AI is everywhere in trucking right now—but most operations are stuck between hype and hesitation.
In this episode of That’s Delivered, Trucking Ray sits down with Evan J. Schwartz, Chief Innovation Officer at AMCS Group, to break down what actually works when you bring AI into real-world logistics—where margins are tight, uptime is critical, and mistakes don’t stay on paper… they hit the street.
This conversation goes beyond buzzwords. We connect AI to real operations—fleet optimization, predictive maintenance, route planning, and the day-to-day reality of drivers and dispatch teams. Evan explains why most “AI hallucinations” aren’t random—they come from unclear instructions, missing context, and poor data. The real game-changer? Not just better models—but better communication, better inputs, and someone accountable for making AI work inside your business.
We also get real about change management. Leading with “cut headcount” is one of the fastest ways to kill adoption. Instead, the smartest companies are using AI to remove repetitive admin work—freeing up people to focus on relationships, service, and decision-making where it actually matters.
Drawing from years of ERP and digital transformation experience, Evan lays out a practical roadmap: avoid hidden spreadsheets, define outcomes that pay for the project, pace the rollout, and never outsource your thinking to a vendor.
If you’re serious about using AI to strengthen your operation—not replace it—this episode is a must-listen.
👇 Key Takeaways
✅ AI fails when tasks are vague—clear instructions and context are everything
✅ “AI hallucinations” are often a data and communication problem, not a model problem
✅ The most important role isn’t the tool—it’s the AI steward guiding it
✅ Leading with job cuts will destroy trust and slow adoption
✅ Start with automating low-value admin work, not high-stakes decisions
✅ Define a few outcomes that pay for the project before scaling
✅ Hidden spreadsheets and undocumented processes will break your AI rollout
✅ Pace the change—teams need time to absorb new systems
✅ AI should enhance human judgment, not replace it
✅ The companies that win will balance speed, data, and trust
Interested in being a guest on the podcast? Click here to join the conversation!
Promo code: TR104U
Get your discount for Truck Parking Club click: here
Need help finding the perfect truck seat? Let us find one for you! Call 269-692-5770, email: sales@seat-specialists.com ,or fill out form and use code TR104U.
Get your Trucking Ray merch to show that you roll with Trucking Ray and That’s Delivered Podcast! click: here
Listen
Watch
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
SPEAKER_00 (00:25):
Welcome back to another show of That's
Delivered.
I'm your host, Trucking Ray, andwe talk with people that are
shaping the future of trucking,logistics, and the industries
that keep America moving.
Today we're diving into AI,digital transformation, and what
actually looks like forreal-world operations, not just
theory.
I'm joined by Evan Schwartz,Chief Innovation Officer of AMCS
(00:50):
Group.
With 35 years of experience,Evan has helped organizations
build AI and data-drivenstrategies into industries like
waste, recycling, and logistics,turning big ideas into real
measurable results.
Evan, welcome to the show.
(01:12):
Living the dream, man.
You?
Yeah, I'm doing well.
So glad to um have you on theshow, talk about what you're
doing.
I'm always excited about theguests that that you know take
the time out of their day tocome on the show and talk about
what they're doing, the changesthat they're making in the
industry and beyond.
So um can't wait to hear more,man.
You know, I know you're outthere traveling and you took the
time out to come on.
(01:32):
So uh what's the week been likeuh getting started?
SPEAKER_01 (01:35):
Oh yeah.
So the week is uh mostlytraveling and it's full of
logistics, right?
The uh air traffic, it's notwhat it used to be.
I I'm old enough to tell you Icould just throw cash down, walk
out on the tarmac, climb on aplane, and be off.
Those days are gone.
It's it's not great these days.
So maybe that's something we canwork on.
Maybe AI can help improve thatlittle strategy.
(01:56):
But I'm in Dublin right now.
Uh we're having an off-sitemeeting.
So I'm speaking to you from thehotel room.
I hope that's okay.
SPEAKER_00 (02:03):
Yeah, that's perfect, man.
So uh let's let's let thelisteners know.
Give us a little bit about yourbackground.
Um, you spent decades out thereacross operations, um,
agriculture, um, I think, andand now innovation.
Um, what's been in the commonthread for your career?
You know, how do you held thoseroles as chief officer,
architect, or COO, now uh was itchief innovation officer?
(02:27):
So how has that perspective ontechnology changed as you've
moved closer to the businessrole?
SPEAKER_01 (02:33):
So, interestingly enough, I started in the gaming
industry as a teenager and Istarted my first company writing
games.
And, you know, I thought I I wasreally attracted to creation,
creating something.
Because I remember when I openedup my PO box and pulled that big
bag of orders for a game I hadbuilt, I was hooked.
I was hooked on business.
Um, later I've refined it andfound that it's not just the
(02:55):
creation, but it's the creationof value.
Seeing how software or digitalsystems improve the lives of
everyone and produce value, I'mreally attracted to that,
particularly around how it getsinto the real world.
Uh, I as I left the gaming worldand went into the first business
venture, which was it started inthe insurance that got me kind
(03:16):
of hooked.
And then I went into what Iwould consider the wealth
industries, where you seetrucking, where you see the
movement of material.
You know, it's the I mybest-selling book, People,
Places and Things, is writtenall about the fact that it
there's people involved.
You got to pick something upfrom one location, move it to
another, buying and sellingthings, right?
And it was in the natural gasindustry, the transportation of
(03:38):
gas around the country, havingto get it sold, how to give it
truck, uh, the trucking with theoil and distribution of gas to
various gas stations introduceda lot of interesting logistics.
That led me in strangely intothe forestry side of the world,
where log trucks had to go outinto the woods and get a whole
load of uh wood.
(03:59):
Some of it was hardwood, some ofit was pulp wood to be able to
make paper.
Um, and again, I was surprisedat the level of complexity built
into those businesses and howhard it was just to be able to
schedule a truck, to be at somepoint in the woods, to pick up a
load of logs to get it to themill, because that mill can't
(04:20):
shut down.
I mean, it's a million dollarsan hour for that thing not to
run.
So it has to be fed.
And then as that industrychanged from virgin fibers into
secondary um, you know, OCC, DLKtype of uh fires that got into
the waste and recycling side ofthe world.
And it was then really youstarted to see all the
(04:42):
connective tissues between allthe industries.
Like even in the forestrybusiness, no part of that tree
would go to waste, right?
The terpenes got extracted outof it during the paper making
and got sold into the chemicalindustry for flavorings and fuel
additives and some of the pulpnon-woven.
There's a sheet of paper behinda flat screen, I'm sure you're
staring at right now to see myugly mug.
(05:03):
All of these industries areconnected and they have waste
components.
And it's trucking, it'slogistics that has to move this
around.
The entire scrap metal industry,which is really a hyper-focused
version of the waste andrecycling side of the business,
will tell you that scrap doesn'tlike to move far.
Every mile you have to take thatand move it down the road, it's
(05:24):
just killing the margins in it.
So there's such a huge focus onthe logistics, the operational
side, keeping trucks in goodworking order, being able to
maintain a fleet of vehicles,making sure that you know what
they're doing, when they'redoing it, every minute they're
not producing is a minute ofwaste opportunity.
So where I really kind of landedwas across that connective
(05:46):
tissue of all of theseindustries at AMCS group.
How can what we've learned inthe forestry business help the
scrap business, help, you know,just general logistics and field
services and predictivemaintenance to make sure that I
can maybe even predict usingalgorithms on whether I need to
take that truck out of service,you know, or I might need to put
(06:07):
that truck up for auction.
I've gotten about all the life Ican get out of that thing, and I
need to go shop for a new truck.
All of those things drive bottomline value.
If you're concerned aboutenvironmental social governance,
making sure that you're having alight touch on the uh the
environment, some of thissoftware, we have 700,000 plus
(06:29):
trucks around the world undermanagement in our software, and
we're shaving 17 gallons ofdiesel a month off of the burn
rate by just optimizing howthose trucks do their work.
unknown (06:43):
Wow.
SPEAKER_01 (06:44):
That's a lot of fuel not being burnt.
That's also a lot of fuel Ididn't have to pay for.
Right?
So we're we're starting to drivebeyond the let's just be
environmentally responsiblebecause it's the right thing to,
oh, you know what?
It also happens to make money,right?
If if I don't have to burn thatfuel and spill that into the
(07:04):
environment, that's money Idon't have to spend.
You know, if I if I know exactlyhow to maintenance my truck
optimally, and I don't have tohave a historical predictive
algorithm going, well, wemaintenance it every six months
before.
Let's keep doing that until itbreaks.
And I can take an oil sample andtell you exactly what's going on
in that engine and what you needto do with it.
(07:25):
Now I can be more responsible inthe way that I'm maintening the
truck.
I'm not taking it out ofservice, I'm making sure that
it's doing its job.
And more importantly, I'm notputting oil into the environment
that has to be cleaned anddisposed.
Everywhere there's a high carbonfootprint in your business is an
opportunity for efficiencies andbottom line impact.
(07:46):
And the only reason why wedidn't run at it like that
before is because we didn't havethe solutions of software or or
the passion to chase it.
And now with AI and some of thetech we've got in this world,
it's easy to do.
But the most thing that bringsme here today, Ray, is the
passion about people, places,and things.
That there is a future of AIwithout people.
(08:08):
Why are we doing it?
I'm trying to make sure I'mgetting the message out that the
future is not happening to us,it happens because of us.
We have a choice, right?
Take ownership, don't be thevictim, own that future.
And the reality is we just spenta year and a half and a whole
ton of money by a lot of smartpeople on AI that didn't work,
(08:30):
didn't succeed.
And they let a lot of goodpeople go with their business.
Sure, this thing was going toreduce headcount, and it was
just the way they did it waswrong, right?
SPEAKER_00 (08:39):
So, what do you say about that?
I mean, AI and real worldoperations, there's a lot of
noise around it, like you'resaying, and the perspective of
how businesses can actuallyleverage AI to modernize
operations or improve efficiencywithout complicating things.
Um, it's got to be a challengein itself, or just uh trying to
package it correctly.
SPEAKER_01 (08:59):
Yeah, it's it's an OCM, organizational change
management.
It's a person problem, it's apeople problem.
The tech is there.
What makes AI useful is theperson who can clearly
communicate to it, give it avery clear instruction, and can
designate the data context itneeds to perform the job that
(09:19):
you just asked it to do.
Where AI goes off the rails isit's like a super smart young
intern that's come in and theyknow a lot, but they don't have
your experience.
They don't have your intuitionyet.
The difference is that internmight be a little self-conscious
and tell you they're not reallysure about this.
