeah.

I think there's a difference in learning something new.

And I think about pool.

When I was learning to play pool, my uncle

taught me about the cheater really early on and

I was like, well, it's called a cheater, so

why don't I just use it all the time?

And he had a really insightful thing.

The cheater doesn't help you cheat to play perfectly.

The cheater helps you learn how you should hold the

cue, the pull stick and hit the cue ball.

I think I'm getting that right.

The cue ball in the way that will push the

other ball in the direction that you want to go.

And so I think about it like this, chat

GPT is the cheater and it helps you get

past that one blocker, but then you have to

be restrained enough if you're in a learning mode.

Well, don't ask chat, GPT, everything, but ask

it to review what you did if you're

going to use it as a cheater, right?

I think about when Google first came out, in my mind,

I was sitting in the library in my high school, and

I'm like, Google just popped up, and I'm like, what's?

Two plus know something really simple and it's like,

four, okay, what's the square root of it?

Starts answering every question I

could ever come up with.

So I'm like, well, I don't need to memorize

any of like, I'm going to geography class.

I don't need to know what the capital of Egypt is.

I'm just going to Google everything.

And it's that same kind of mentality.

Like, if you're in a learning mode and

you're just going to use the cheater, well,

then just use the cheater all the time.

But if you really want to learn,

use it to review what you're doing.

So I don't know where we're going with

that, but that's what that reminded me of.

Do you see when we think about particularly the

sentiment, this ruined me as a programmer, have you

had that experience in using these tools?

Because you have used it in

a much more technical capacity.

I think for people that are just first learning how

to code, you're stepping off on the wrong foot because

it's basically like, well, do this for me, right?

Write this code for me.

And part of learning coding is actually writing the

code yourself because you're going to get it wrong.

How do you identify, like, if you have

never done it before, how can you discern

what is good code versus bad code?

Because that's what I would say.

Yeah, experience.

Experience doing it over and over again.

And people are all over on LinkedIn,

you got to fail a million.

You got to learn how to fail.

If you're not failing fast, you're

failing slow, whatever they say.

But there's a real sentiment to that because

if you're not failing, you're not learning.

And if you're succeeding every time while you're just getting

lucky, right, you're not going to learn how to code

a massive program just right out of the gate.

You have to do it for some of your pet projects.

Your clients have to tell you, this sucks,

and you should have done it this way.

And you have kind of this internalization of,

I should have done it this way, or

all these bugs are popping up.

Now I know how to debug faster, or I

know how a better way to do it.

And it's through that experience that you really learn

how to be a better coder, particularly if you're

rolling out some sort of product to a marketplace

and you're talking about collecting inputs, right?

So collecting inputs from your boss, your customer,

your prospect, your mentors, your family, your.

Friends that's likely going to be

in some sort of qualitative form.

It's not always going to be over text or email.

And so there's still that human assisted AI, right?

And so that's another area where I don't see

AI just being able to fully automate or replace.

I think that day is coming because the

AI will eventually be able to iterate and

think through and have lessons learned from all

of those transcripts that it's digesting.

Because it could have the same type of experience.

Assuming that there's recordings of calls. Right.

It has access to the code.

But I think until you give it everything,

it's not going to get to that point.

It's not going to really truly learn from experience.

The AI isn't going to be able

to understand the very unique intricacies in

our particular market in this particular process.

When I got on that phone with that

particular prospect, that's a really long way away.

So if you're in that kind of I

want to call it a business analyst, right?

But this idea of kind of a consultative

type approach, and everybody in whatever you do

is in some kind of consultative approach.

Even the guys that are selling the cookies, right?

There's a new cookie place.

It's a really good cookie place.

Tumblr, tumbler, crumble. Crumble.

Yes, really good cookies. Right?

Well, they're even consulting you in that moment,

like, well, this cookie is really good because

it has cinnamon and DA DA DA.

And this is why it's $10 a cookie.

They are good cookies. They are really good cookies.

There's some in my fridge right now.

But yeah, I think if you're in a

customer service element in general, so there's always

that, I don't know, the human connection.

That's like kind of a buzzy term, but there's the

people who want a neck to ring, and it doesn't

feel as good to ring a neck of an AI. Right.

So unfortunately, like, having that, when there's a moment

of conflict and you want to give me your

manager, I think that there's still going to be

very much a place for that.

But then also to share and celebrating those

successes, it just doesn't feel as good if

you can't do it with people.

Yeah, I don't know.

I was exchanging some emails with our partner agency

who's redesigning our website, and we were going back

and forth about football scorers, and they were talking

about the Iowa Iowa State game.

I'm talking about Nebraska, Colorado, and

mean, they're talking about how they

were physically going to the game.

And I had gone to the game

last week, and AI cannot replicate that.

That human connection is so pure.

So those are the things that that's

not that's not going away totally.

You know, it is going away right before we wrap up.

This is what's going away, Karen.

She's going away because guess what?

When everything is automated and AI is

taking over everything, they're going to be

like, can I talk to your manager?

And the AI is going to be like, I am the manager.

And she's like, well, let me talk

to the people that own you.

And they're like, I am the owner.

I know you're laughing really hard, but we did this on

the road to a wedding we were going to on Saturday,

and somebody in the car asked something really silly.

Sorry, Karen.

I have a neighbor named Karen. She's awesome.

She's real sweet. Actually, I do too.

I was in ministry with a Karen. She's awesome.

She has three little kids. She's amazing.

She lives in Austin.

Karen, love you. You're awesome. Cool.

But for the the overarching, like that term

that somehow came about and coined this. You know what?

The AI is going to be like, oh, we need a manager.

And the manager is going to come out and

Karen the Karen you know so well, per usual,

this was a very entertaining discussion, Chase.

I look forward to our next session where

we're talking about all things AI and automation.

You guys know this. It's not going away.

We do want to hear from you.

If you like the show, please share it with a friend.

Like subscribe, do all that, maybe give us a rating.

But we also want to know

what questions you have about AI.

So send us an email at thejunction@ventechnology.com.

In the meantime, keep it automated.