AI, on the other hand, superconfident because it knows
(09:41):
everything there is to knowabout your industry and your
business, but it knows nothingabout your business, the way you
want to operate, the way youwork.
So it's so confident knowingeverything, it doesn't come back
and go, yeah, you didn't exactlygive me enough context.
So what I'm gonna do is I'mgonna look down in my super big
brain that knows everything, andI'm gonna get the answer from
(10:02):
that.
And that might not apply to yourbusiness, and that's what we
call today hallucinations.
Every time you see ahallucination from an AI, it's
because the task we asked it wastoo vague, or the data we gave
it to be able to perform itsactivity was incomplete, the
context was incomplete.
SPEAKER_00 (10:20):
And then the input that's a lot of work to get it
right.
SPEAKER_01 (10:24):
Yeah, it is a lot, and then the third leg of what
I'm calling the AI steward, thatperson that sits over top of a
bunch of AIs to get a lot done,has to know what good looks
like.
You don't have to know theanswer, right?
That that's I get some pushbacksometimes.
Well, if I know the answer, whatdo I need AI for?
You don't need to know theanswer, you just need to know
what it gave you doesn't lookright.
(10:44):
And now I'm gonna dig in andjust validate it, or it does
look right.
It that's about where I expectit to be.
Let's move forward.
So, stewards are really good atcommunicating that task, knowing
the data it needs, and knowingwhat good looks like.
We're moving away fromrepetitiveness as a burden of
humanity and into rapiditeration.
(11:07):
I need to be able to do a lotmore.
I don't mind if it messes upbecause the cost of doing that
iteration is so low, I can justfix it, do it again.
And when I get it right, I cando a hundred thousand of those
things where I couldn't comeclose to that level of iteration
without the power of AI, right?
That's right.
SPEAKER_00 (11:26):
Yeah, I mean, that's there's pluses and pros, you
know, that or cons that comewith everything.
So uh there's gotta be you gottalook at both sides of the coin.
I mean, when these companies areworking with uh all this AI and
automation, what's the biggestmistakes you see when they try
to adopt AI or automation?
We just kind of call that out.
SPEAKER_01 (11:44):
Yeah, look, the biggest number one mistake is
thinking that I'm gonna saveheadcount.
I'm gonna put this thing in,it's gonna run this little
section perfectly.
And if I had 10 people, I cannow cut it down to two.
And no, it's not gonna work thatway first.
You when you're applying AI, getyourself a positive outcome
that's gonna justify the cost ofit outside of headcount.
(12:07):
At the end of the journey, Ray,if you want to right size your
business, fine.
But what AI should be doing isasymmetric growth.
I should, with the number ofpeople I've got right now, I
should be able to grow mybusiness not by an extra 20%.
That's that that's a childchange.
I should be able to use thistech to go 3x, 4x, 8x in growth,
(12:29):
right?
And then if I've looked aroundand I've grown as big as I want,
by all means, right size yourbusiness.
I'm not gonna tell you how tonot right size your business.
But if you go in with thatsentiment first, you're gonna
get caught with your pants down.
There is a change managementthat you're gonna go through.
Now, there is gonna be a certainnatural attrition if you decide
(12:50):
to adopt AI.
And maybe the comfort of a lotof your listeners, except for
those who are running their ownbusinesses, is a lot of this is
gonna impact the SGNA side ofyour business, the back office
work.
Those guys that are boots on theground with a blowtorch or a
wrench getting it done today,their life is gonna get much,
much easier, but not changesignificantly for a while.
(13:13):
But the back office is gonnachange significantly.
And today, you can't go out intothe market and hire an AI
steward.
Now, my role at JU, we'rebuilding a curriculum around
getting our first master'sgraduate as an AI steward.
And they have domain domainexpertise.
So you'll have AI stewards thatare good at finance and AI
(13:34):
stewards that are good atlogistics and dispatch, and AI
stewards that are really good atfront office customer service
modules.
So they have breadth ofknowledge, they're clear
communicators, and theyunderstand when I ask the AI to
do something, it's gonna needaccess to this data to achieve
it.
Right.
And then when I can watch it,because I've got metrics and I
know what good looks like.
(13:55):
So now I could manage maybe ahundred of these agents, and
these agents are callingsubagents to be able to do
refined work.
That's gonna be the back office.
And if anything, they're able togrow that business so rapidly
that you're gonna see,especially if your business
deals around logistics, servicevehicles, the truck is tied to
(14:16):
the revenue, like a garbagetruck goes out and does it.
So that's that's where the moneyis.
That's the offer, right?
That's the service.
You're gonna see that grow.
You're gonna be adding 10, 15,20 trucks to keep up with how
quick the business has grown.
That's gonna be one of itsbiggest problems.
But thinking that you're goingto reduce headcount as a first
measure of success almost alwaysends in failure.
(14:40):
Almost always.
SPEAKER_00 (14:41):
Yeah, so we got to change the culture.
I mean, it's a lot of industriesout there that are adopted that,
thinking that hey, that's thebest way to go because maybe
that's how we used to do thingsback in the day is cut back on
on headcount.
But moving forward, you got tokeep that experience.
You don't want to lose thatvalue.
So what you're saying.
SPEAKER_01 (14:57):
Um AI doesn't have yet is your experience.
There's a difference betweenknowledge and experience, big
difference, right?
We got tons of knowledge in abook, but a book can't look at
the the walk around a truck andtell you whether or not that
vehicle is in trouble, you know,whether those tires need to be
changed.
You're gonna have to, there's acertain amount of experience
that you have to know how tolook under things, check your
(15:19):
chains, make sure if you're alog truck, how you've got those
logs situated, what theunbinding station is, how many
trips a day can you make to theto the woods?
That all comes out ofexperience, grading the
material.
If you're the driver and you'reresponsible for the material and
you're just taking a quick lookat the material you picked up,
because your livelihood countson that material getting across
a scale somewhere.
(15:41):
AI is not doing that yet.
SPEAKER_00 (15:43):
And it's not going to do it for a long, long time.
So, I mean, margins are tightout there for trucking
industries.
You got fuel costs, like youmentioned.
Uh, what does AI do?
That uh, you know, where does itcreate the most impact for
immediately for companies?
SPEAKER_01 (16:00):
Yeah, look, that's a great question.
So, from where you see a lot ofit, and it's not the AI that
everyone talks about, the thechatty transformer AI, it's in
predictive algorithmic AI is therouting optimization.
If I'm if I'm doing servicingwithin a city and I'm running
the same routes over and overagain, optimizing the way that
those routes should be run isthe immediate best say.
(16:23):
But where the new transformers,the agentic AI is really adding
value into anything to do withtrucking, long haul, short haul,
logistics, equipment,maintenance, all of that is the
way it's transforming thedriver's existence.
The driver gets up in themorning, let's just say that
you've got a route, you've got2,000 stops, you got a service
today.
You you basically have a morningdebrief, a stand-up, you look at
(16:46):
what your day's work is going tobe.
So that's eating up some time,you get in your truck, and
you're now also the guy that'shaving to collect details on any
exceptions that occur along theroute.
Take all that away.
Get in your truck, do your job.
It'll tell you where you need togo.
You have faith that someone hasworked out all the logistics,
what the servicing is, what youneed to do, make sure that all
(17:09):
if you've got any repair, ifyou're doing servicing, that
your van has all the stuff in itthat it needs to have.
That has all been done for you.
You were to maintain safety, getto the service location, perform
the service, and then get backin your truck and go on to the
next one.
And that becomes the drive is tomaintain safety.
You know, we were talking beforerecording, is that you could go
(17:31):
back eight years.
There was no tech on a truck.
And now they're mobile datacenters.
I mean, they're practicallymobile offices, they've got
their own cell uplink, theirIoT.
You can learn everything thereis to learn about that engine on
the road.
There's cameras all around itrecording full video feeds
constantly for safety, but alsothose are now being able to be
(17:53):
tapped in by AI so that I cansee things.
I can look out at the servicearea.
One of the things that we'relooking at right now is to be
able to tap into the publicfeeds.
Most cities have public feedsfor cameras.
So if I'm a garbage truck andI'm going to maybe a restaurant
to pick up their container andempty it, and his bread truck is
(18:14):
blocking it.
If the cameras in the city cansee that block 15, 20 minutes
ahead of the truck, so now AIcan call that restaurant and
say, You've got a bread truckblock in your container.
My truck's going to be there in20 minutes.
If you can get him to move, youwon't have a missed collection.
Wow.
Imagine the efficiency in that.
Now you're getting consistencyof service area every time you
(18:38):
go to a particular service.
You're going to see it in thosekinds of ways.
Like right now, I get up in themorning and I do my usual
routine.
I get up, I eat breakfast, havecoffee, take a shower, and then
I go and I get my car, and thefirst thing that pops up is my
phone pops up and says, Oh,you're 20 minutes from work.
I didn't ask it to do that, itjust did it.
You're going to see thatintuitiveness start to permeate
(19:02):
all aspects of our lives.
We're going to go somewhere andwe're just going to be prepped
for where we're going.
It's going to be ready.
I don't have to do a whole lotof activity.
If you've got um listeners thatdrive across a scale, they have
to be weighed to get into afacility and go dump it.
Used to be I had to get upthere, position my truck, get
out of it, go talk to thescaler, or if it's unmanned, I
(19:24):
had to sit there and figure outwhat punch screen I need to hit.
And now I just drive up, itknows it's me.
It knew why I came there.
It just validates that I'mthere, takes the weight, opens
the gate arm, I go on in, and Inever get out of my truck.
Right?
You're gonna see those kinds ofefficiencies impacting the edge.
SPEAKER_00 (19:43):
Man, that's awesome! So digital transformation, you
got ERP.
Um, you've done extensive workaround those implementations.
Why do so many ERP rolloutsstruggle or fail?
SPEAKER_01 (19:55):
Yeah, so it's that is a digital transformation, you
can tie it right back to therollout of AI.
It's the same thing.
I'm implementing a piece oftech, and what I mostly have
about my business is myths andlegends.
So if you've ever worked with abusiness or something and you
start talking to everyone talkslike they know the process, and
then when you dig in, that's notreally how it's run.
(20:15):
You know, they find out there'san Excel spreadsheet that
someone has to do a manualupdate, or there's some you
know, access database in thecorner that's somehow holding up
your entire business that wasonly discovered once you started
the implementation of therollout of the ERP.
It's it's kind of like if Iasked you to chart a trip to New
York, but I didn't tell youwhere you're starting.
(20:37):
And I didn't tell you how longyou're gonna be there.
What are you gonna pack and howyou're gonna chart that where
are you gonna chart it from?
And so it's the same thing.
If you don't really understandwhat your business has when
you're trying to roll out thissoftware, there's no way to put
book ins.
I have my as is process andarchitecture.
Here's the 2B, that's where I'mtrying to get.
Now I can plot the journey fromhere to there.
(20:58):
I know what my decommissionsare, I know what the changes in
my organization are gonna be,and what I'm trying to get to.
Nothing's more frustrating.
I'm sure your listeners have ranto this where all of a sudden
they come in one morning, andnow here's a new piece of
software I got to deal with thatI've never seen before.
It clutches everything up.
The ivory tower doesn't alwaysunderstand what the boots on the
(21:20):
ground are doing, they'rehitting their metrics, but as
far as the process, maybe eightyears ago we loosely drew out a
process and I haven't touched itsince, but it's evolved because
that's what happens.
So when you look going back tothe AI discussion, that's
exactly why AI needs a humansteward.
Because the second you take itout of your POC or the lab and
(21:41):
you put it into the wild,entropy hits it.
The world that you had yesterdayhas changed.
And it's the same problem thatyou have with ERPs.
I buy a piece of software, Ihave this nice requirements list
that I want.
So then I spend all this timetrying to get the vendor to tick
off every single box I wanted.
And you spend 18, 20 months.
Well, here's the problem thesoftware you need today isn't
(22:02):
the software you're gonna need18 to 20 months from now.
So now you're in a perpetual,constantly trying to evolve that
software to catch up to you.
It's not gonna work, and you'regonna end up having scope creep.
And before you know it, you'regonna the vendor is gonna get
exhausted, they're gonna walkaway, and you're gonna end up
paying for a piece of softwareover the next five years that
(22:22):
doesn't do what you need it todo, but you're still paying for.
You can't even use the software.
I've seen it too many times.
You're just going to write itdown as a failed implementation,
get mad at the vendor.
And nine times out of 10, thisis what I see a lot of companies
do.
They grab the keys to theirbusiness, hand it to the vendor,
and say, Call me when you'redone.
No one's going to take care ofyour house like you.
(22:44):
Right.
Own your customer journey.
SPEAKER_00 (22:48):
Yeah.
The experience that goes alongwith what they're selling is a
lot that has to do with it beingsuccessful.
I mean, you can't just uh rollsomething out and expect people
to buy into it because theydon't really have a great
experience to attach with it.
So that's really that's reallyhuge.
I mean, those relationships,strategy, execution, uh, one
thing that doesn't get talkedabout enough is the relationship
(23:09):
between companies and theirtechnology vendors.
So I I really appreciate youbringing that out.
So for more additionalindustries, you know,
traditional industries, excuseme.
Uh, what's your advice forimplementing digital
transformation withoutoverwhelming the team?
I know you say you come in andyou see things coming in.
What would you say to them to bemore successful?
SPEAKER_01 (23:29):
Understand that it's put people at the top of your
plane.
It is a transformation, not ofthe systems, not really, it's a
transformation of the peoplethat are running those systems.
People have three propertiesabout them, two of which aren't
great.
We're lazy, we're greedy, andwe're messy.
(23:49):
So you have to steward peopleacross to change.
So if you're gonna come in andchange everything about every
facet of your business all atonce, that's too much change for
your business can to take.
So once you've got your bookins, once you know what your as
is is and what your to be isgonna be.
Let's just say that software'smissing 20% of the stuff.
(24:11):
The problem is so is thatsoftware when you first bought
it.
You've grown into it over 15 to20 years and you've worked with
your vendor to get it to thatnice, comfortable, fitting pair
of jeans that you're enjoyingtoday.
It's just outlived itsusefulness, and you need to go
to this new software.
Uh there should be three tops,four things in the vision that
(24:32):
pay for that entire software.
But no upfront, you're not goingto get every single thing you
want.
And it sure as hell isn't goingto replace the software checkbox
by checkbox.
Stop bringing the solution toyour vendor, bring the problem
to your vendor.
You're hiring them because theydo this for a living.
How would you like it if I cameto you and I started describing
(24:53):
to you how you should manageyour trust to service my
account?
You get a little upset.
It's like, look, this is my job.
I do this, right?
I I'll I'll do that part.
What do you need?
And then I'll tell you how thebest way is for my vehicles and
my my company to go handle thatbusiness.
It's the same thing withsoftware.
Invest in your relationship withyour vendor because I guarantee
you what you think you needtoday is not going to be what
(25:15):
you need 18 months from now.
It isn't.
The world's changing, everythingaround it changes, compliance,
governments change, presidentschange, everything's always in a
state of change.
So you're looking for a vendorthat will walk that road with
you, walk that journey, be ableto help you get the pieces that
you need in your software,whether it's AI or it's an ERP
(25:35):
system under the hood, to thatgets you the greatest value.
I've seen entire projects getderailed because of a checkbox
on the customer screen that sayshas or has not done the credit
check.
Well, that's great.
Software company added acheckbox there.
We added it.
Well, then now it hits thisother thing.
Well, I needed it to be able todo this.
(25:56):
You didn't ask for that.
You asked for a checkbox on thecustomer scene.
It says credit check or not.
So then before you know, they'respoon feeding that requirement.
And it's 18 months later,tempers are hot.
Nobody can explain how we gothere.
And that's just one paper cut ofthousands that occur doing a
major ERP.
Pick your battles, figure outthe best way to implement it.
(26:18):
This is particularly true withAI.
AI will increase by orders ofmagnitude and asymmetric ability
to deliver.
So if I just look across mybusiness and say this thing has
the greatest opportunity formargins, and I go and I up-level
it with AI, and there'ssomething that they feed, you're
(26:39):
about to overrun that next step.
Because this step is nowproducing 18 times faster than
it was, but the step it feedsinto is still the same rate that
it was before.
So you just you're eitherrunning them over or they're
asking you to slow down.
So now you spend a bunch ofmoney improving this one process
in your business with no way toget money out of it because you
(27:00):
have to slow down so you canmove up to the next one and then
move up to the next one.
Get a partner that can help lookat your processes and find how
am I going to implement this?
What's my transformation modellook like where I can get
maximum value out of AI duringthe rollout?
I don't want to get a success ina pilot and then have to pump
the brakes because the next guyin line is overrun by my new
(27:23):
efficiency.
That doesn't give me any valuefor the product.
SPEAKER_00 (27:28):
A lot of respect that you have to give to the
individuals out there.
I mean, that's huge.
I mean, you gotta you can't justgo around assuming that you
know, because you'reknowledgeable and you got a lot
of experience, that uh, youknow, you got to spread that
knowledge with everybody.
No, you got to actually worktogether.
And that's yeah.
I mean, I think you you'reimmolating that.
That's amazing.
How did you uh come about tohave those qualities or to
(27:50):
implement that with your yourwork environment?
SPEAKER_01 (27:53):
A lot of face planning.
All of these, I'm not coming toyou smarter than anybody.
I just tripped and face plantedon the ground over stuff.
I mean, I can tell you my firstimplementation of my software, I
was so proud of myself.
It was at a scale, and we we hadone of these things, it's a
mouse, and I'm old enough totell you we introduced it.
And this is a scaler, right?
So, this is uh probably asecurity guard or something that
(28:16):
you know was sitting there doinghis job, and I saw that guy grab
that mouse and white knuckle itinto the desk trying to get that
pointer to move, just he'd neverworked with one before.
And of course, I'm a nerd fornerds, so I'm thinking, yeah,
you don't use a mouse, that'sridiculous.
And I had a mentor at the timesaying, You've got the messy
part of people, you left peopleout of your solution, Mr.
(28:38):
Wizard.
I I used to get a lot of Mr.
Wizard when I was growing up.
It wasn't a very an endearingstatement.
He's like, So, because youdidn't include people, your your
product that you love so much isgoing to be rejected because
people can't use it.
So then we took it back, and Iwas like, Okay, let me think
about this from a person.
Let me go talk to a scaler who'snever used a mouse.
He goes, Look, I can hit thekeyboard or I can see that
(29:01):
screen.
And back then we didn't havetouch screens, we had punch
screens, they were like gooeyscreens, if you remember.
So he's knuckle punch it.
When we switched to that and gotrid of the mouse, and it just
popped up, that guy would see atruck come up on the scale and
the tickets up before that guycan even really settle his
truck.
I was like, Are you reading thescreens?
And he's like, No, I just gotmuscle memory, and that's but
(29:23):
that that's learning how peopleinterface with the solution,
right?
I regardless of what you thinkpeople are intelligent or not
intelligent, they're diabolical,all of them.
They will find a way to do thebest at their job, even if they
come up with undocumentedfeatures of your software, Mr.
Witcher, that you didn't thinkof to use.
(29:43):
They're getting their job done.
So all of it, and in the book,I've got 35 years of just
narrative after narrative afternarrative explaining where a lot
of these principles and conceptscame from.
But I can tell you whether it'san ERP system or AI, keep people
at the core of your vision ofthe project.
(30:05):
Have a vision.
I can tell you the number ofpeople complaining.
I didn't even know this softwarewas changing.
You didn't tell your front linespeople you were swapping out the
software, and they're surprisedon the day it disappears, and
you're wondering why it's goingso bad.
You know, and most of most ofthe time, because uh this
happened to me, I was we spentmonths around uh a conference
room table prepping this um pulpand paper company.
(30:28):
So this was the mill, so theybrought chips in, cooked them
up, and made paper.
And we're rolling it out.
Everybody's proud of themselves,you know, cutting ribbons and
popping champagne.
I'm saying this old scale, hemust have been 60.
This must have been like aretirement job for him.
So he's sitting there and hedidn't do anything, he was just
there in case a driver had aproblem working with the screen.
He was just supposed to help himand be secured.
(30:48):
So he's sitting there and hegoes, Well, um, how do you get a
chip truck out if you sell somechips?
I was a much younger then, sodon't judge me too harshly.
But I was like, No, no, you see,this is the mill.
You don't sell chips here, youyou bring chips in and cook.
So I'm now this guy's had thisjob for 30 years, and I'm
arrogant enough to now proceedto explain to him how paper's
(31:11):
being made from chips.
Like I said, much younger, fullof piss and vinegar at the time.
And he goes, Well, that's fine,Mr.
Wizard.
Tell me how there's a chip truckright there that the Foresters
want to sell because those chipsare worth more than that paper.
So you tell me how I get thatchip.
So I had to call up there andgo, Do you guys sell chips out
of the mill?
And he's like, Oh, we only dothat every once in a while.
(31:32):
Well, today it's once in awhile.
So every one of those storiesare just tripped, fell, had to
learn something from it.
And they all come back to thesame message it's people, right?
So that's your software vendorneeds to be people.
You need to have someone in yourcompany very close to the people
in that vendor company.
They all everybody wants to dothe right thing.
(31:52):
No one's not wanting to besuccessful, everybody wants a
successful implementation.
So if you build that strongrelationship with your vendor,
it's going to be successful, ifyou be honest with it.
If you don't know, it's like Ithink I know how this thing
runs, but I don't really,because Barbara's been here for
35 years and she really runs it.
God knows what process she'scome up with to make it her life
(32:13):
easier.
Fine, we need a discovery.
You won't be successful if wejust throw this thing down
everyone's throats.
Have a vision.
What are the four things youwant to get out of this new
implementation?
Just four.
It has to be able to do yourcore business, but it's not
going to do it the same way asyour old software.
And neither will AI.
AI will not do it exactly theway you have it today.
(32:36):
You've built a process forpeople.
And this is a wholly differentparadigm.
It's meant to repeat iterativelyvery quickly.
That's what it's designed to do.
So you and my best advice takethe necessary low value steps of
a process and delegate that toAI.
(32:57):
The high value, bring that andkeep it with your humans.
These, you know, I like to calla customer success.
So we we've learned this atAMCS.
We have our customer successpeople.
They call and interact with ourcustomers, make sure they're
happy, they're getting value outof the product, all of that
stuff.
But at the end of that call,that same CSR has to go and
write an email talking about theconversation we just had.
(33:20):
Are there any action items?
Is there anyone else in theorganization they need to
connect?
Let's put some meetings oncalendars and all that stuff.
So there was a maximum number ofcustomers they could call
because of all of thisnecessary, but that's low value
work.
Now AI does all that for him.
So he can flip a switch on acall, have a meeting like this.
(33:40):
It all goes into the AI.
The AI handles all thefollow-up, the action items,
schedules, all the people insidethat need to have the meetings,
and he can just go right on tothe next customer.
That has dramatically improvedcustomer satisfaction in our
company.
SPEAKER_00 (33:55):
That's what we want.
Customer satisfaction thatdrives revenue, uh, keep people
coming back for more.
I mean, what would you say to alot of the younger ones out
there that they may not have theexperience and they see AI as a
way to get into these doors andsay, hey, you know, I can do
that job.
I see what it is.
But uh, you know, like you said,too, you had some experience um,
you know, face planting.
Um, what would you say to theyounger ones out there?
(34:16):
I mean, maybe they just theythey're afraid to do the face
plant.
SPEAKER_01 (34:20):
So, well, don't, right?
But you gotta go fearless now,right?
We got to get back to a fear.
Think back when your childhoodwhere anything's falling,
because AI makes the cost offailing very small.
I can try something, see that itdidn't work, change my game up,
try it again, change my game up,try it again.
Adopt an iterative approach,learn through doing, and
(34:42):
understand the fundamentals ofthe domain you're gonna be the
steward of.
So, like uh one of the classeslast year that I taught at JU,
we had a concern.
AI is doing all this coding.
Where so programmers arebecoming a bit of a unicorn at
this point, but we still needarchitects.
Where do you get architectsfrom?
(35:02):
You get architects from peoplethat's been coding for hundreds
of thousands of lines, handjamming in code and
understanding how it works, andthen they become systems
architects or applicationarchitects or some other
architect.
But if I don't have humans doingthe coding bit anymore, where do
I get these guys?
And you know, they're allstarting to age out.
We're gonna be in trouble realquick if we lose them.
(35:23):
How do you get there?
So the class put down, well,what if we started with the
juniors that understood thefundamentals, but they used AI
exclusively and then confirmedtheir suspicion with the output
so I can see what good lookslike.
I may not know that these linesand these characters produce
that, but I understand theprinciple of solid architecture.
(35:45):
So, where did that come from?
There was a meme, you may haveeven seen this, that a
photographer took a selfie ofhimself in a t-shirt and he fed
it a prompt, you know, make itstark, dark and white, water
dripping, stern look.
And that thing came back amuseum work of art.
And that's what this guy's wholejob was taking photos that would
hang in a museum.
(36:07):
And he was devastated.
He's like, I'm out of a job.
I'm out of a job.
He had an hour of reflection.
He came back and goes, Well,wait, wait a minute.
The only thing that reallychanged was the camera.
I had to tell it the lens, I hadto tell it the refractal of the
light.
I had to describe the boxlights, the scene, the guy, the
(36:27):
picture, this the way the colortone needed to be.
I had to describe all that, andI used terms that an artist
would know, a photo, aphotographer would know.
And just like, you know, we usethese things and we say we're
gonna dial someone.
No one's dialed anyone indecades, but we still use the
term and we know what it means.
We know that when I'm dialingsomething, no dial, I'm calling
(36:50):
somebody, right?
That's what it means.
So we did the same thing withthis architecture group.
We taught solid architecturalprinciples, fundamentals.
We gave them an AI tool togenerate the output and then
have them confirm the output.
What we learned is that firstyears it was too abstract a
thought.
So you need a lot offundamentals.
Our second years, who had plentyof fundamentals, actually did
(37:13):
the best, spoiler alert.
My seniors fought the AI toothand nail because they had fallen
in love with code.
Oh, I don't like the way it didit.
I don't like it, it doesn'tmatter.
It did it, it produced theresults you're looking for.
You may have coded itdifferently, and they could not
get past Go because they werefighting the AI.
So, what we learned is that at acertain level of fundamentals,
(37:36):
knowing your business, you cancommunicate into AI, you
understand what it needs, andyou understand what good looks
like.
These kids produced an entirecapstone project in a four-month
class.
And if you've ever gone into amaster's program, these things
are passed down class to class.
You know, I'll just start on theauthorization, pass it to the
(37:58):
next master's group, and thenthey'll build this piece and
pass on.
These kids built the entirething, start to finish.
We were in the uh Invent for thePlanet.
Uh, it's 80 colleges around theworld, and AMCS got to sponsor
it and I was very excited aboutthat.
I got to be over there and be amentor.
And we gave them hard problemsthat we learned folks couldn't
solve, like world hunger andfood waste at a global level.
(38:20):
40% of all food goes to thegarbage and never gets eaten.
As a human, which I'membarrassed that anyone's
starving in the world at thatkind of waste, right?
We all should be.
So I gave this problem to thesekids, and they took in the
problem on a Friday night.
They had all of Saturday to workon it, and then Sunday they
pitched it Shark Tank style tojudges, and they had built an
(38:44):
entire app back in commercialmarket everything in a day that
I honestly believe would solvepotentially world hunger and
most definitely food waste, tothe point they got two of the
judges willing to offer them amillion dollars in finance to
take the program forward, toinvest in it.
That's awesome.
The increased iterativecapability.
(39:06):
There, we're we're getting outof being cogs, we got to change
our mindset, and I'm notmeasuring my value from that
work anymore.
And that's where the attritionis gonna happen is people who
judge their value by that workrather than the output of that
work are gonna have a hard timeadopting this new world of AI to
(39:26):
be a steward.
SPEAKER_00 (39:28):
You mentioned the cog, you mentioned stewards.
Um, you know, get away from thecog, be more of a creator
mindset.
Um, more so, I feel like youknow, a business mindset.
You know, you gotta be aself-starter, you gotta start
somewhere.
Yeah.
And that's great.
And great examples to show that,you know, you can be successful.
You just gotta put your feetdown, start one foot at a time,
(39:51):
and uh you'll you'll get there.
Um, sometimes you feel likeyou're running out of time, but
that's just something I thinkanxiety or something is getting
in the way.
So push past that.
SPEAKER_01 (40:00):
Yeah.
Yeah, look, creativity is stillthe dominion of humanity.
So you have plenty of timebecause you'd be surprised how
quickly you can learn the skillsto be a steward over it, as long
as you already have thefundamental knowledge, you can
quickly take that on.
Yeah, I would be shocked andamazing that if say a per I'd
(40:21):
say more than 50% of youraudience decided right now,
having that if they nevertouched AI to go, I'm gonna go
learn this.
And I know a lot about mytrucking, I know a lot about my
industry and business.
They could come up with a reallycool app or a really cool
agentic tool and have that thingdone within three weeks out
there, earning them thousandsand thousands and thousands of
(40:42):
dollars.
That's the power that we nowhave available to us.
SPEAKER_00 (40:47):
Nice.
I mean, that's really good.
I mean, it sounds like you are agreat leader.
Um, you also uh do a lot of uhpersonal work too in your career
that you bring forward and bringlight to issues that people need
to see that they can they cansurprise themselves along the
way.
So um, you're also mentoring andteaching, uh, I believe I get
that correctly, or you're seeingon the next generation enter
(41:08):
into the space.
So uh what a what a greatexample for a lot of leaders out
there to help the young ones.
Um, so yeah, what what would yousay to leaders uh to to be more
inspiring to individuals outthere?
SPEAKER_01 (41:20):
Don't buy into the hype that the oligarchs are
putting out there that this issome apocalyptic outcome.
Future is not happening to us,it's happening because of us.
We get to choose it.
So if you're gonna take aleadership role, inspire those
around you.
I'm a huge believer in servantleadership.
Your job there is to enablethose that you serve to be able
(41:43):
to do the best of their job.
Introduce them to thecapabilities and tools to be
able to operate in this world.
Get them to maybe change abouthow they measure their
self-worth.
That data in an Excelspreadsheet that produces that
bar chart, that's not yourvalue, right?
That is, it can't be your valueand it won't be anymore.
The output of that, thedecisions that can be made off
(42:06):
of that, is what you're workingtowards.
And you're driving agents toproduce that quicker, faster, so
those decisions can be made.
If I would ask anything ofleaders is to just stop
spreading this apocalyptic viewthat AI is wiping everything and
this universal basic incomenonsense.
I believe that we have a chancereally to put our arms around
(42:29):
this and build that future thatwe want.
It just needs to go all the wayback to elementary school.
We need to change the way we'reeducating.
We don't need to build cogs andbusiness that have repeatable
excellence.
Like you said before, you know,the reduction of people is going
to happen naturally because Igot a tractor.
Used to have 200 guys out therewith hand plows to plow an acre
to grow food to feed everyone.
(42:50):
I can now do it with two guys ina tractor.
And I think even today you cando it with just the tractor and
the will for it to go out thereand do the work.
We're doing the same thing,except we're getting out of the
repetitive service side of theworld.
That burden is now going to beAIs.
Our job is to be able to commandas many of those as we can
within the domain of expertisethat we have to make sure that
(43:11):
they're producing the output.
The work is now going to bedelegated to the AI.
I don't need to do it.
Elon Musk framed this to best,in my opinion, where he said,
you take AI out of the equation,you go back 30, 40 years, there
were skyscrapers where therewere three floors full of
business analysts withcalculators and paper crunching
(43:32):
the numbers for businesses tomake decisions on.
And today, one guy with an Excelworkbook could cook all of
those, that whole three floorsof analysts, right?
So that's just him using abetter tool.
Put AI where it belongs, a toolto be able to produce output in
orders of magnitude that we'rereally starting to play with.
(43:54):
And there's no experts outthere.
I'm not even one.
I'm learning new stuff every daythat you can do with this.
So just go in there, leap intoit, see what it can do, see what
you can get out of it.
You might surprise yourself.
SPEAKER_00 (44:09):
Well, yeah.
I mean, uh, I got a chance to goto Japan and see all the things
they do there, the height of alot of their infrastructure, um,
a lot of their customer servicethat they're able to give.
Um, I mean, we can bring a lotof that too when we travel and
you see that and you Hey, let'show do we raise the bar and
raise the experience?
So when someone puts the productin their hand, um, there's
(44:31):
there's that just greatexperience that comes with it.
And I think you're doing that.
I think your company isemulating that.
So keep up the good work.
Um, I think we got a long waysto go, and uh it's like you
said, the doom, the doom stuff,we gotta stop with that.
I may get attention for maybe acouple seconds, but then it
quickly dies off because westill got work to do.
(44:52):
So yeah.
SPEAKER_01 (44:53):
Very true.
Well, if you're interested inthe stories, grab my book on my
website.
I have all of this information.
You can get a deck for free.
So if you're wanting to bring AIinto your business, I don't
charge for it.
I'm happy to just send me a DMon LinkedIn.
I'll I'll share everything thatI have so that you can be
successful at.
SPEAKER_00 (45:12):
That's that's great.
And I think a lot of people umshould follow your lead on that,
man.
So we don't have to hold on tothese secrets or uh industry
secrets that people think, hey,this is what makes me special.
No, we can share them and knowthat people, hey, are gonna be
successful out there becausethey're also sharing that same
information and we can worktogether.
So, man, Evan, it's been a greatconversation.
Um, really appreciate you comingon, talking about what you guys
(45:34):
are doing over there.
Um, so yeah, don't don't stop,keep going.
Uh, for anyone out therelistening, wants to also
connect.
So appreciate the uh theLinkedIn.
Um, is there a handle and justyour last name?
SPEAKER_01 (45:46):
Uh it's Evan Schwartz Live, or just go to
evanjschwartz.com, all one word,and you can get to anything and
everything I'm doing, latestevents, podcasts like this, all
from that site.
SPEAKER_00 (45:57):
That's great, man.
So I really appreciate you doingthis.
Um anything else you want thelisteners to know before we go?
Um, maybe something I missed oryou wanted to touch on.
SPEAKER_01 (46:07):
Just it's a lot more approachable than you think.
It's not hard.
Just try it.
You'll be you will surpriseyourself.
SPEAKER_00 (46:16):
Man, I love it.
And you know, that's that's whywe have these shows, and yeah,
we want people to feelemboldened to do more, uh, to
take their you know, theirknowledge to another level and
to deliver that to not just uhthe packages and products, but
also the knowledge.
So thank you so much.
SPEAKER_01 (46:33):
Your experience is what makes you unique right now.
SPEAKER_00 (46:35):
Take advantage of no one can take it away from you,
you can only use it to and thenadd to it.
So, yeah, great advice, Evan.
So thanks for breaking that downfor us in a big way.
We really appreciate youconnecting with uh listeners out
there.
For anyone that wants to learnmore, please uh share this
episode.
If you like what you see andlike what you hear, uh please
share it with an individual outthere.
And uh, I think we got moregreat stories coming up for
(46:57):
individuals like yourself andmany other uh industry experts
that are trying to make it makesuccess stories out of what's
going on in this world.
So really appreciate that, man.
So if you got uh this value fromthis episode, be sure to
subscribe, share it with someonein your network, and uh that's
delivered.
Welcome back to another show of That's
Delivered.
I'm your host, Trucking Ray, andwe talk with people that are
shaping the future of trucking,logistics, and the industries
that keep America moving.
Today we're diving into AI,digital transformation, and what
actually looks like forreal-world operations, not just
theory.
I'm joined by Evan Schwartz,Chief Innovation Officer of AMCS
(00:50):
Group.
With 35 years of experience,Evan has helped organizations
build AI and data-drivenstrategies into industries like
waste, recycling, and logistics,turning big ideas into real
measurable results.
Evan, welcome to the show.
(01:12):
Living the dream, man.
You?
Yeah, I'm doing well.
So glad to um have you on theshow, talk about what you're
doing.
I'm always excited about theguests that that you know take
the time out of their day tocome on the show and talk about
what they're doing, the changesthat they're making in the
industry and beyond.
So um can't wait to hear more,man.
You know, I know you're outthere traveling and you took the
time out to come on.
(01:32):
So uh what's the week been likeuh getting started?
SPEAKER_01 (01:35):
Oh yeah.
So the week is uh mostlytraveling and it's full of
logistics, right?
The uh air traffic, it's notwhat it used to be.
I I'm old enough to tell you Icould just throw cash down, walk
out on the tarmac, climb on aplane, and be off.
Those days are gone.
It's it's not great these days.
So maybe that's something we canwork on.
Maybe AI can help improve thatlittle strategy.
(01:56):
But I'm in Dublin right now.
Uh we're having an off-sitemeeting.
So I'm speaking to you from thehotel room.
I hope that's okay.
SPEAKER_00 (02:03):
Yeah, that's perfect, man.
So uh let's let's let thelisteners know.
Give us a little bit about yourbackground.
Um, you spent decades out thereacross operations, um,
agriculture, um, I think, andand now innovation.
Um, what's been in the commonthread for your career?
You know, how do you held thoseroles as chief officer,
architect, or COO, now uh was itchief innovation officer?
(02:27):
So how has that perspective ontechnology changed as you've
moved closer to the businessrole?
SPEAKER_01 (02:33):
So, interestingly enough, I started in the gaming
industry as a teenager and Istarted my first company writing
games.
And, you know, I thought I I wasreally attracted to creation,
creating something.
Because I remember when I openedup my PO box and pulled that big
bag of orders for a game I hadbuilt, I was hooked.
I was hooked on business.
Um, later I've refined it andfound that it's not just the
(02:55):
creation, but it's the creationof value.
Seeing how software or digitalsystems improve the lives of
everyone and produce value, I'mreally attracted to that,
particularly around how it getsinto the real world.
Uh, I as I left the gaming worldand went into the first business
venture, which was it started inthe insurance that got me kind
(03:16):
of hooked.
And then I went into what Iwould consider the wealth
industries, where you seetrucking, where you see the
movement of material.
You know, it's the I mybest-selling book, People,
Places and Things, is writtenall about the fact that it
there's people involved.
You got to pick something upfrom one location, move it to
another, buying and sellingthings, right?
And it was in the natural gasindustry, the transportation of
(03:38):
gas around the country, havingto get it sold, how to give it
truck, uh, the trucking with theoil and distribution of gas to
various gas stations introduceda lot of interesting logistics.
That led me in strangely intothe forestry side of the world,
where log trucks had to go outinto the woods and get a whole
load of uh wood.
(03:59):
Some of it was hardwood, some ofit was pulp wood to be able to
make paper.
Um, and again, I was surprisedat the level of complexity built
into those businesses and howhard it was just to be able to
schedule a truck, to be at somepoint in the woods, to pick up a
load of logs to get it to themill, because that mill can't
(04:20):
shut down.
I mean, it's a million dollarsan hour for that thing not to
run.
So it has to be fed.
And then as that industrychanged from virgin fibers into
secondary um, you know, OCC, DLKtype of uh fires that got into
the waste and recycling side ofthe world.
And it was then really youstarted to see all the
(04:42):
connective tissues between allthe industries.
Like even in the forestrybusiness, no part of that tree
would go to waste, right?
The terpenes got extracted outof it during the paper making
and got sold into the chemicalindustry for flavorings and fuel
additives and some of the pulpnon-woven.
There's a sheet of paper behinda flat screen, I'm sure you're
staring at right now to see myugly mug.
(05:03):
All of these industries areconnected and they have waste
components.
And it's trucking, it'slogistics that has to move this
around.
The entire scrap metal industry,which is really a hyper-focused
version of the waste andrecycling side of the business,
will tell you that scrap doesn'tlike to move far.
Every mile you have to take thatand move it down the road, it's
(05:24):
just killing the margins in it.
So there's such a huge focus onthe logistics, the operational
side, keeping trucks in goodworking order, being able to
maintain a fleet of vehicles,making sure that you know what
they're doing, when they'redoing it, every minute they're
not producing is a minute ofwaste opportunity.
So where I really kind of landedwas across that connective
(05:46):
tissue of all of theseindustries at AMCS group.
How can what we've learned inthe forestry business help the
scrap business, help, you know,just general logistics and field
services and predictivemaintenance to make sure that I
can maybe even predict usingalgorithms on whether I need to
take that truck out of service,you know, or I might need to put
(06:07):
that truck up for auction.
I've gotten about all the life Ican get out of that thing, and I
need to go shop for a new truck.
All of those things drive bottomline value.
If you're concerned aboutenvironmental social governance,
making sure that you're having alight touch on the uh the
environment, some of thissoftware, we have 700,000 plus
(06:29):
trucks around the world undermanagement in our software, and
we're shaving 17 gallons ofdiesel a month off of the burn
rate by just optimizing howthose trucks do their work.
unknown (06:43):
Wow.
SPEAKER_01 (06:44):
That's a lot of fuel not being burnt.
That's also a lot of fuel Ididn't have to pay for.
Right?
So we're we're starting to drivebeyond the let's just be
environmentally responsiblebecause it's the right thing to,
oh, you know what?
It also happens to make money,right?
If if I don't have to burn thatfuel and spill that into the
(07:04):
environment, that's money Idon't have to spend.
You know, if I if I know exactlyhow to maintenance my truck
optimally, and I don't have tohave a historical predictive
algorithm going, well, wemaintenance it every six months
before.
Let's keep doing that until itbreaks.
And I can take an oil sample andtell you exactly what's going on
in that engine and what you needto do with it.
(07:25):
Now I can be more responsible inthe way that I'm maintening the
truck.
I'm not taking it out ofservice, I'm making sure that
it's doing its job.
And more importantly, I'm notputting oil into the environment
that has to be cleaned anddisposed.
Everywhere there's a high carbonfootprint in your business is an
opportunity for efficiencies andbottom line impact.
(07:46):
And the only reason why wedidn't run at it like that
before is because we didn't havethe solutions of software or or
the passion to chase it.
And now with AI and some of thetech we've got in this world,
it's easy to do.
But the most thing that bringsme here today, Ray, is the
passion about people, places,and things.
That there is a future of AIwithout people.
(08:08):
Why are we doing it?
I'm trying to make sure I'mgetting the message out that the
future is not happening to us,it happens because of us.
We have a choice, right?
Take ownership, don't be thevictim, own that future.
And the reality is we just spenta year and a half and a whole
ton of money by a lot of smartpeople on AI that didn't work,
(08:30):
didn't succeed.
And they let a lot of goodpeople go with their business.
Sure, this thing was going toreduce headcount, and it was
just the way they did it waswrong, right?
SPEAKER_00 (08:39):
So, what do you say about that?
I mean, AI and real worldoperations, there's a lot of
noise around it, like you'resaying, and the perspective of
how businesses can actuallyleverage AI to modernize
operations or improve efficiencywithout complicating things.
Um, it's got to be a challengein itself, or just uh trying to
package it correctly.
SPEAKER_01 (08:59):
Yeah, it's it's an OCM, organizational change
management.
It's a person problem, it's apeople problem.
The tech is there.
What makes AI useful is theperson who can clearly
communicate to it, give it avery clear instruction, and can
designate the data context itneeds to perform the job that
(09:19):
you just asked it to do.
Where AI goes off the rails isit's like a super smart young
intern that's come in and theyknow a lot, but they don't have
your experience.
They don't have your intuitionyet.
The difference is that internmight be a little self-conscious
and tell you they're not reallysure about this.
AI, on the other hand, superconfident because it knows
(09:41):
everything there is to knowabout your industry and your
business, but it knows nothingabout your business, the way you
want to operate, the way youwork.
So it's so confident knowingeverything, it doesn't come back
and go, yeah, you didn't exactlygive me enough context.
So what I'm gonna do is I'mgonna look down in my super big
brain that knows everything, andI'm gonna get the answer from
(10:02):
that.
And that might not apply to yourbusiness, and that's what we
call today hallucinations.
Every time you see ahallucination from an AI, it's
because the task we asked it wastoo vague, or the data we gave
it to be able to perform itsactivity was incomplete, the
context was incomplete.
SPEAKER_00 (10:20):
And then the input that's a lot of work to get it
right.
SPEAKER_01 (10:24):
Yeah, it is a lot, and then the third leg of what
I'm calling the AI steward, thatperson that sits over top of a
bunch of AIs to get a lot done,has to know what good looks
like.
You don't have to know theanswer, right?
That that's I get some pushbacksometimes.
Well, if I know the answer, whatdo I need AI for?
You don't need to know theanswer, you just need to know
what it gave you doesn't lookright.
(10:44):
And now I'm gonna dig in andjust validate it, or it does
look right.
It that's about where I expectit to be.
Let's move forward.
So, stewards are really good atcommunicating that task, knowing
the data it needs, and knowingwhat good looks like.
We're moving away fromrepetitiveness as a burden of
humanity and into rapiditeration.
(11:07):
I need to be able to do a lotmore.
I don't mind if it messes upbecause the cost of doing that
iteration is so low, I can justfix it, do it again.
And when I get it right, I cando a hundred thousand of those
things where I couldn't comeclose to that level of iteration
without the power of AI, right?
That's right.
SPEAKER_00 (11:26):
Yeah, I mean, that's there's pluses and pros, you
know, that or cons that comewith everything.
So uh there's gotta be you gottalook at both sides of the coin.
I mean, when these companies areworking with uh all this AI and
automation, what's the biggestmistakes you see when they try
to adopt AI or automation?
We just kind of call that out.
SPEAKER_01 (11:44):
Yeah, look, the biggest number one mistake is
thinking that I'm gonna saveheadcount.
I'm gonna put this thing in,it's gonna run this little
section perfectly.
And if I had 10 people, I cannow cut it down to two.
And no, it's not gonna work thatway first.
You when you're applying AI, getyourself a positive outcome
that's gonna justify the cost ofit outside of headcount.
(12:07):
At the end of the journey, Ray,if you want to right size your
business, fine.
But what AI should be doing isasymmetric growth.
I should, with the number ofpeople I've got right now, I
should be able to grow mybusiness not by an extra 20%.
That's that that's a childchange.
I should be able to use thistech to go 3x, 4x, 8x in growth,
(12:29):
right?
And then if I've looked aroundand I've grown as big as I want,
by all means, right size yourbusiness.
I'm not gonna tell you how tonot right size your business.
But if you go in with thatsentiment first, you're gonna
get caught with your pants down.
There is a change managementthat you're gonna go through.
Now, there is gonna be a certainnatural attrition if you decide
(12:50):
to adopt AI.
And maybe the comfort of a lotof your listeners, except for
those who are running their ownbusinesses, is a lot of this is
gonna impact the SGNA side ofyour business, the back office
work.
Those guys that are boots on theground with a blowtorch or a
wrench getting it done today,their life is gonna get much,
much easier, but not changesignificantly for a while.
(13:13):
But the back office is gonnachange significantly.
And today, you can't go out intothe market and hire an AI
steward.
Now, my role at JU, we'rebuilding a curriculum around
getting our first master'sgraduate as an AI steward.
And they have domain domainexpertise.
So you'll have AI stewards thatare good at finance and AI
(13:34):
stewards that are good atlogistics and dispatch, and AI
stewards that are really good atfront office customer service
modules.
So they have breadth ofknowledge, they're clear
communicators, and theyunderstand when I ask the AI to
do something, it's gonna needaccess to this data to achieve
it.
Right.
And then when I can watch it,because I've got metrics and I
know what good looks like.
(13:55):
So now I could manage maybe ahundred of these agents, and
these agents are callingsubagents to be able to do
refined work.
That's gonna be the back office.
And if anything, they're able togrow that business so rapidly
that you're gonna see,especially if your business
deals around logistics, servicevehicles, the truck is tied to
(14:16):
the revenue, like a garbagetruck goes out and does it.
So that's that's where the moneyis.
That's the offer, right?
That's the service.
You're gonna see that grow.
You're gonna be adding 10, 15,20 trucks to keep up with how
quick the business has grown.
That's gonna be one of itsbiggest problems.
But thinking that you're goingto reduce headcount as a first
measure of success almost alwaysends in failure.
(14:40):
Almost always.
SPEAKER_00 (14:41):
Yeah, so we got to change the culture.
I mean, it's a lot of industriesout there that are adopted that,
thinking that hey, that's thebest way to go because maybe
that's how we used to do thingsback in the day is cut back on
on headcount.
But moving forward, you got tokeep that experience.
You don't want to lose thatvalue.
So what you're saying.
SPEAKER_01 (14:57):
Um AI doesn't have yet is your experience.
There's a difference betweenknowledge and experience, big
difference, right?
We got tons of knowledge in abook, but a book can't look at
the the walk around a truck andtell you whether or not that
vehicle is in trouble, you know,whether those tires need to be
changed.
You're gonna have to, there's acertain amount of experience
that you have to know how tolook under things, check your
(15:19):
chains, make sure if you're alog truck, how you've got those
logs situated, what theunbinding station is, how many
trips a day can you make to theto the woods?
That all comes out ofexperience, grading the
material.
If you're the driver and you'reresponsible for the material and
you're just taking a quick lookat the material you picked up,
because your livelihood countson that material getting across
a scale somewhere.
(15:41):
AI is not doing that yet.
SPEAKER_00 (15:43):
And it's not going to do it for a long, long time.
So, I mean, margins are tightout there for trucking
industries.
You got fuel costs, like youmentioned.
Uh, what does AI do?
That uh, you know, where does itcreate the most impact for
immediately for companies?
SPEAKER_01 (16:00):
Yeah, look, that's a great question.
So, from where you see a lot ofit, and it's not the AI that
everyone talks about, the thechatty transformer AI, it's in
predictive algorithmic AI is therouting optimization.
If I'm if I'm doing servicingwithin a city and I'm running
the same routes over and overagain, optimizing the way that
those routes should be run isthe immediate best say.
(16:23):
But where the new transformers,the agentic AI is really adding
value into anything to do withtrucking, long haul, short haul,
logistics, equipment,maintenance, all of that is the
way it's transforming thedriver's existence.
The driver gets up in themorning, let's just say that
you've got a route, you've got2,000 stops, you got a service
today.
You you basically have a morningdebrief, a stand-up, you look at
(16:46):
what your day's work is going tobe.
So that's eating up some time,you get in your truck, and
you're now also the guy that'shaving to collect details on any
exceptions that occur along theroute.
Take all that away.
Get in your truck, do your job.
It'll tell you where you need togo.
You have faith that someone hasworked out all the logistics,
what the servicing is, what youneed to do, make sure that all
(17:09):
if you've got any repair, ifyou're doing servicing, that
your van has all the stuff in itthat it needs to have.
That has all been done for you.
You were to maintain safety, getto the service location, perform
the service, and then get backin your truck and go on to the
next one.
And that becomes the drive is tomaintain safety.
You know, we were talking beforerecording, is that you could go
(17:31):
back eight years.
There was no tech on a truck.
And now they're mobile datacenters.
I mean, they're practicallymobile offices, they've got
their own cell uplink, theirIoT.
You can learn everything thereis to learn about that engine on
the road.
There's cameras all around itrecording full video feeds
constantly for safety, but alsothose are now being able to be
(17:53):
tapped in by AI so that I cansee things.
I can look out at the servicearea.
One of the things that we'relooking at right now is to be
able to tap into the publicfeeds.
Most cities have public feedsfor cameras.
So if I'm a garbage truck andI'm going to maybe a restaurant
to pick up their container andempty it, and his bread truck is
(18:14):
blocking it.
If the cameras in the city cansee that block 15, 20 minutes
ahead of the truck, so now AIcan call that restaurant and
say, You've got a bread truckblock in your container.
My truck's going to be there in20 minutes.
If you can get him to move, youwon't have a missed collection.
Wow.
Imagine the efficiency in that.
Now you're getting consistencyof service area every time you
(18:38):
go to a particular service.
You're going to see it in thosekinds of ways.
Like right now, I get up in themorning and I do my usual
routine.
I get up, I eat breakfast, havecoffee, take a shower, and then
I go and I get my car, and thefirst thing that pops up is my
phone pops up and says, Oh,you're 20 minutes from work.
I didn't ask it to do that, itjust did it.
You're going to see thatintuitiveness start to permeate
(19:02):
all aspects of our lives.
We're going to go somewhere andwe're just going to be prepped
for where we're going.
It's going to be ready.
I don't have to do a whole lotof activity.
If you've got um listeners thatdrive across a scale, they have
to be weighed to get into afacility and go dump it.
Used to be I had to get upthere, position my truck, get
out of it, go talk to thescaler, or if it's unmanned, I
(19:24):
had to sit there and figure outwhat punch screen I need to hit.
And now I just drive up, itknows it's me.
It knew why I came there.
It just validates that I'mthere, takes the weight, opens
the gate arm, I go on in, and Inever get out of my truck.
Right?
You're gonna see those kinds ofefficiencies impacting the edge.
SPEAKER_00 (19:43):
Man, that's awesome! So digital transformation, you
got ERP.
Um, you've done extensive workaround those implementations.
Why do so many ERP rolloutsstruggle or fail?
SPEAKER_01 (19:55):
Yeah, so it's that is a digital transformation, you
can tie it right back to therollout of AI.
It's the same thing.
I'm implementing a piece oftech, and what I mostly have
about my business is myths andlegends.
So if you've ever worked with abusiness or something and you
start talking to everyone talkslike they know the process, and
then when you dig in, that's notreally how it's run.
(20:15):
You know, they find out there'san Excel spreadsheet that
someone has to do a manualupdate, or there's some you
know, access database in thecorner that's somehow holding up
your entire business that wasonly discovered once you started
the implementation of therollout of the ERP.
It's it's kind of like if Iasked you to chart a trip to New
York, but I didn't tell youwhere you're starting.
(20:37):
And I didn't tell you how longyou're gonna be there.
What are you gonna pack and howyou're gonna chart that where
are you gonna chart it from?
And so it's the same thing.
If you don't really understandwhat your business has when
you're trying to roll out thissoftware, there's no way to put
book ins.
I have my as is process andarchitecture.
Here's the 2B, that's where I'mtrying to get.
Now I can plot the journey fromhere to there.
(20:58):
I know what my decommissionsare, I know what the changes in
my organization are gonna be,and what I'm trying to get to.
Nothing's more frustrating.
I'm sure your listeners have ranto this where all of a sudden
they come in one morning, andnow here's a new piece of
software I got to deal with thatI've never seen before.
It clutches everything up.
The ivory tower doesn't alwaysunderstand what the boots on the
(21:20):
ground are doing, they'rehitting their metrics, but as
far as the process, maybe eightyears ago we loosely drew out a
process and I haven't touched itsince, but it's evolved because
that's what happens.
So when you look going back tothe AI discussion, that's
exactly why AI needs a humansteward.
Because the second you take itout of your POC or the lab and
(21:41):
you put it into the wild,entropy hits it.
The world that you had yesterdayhas changed.
And it's the same problem thatyou have with ERPs.
I buy a piece of software, Ihave this nice requirements list
that I want.
So then I spend all this timetrying to get the vendor to tick
off every single box I wanted.
And you spend 18, 20 months.
Well, here's the problem thesoftware you need today isn't
(22:02):
the software you're gonna need18 to 20 months from now.
So now you're in a perpetual,constantly trying to evolve that
software to catch up to you.
It's not gonna work, and you'regonna end up having scope creep.
And before you know it, you'regonna the vendor is gonna get
exhausted, they're gonna walkaway, and you're gonna end up
paying for a piece of softwareover the next five years that
(22:22):
doesn't do what you need it todo, but you're still paying for.
You can't even use the software.
I've seen it too many times.
You're just going to write itdown as a failed implementation,
get mad at the vendor.
And nine times out of 10, thisis what I see a lot of companies
do.
They grab the keys to theirbusiness, hand it to the vendor,
and say, Call me when you'redone.
No one's going to take care ofyour house like you.
(22:44):
Right.
Own your customer journey.
SPEAKER_00 (22:48):
Yeah.
The experience that goes alongwith what they're selling is a
lot that has to do with it beingsuccessful.
I mean, you can't just uh rollsomething out and expect people
to buy into it because theydon't really have a great
experience to attach with it.
So that's really that's reallyhuge.
I mean, those relationships,strategy, execution, uh, one
thing that doesn't get talkedabout enough is the relationship
(23:09):
between companies and theirtechnology vendors.
So I I really appreciate youbringing that out.
So for more additionalindustries, you know,
traditional industries, excuseme.
Uh, what's your advice forimplementing digital
transformation withoutoverwhelming the team?
I know you say you come in andyou see things coming in.
What would you say to them to bemore successful?
SPEAKER_01 (23:29):
Understand that it's put people at the top of your
plane.
It is a transformation, not ofthe systems, not really, it's a
transformation of the peoplethat are running those systems.
People have three propertiesabout them, two of which aren't
great.
We're lazy, we're greedy, andwe're messy.
(23:49):
So you have to steward peopleacross to change.
So if you're gonna come in andchange everything about every
facet of your business all atonce, that's too much change for
your business can to take.
So once you've got your bookins, once you know what your as
is is and what your to be isgonna be.
Let's just say that software'smissing 20% of the stuff.
(24:11):
The problem is so is thatsoftware when you first bought
it.
You've grown into it over 15 to20 years and you've worked with
your vendor to get it to thatnice, comfortable, fitting pair
of jeans that you're enjoyingtoday.
It's just outlived itsusefulness, and you need to go
to this new software.
Uh there should be three tops,four things in the vision that
(24:32):
pay for that entire software.
But no upfront, you're not goingto get every single thing you
want.
And it sure as hell isn't goingto replace the software checkbox
by checkbox.
Stop bringing the solution toyour vendor, bring the problem
to your vendor.
You're hiring them because theydo this for a living.
How would you like it if I cameto you and I started describing
(24:53):
to you how you should manageyour trust to service my
account?
You get a little upset.
It's like, look, this is my job.
I do this, right?
I I'll I'll do that part.
What do you need?
And then I'll tell you how thebest way is for my vehicles and
my my company to go handle thatbusiness.
It's the same thing withsoftware.
Invest in your relationship withyour vendor because I guarantee
you what you think you needtoday is not going to be what
(25:15):
you need 18 months from now.
It isn't.
The world's changing, everythingaround it changes, compliance,
governments change, presidentschange, everything's always in a
state of change.
So you're looking for a vendorthat will walk that road with
you, walk that journey, be ableto help you get the pieces that
you need in your software,whether it's AI or it's an ERP
(25:35):
system under the hood, to thatgets you the greatest value.
I've seen entire projects getderailed because of a checkbox
on the customer screen that sayshas or has not done the credit
check.
Well, that's great.
Software company added acheckbox there.
We added it.
Well, then now it hits thisother thing.
Well, I needed it to be able todo this.
(25:56):
You didn't ask for that.
You asked for a checkbox on thecustomer scene.
It says credit check or not.
So then before you know, they'respoon feeding that requirement.
And it's 18 months later,tempers are hot.
Nobody can explain how we gothere.
And that's just one paper cut ofthousands that occur doing a
major ERP.
Pick your battles, figure outthe best way to implement it.
(26:18):
This is particularly true withAI.
AI will increase by orders ofmagnitude and asymmetric ability
to deliver.
So if I just look across mybusiness and say this thing has
the greatest opportunity formargins, and I go and I up-level
it with AI, and there'ssomething that they feed, you're
(26:39):
about to overrun that next step.
Because this step is nowproducing 18 times faster than
it was, but the step it feedsinto is still the same rate that
it was before.
So you just you're eitherrunning them over or they're
asking you to slow down.
So now you spend a bunch ofmoney improving this one process
in your business with no way toget money out of it because you
(27:00):
have to slow down so you canmove up to the next one and then
move up to the next one.
Get a partner that can help lookat your processes and find how
am I going to implement this?
What's my transformation modellook like where I can get
maximum value out of AI duringthe rollout?
I don't want to get a success ina pilot and then have to pump
the brakes because the next guyin line is overrun by my new
(27:23):
efficiency.
That doesn't give me any valuefor the product.
SPEAKER_00 (27:28):
A lot of respect that you have to give to the
individuals out there.
I mean, that's huge.
I mean, you gotta you can't justgo around assuming that you
know, because you'reknowledgeable and you got a lot
of experience, that uh, youknow, you got to spread that
knowledge with everybody.
No, you got to actually worktogether.
And that's yeah.
I mean, I think you you'reimmolating that.
That's amazing.
How did you uh come about tohave those qualities or to
(27:50):
implement that with your yourwork environment?
SPEAKER_01 (27:53):
A lot of face planning.
All of these, I'm not coming toyou smarter than anybody.
I just tripped and face plantedon the ground over stuff.
I mean, I can tell you my firstimplementation of my software, I
was so proud of myself.
It was at a scale, and we we hadone of these things, it's a
mouse, and I'm old enough totell you we introduced it.
And this is a scaler, right?
So, this is uh probably asecurity guard or something that
(28:16):
you know was sitting there doinghis job, and I saw that guy grab
that mouse and white knuckle itinto the desk trying to get that
pointer to move, just he'd neverworked with one before.
And of course, I'm a nerd fornerds, so I'm thinking, yeah,
you don't use a mouse, that'sridiculous.
And I had a mentor at the timesaying, You've got the messy
part of people, you left peopleout of your solution, Mr.
(28:38):
Wizard.
I I used to get a lot of Mr.
Wizard when I was growing up.
It wasn't a very an endearingstatement.
He's like, So, because youdidn't include people, your your
product that you love so much isgoing to be rejected because
people can't use it.
So then we took it back, and Iwas like, Okay, let me think
about this from a person.
Let me go talk to a scaler who'snever used a mouse.
He goes, Look, I can hit thekeyboard or I can see that
(29:01):
screen.
And back then we didn't havetouch screens, we had punch
screens, they were like gooeyscreens, if you remember.
So he's knuckle punch it.
When we switched to that and gotrid of the mouse, and it just
popped up, that guy would see atruck come up on the scale and
the tickets up before that guycan even really settle his
truck.
I was like, Are you reading thescreens?
And he's like, No, I just gotmuscle memory, and that's but
(29:23):
that that's learning how peopleinterface with the solution,
right?
I regardless of what you thinkpeople are intelligent or not
intelligent, they're diabolical,all of them.
They will find a way to do thebest at their job, even if they
come up with undocumentedfeatures of your software, Mr.
Witcher, that you didn't thinkof to use.
(29:43):
They're getting their job done.
So all of it, and in the book,I've got 35 years of just
narrative after narrative afternarrative explaining where a lot
of these principles and conceptscame from.
But I can tell you whether it'san ERP system or AI, keep people
at the core of your vision ofthe project.
(30:05):
Have a vision.
I can tell you the number ofpeople complaining.
I didn't even know this softwarewas changing.
You didn't tell your front linespeople you were swapping out the
software, and they're surprisedon the day it disappears, and
you're wondering why it's goingso bad.
You know, and most of most ofthe time, because uh this
happened to me, I was we spentmonths around uh a conference
room table prepping this um pulpand paper company.
(30:28):
So this was the mill, so theybrought chips in, cooked them
up, and made paper.
And we're rolling it out.
Everybody's proud of themselves,you know, cutting ribbons and
popping champagne.
I'm saying this old scale, hemust have been 60.
This must have been like aretirement job for him.
So he's sitting there and hedidn't do anything, he was just
there in case a driver had aproblem working with the screen.
He was just supposed to help himand be secured.
(30:48):
So he's sitting there and hegoes, Well, um, how do you get a
chip truck out if you sell somechips?
I was a much younger then, sodon't judge me too harshly.
But I was like, No, no, you see,this is the mill.
You don't sell chips here, youyou bring chips in and cook.
So I'm now this guy's had thisjob for 30 years, and I'm
arrogant enough to now proceedto explain to him how paper's
(31:11):
being made from chips.
Like I said, much younger, fullof piss and vinegar at the time.
And he goes, Well, that's fine,Mr.
Wizard.
Tell me how there's a chip truckright there that the Foresters
want to sell because those chipsare worth more than that paper.
So you tell me how I get thatchip.
So I had to call up there andgo, Do you guys sell chips out
of the mill?
And he's like, Oh, we only dothat every once in a while.
(31:32):
Well, today it's once in awhile.
So every one of those storiesare just tripped, fell, had to
learn something from it.
And they all come back to thesame message it's people, right?
So that's your software vendorneeds to be people.
You need to have someone in yourcompany very close to the people
in that vendor company.
They all everybody wants to dothe right thing.
(31:52):
No one's not wanting to besuccessful, everybody wants a
successful implementation.
So if you build that strongrelationship with your vendor,
it's going to be successful, ifyou be honest with it.
If you don't know, it's like Ithink I know how this thing
runs, but I don't really,because Barbara's been here for
35 years and she really runs it.
God knows what process she'scome up with to make it her life
(32:13):
easier.
Fine, we need a discovery.
You won't be successful if wejust throw this thing down
everyone's throats.
Have a vision.
What are the four things youwant to get out of this new
implementation?
Just four.
It has to be able to do yourcore business, but it's not
going to do it the same way asyour old software.
And neither will AI.
AI will not do it exactly theway you have it today.
(32:36):
You've built a process forpeople.
And this is a wholly differentparadigm.
It's meant to repeat iterativelyvery quickly.
That's what it's designed to do.
So you and my best advice takethe necessary low value steps of
a process and delegate that toAI.
(32:57):
The high value, bring that andkeep it with your humans.
These, you know, I like to calla customer success.
So we we've learned this atAMCS.
We have our customer successpeople.
They call and interact with ourcustomers, make sure they're
happy, they're getting value outof the product, all of that
stuff.
But at the end of that call,that same CSR has to go and
write an email talking about theconversation we just had.
(33:20):
Are there any action items?
Is there anyone else in theorganization they need to
connect?
Let's put some meetings oncalendars and all that stuff.
So there was a maximum number ofcustomers they could call
because of all of thisnecessary, but that's low value
work.
Now AI does all that for him.
So he can flip a switch on acall, have a meeting like this.
(33:40):
It all goes into the AI.
The AI handles all thefollow-up, the action items,
schedules, all the people insidethat need to have the meetings,
and he can just go right on tothe next customer.
That has dramatically improvedcustomer satisfaction in our
company.
SPEAKER_00 (33:55):
That's what we want.
Customer satisfaction thatdrives revenue, uh, keep people
coming back for more.
I mean, what would you say to alot of the younger ones out
there that they may not have theexperience and they see AI as a
way to get into these doors andsay, hey, you know, I can do
that job.
I see what it is.
But uh, you know, like you said,too, you had some experience um,
you know, face planting.
Um, what would you say to theyounger ones out there?
(34:16):
I mean, maybe they just theythey're afraid to do the face
plant.
SPEAKER_01 (34:20):
So, well, don't, right?
But you gotta go fearless now,right?
We got to get back to a fear.
Think back when your childhoodwhere anything's falling,
because AI makes the cost offailing very small.
I can try something, see that itdidn't work, change my game up,
try it again, change my game up,try it again.
Adopt an iterative approach,learn through doing, and
(34:42):
understand the fundamentals ofthe domain you're gonna be the
steward of.
So, like uh one of the classeslast year that I taught at JU,
we had a concern.
AI is doing all this coding.
Where so programmers arebecoming a bit of a unicorn at
this point, but we still needarchitects.
Where do you get architectsfrom?
(35:02):
You get architects from peoplethat's been coding for hundreds
of thousands of lines, handjamming in code and
understanding how it works, andthen they become systems
architects or applicationarchitects or some other
architect.
But if I don't have humans doingthe coding bit anymore, where do
I get these guys?
And you know, they're allstarting to age out.
We're gonna be in trouble realquick if we lose them.
(35:23):
How do you get there?
So the class put down, well,what if we started with the
juniors that understood thefundamentals, but they used AI
exclusively and then confirmedtheir suspicion with the output
so I can see what good lookslike.
I may not know that these linesand these characters produce
that, but I understand theprinciple of solid architecture.
(35:45):
So, where did that come from?
There was a meme, you may haveeven seen this, that a
photographer took a selfie ofhimself in a t-shirt and he fed
it a prompt, you know, make itstark, dark and white, water
dripping, stern look.
And that thing came back amuseum work of art.
And that's what this guy's wholejob was taking photos that would
hang in a museum.
(36:07):
And he was devastated.
He's like, I'm out of a job.
I'm out of a job.
He had an hour of reflection.
He came back and goes, Well,wait, wait a minute.
The only thing that reallychanged was the camera.
I had to tell it the lens, I hadto tell it the refractal of the
light.
I had to describe the boxlights, the scene, the guy, the
(36:27):
picture, this the way the colortone needed to be.
I had to describe all that, andI used terms that an artist
would know, a photo, aphotographer would know.
And just like, you know, we usethese things and we say we're
gonna dial someone.
No one's dialed anyone indecades, but we still use the
term and we know what it means.
We know that when I'm dialingsomething, no dial, I'm calling
(36:50):
somebody, right?
That's what it means.
So we did the same thing withthis architecture group.
We taught solid architecturalprinciples, fundamentals.
We gave them an AI tool togenerate the output and then
have them confirm the output.
What we learned is that firstyears it was too abstract a
thought.
So you need a lot offundamentals.
Our second years, who had plentyof fundamentals, actually did
(37:13):
the best, spoiler alert.
My seniors fought the AI toothand nail because they had fallen
in love with code.
Oh, I don't like the way it didit.
I don't like it, it doesn'tmatter.
It did it, it produced theresults you're looking for.
You may have coded itdifferently, and they could not
get past Go because they werefighting the AI.
So, what we learned is that at acertain level of fundamentals,
(37:36):
knowing your business, you cancommunicate into AI, you
understand what it needs, andyou understand what good looks
like.
These kids produced an entirecapstone project in a four-month
class.
And if you've ever gone into amaster's program, these things
are passed down class to class.
You know, I'll just start on theauthorization, pass it to the
(37:58):
next master's group, and thenthey'll build this piece and
pass on.
These kids built the entirething, start to finish.
We were in the uh Invent for thePlanet.
Uh, it's 80 colleges around theworld, and AMCS got to sponsor
it and I was very excited aboutthat.
I got to be over there and be amentor.
And we gave them hard problemsthat we learned folks couldn't
solve, like world hunger andfood waste at a global level.
(38:20):
40% of all food goes to thegarbage and never gets eaten.
As a human, which I'membarrassed that anyone's
starving in the world at thatkind of waste, right?
We all should be.
So I gave this problem to thesekids, and they took in the
problem on a Friday night.
They had all of Saturday to workon it, and then Sunday they
pitched it Shark Tank style tojudges, and they had built an
(38:44):
entire app back in commercialmarket everything in a day that
I honestly believe would solvepotentially world hunger and
most definitely food waste, tothe point they got two of the
judges willing to offer them amillion dollars in finance to
take the program forward, toinvest in it.
That's awesome.
The increased iterativecapability.
(39:06):
There, we're we're getting outof being cogs, we got to change
our mindset, and I'm notmeasuring my value from that
work anymore.
And that's where the attritionis gonna happen is people who
judge their value by that workrather than the output of that
work are gonna have a hard timeadopting this new world of AI to
(39:26):
be a steward.
SPEAKER_00 (39:28):
You mentioned the cog, you mentioned stewards.
Um, you know, get away from thecog, be more of a creator
mindset.
Um, more so, I feel like youknow, a business mindset.
You know, you gotta be aself-starter, you gotta start
somewhere.
Yeah.
And that's great.
And great examples to show that,you know, you can be successful.
You just gotta put your feetdown, start one foot at a time,
(39:51):
and uh you'll you'll get there.
Um, sometimes you feel likeyou're running out of time, but
that's just something I thinkanxiety or something is getting
in the way.
So push past that.
SPEAKER_01 (40:00):
Yeah.
Yeah, look, creativity is stillthe dominion of humanity.
So you have plenty of timebecause you'd be surprised how
quickly you can learn the skillsto be a steward over it, as long
as you already have thefundamental knowledge, you can
quickly take that on.
Yeah, I would be shocked andamazing that if say a per I'd
(40:21):
say more than 50% of youraudience decided right now,
having that if they nevertouched AI to go, I'm gonna go
learn this.
And I know a lot about mytrucking, I know a lot about my
industry and business.
They could come up with a reallycool app or a really cool
agentic tool and have that thingdone within three weeks out
there, earning them thousandsand thousands and thousands of
(40:42):
dollars.
That's the power that we nowhave available to us.
SPEAKER_00 (40:47):
Nice.
I mean, that's really good.
I mean, it sounds like you are agreat leader.
Um, you also uh do a lot of uhpersonal work too in your career
that you bring forward and bringlight to issues that people need
to see that they can they cansurprise themselves along the
way.
So um, you're also mentoring andteaching, uh, I believe I get
that correctly, or you're seeingon the next generation enter
(41:08):
into the space.
So uh what a what a greatexample for a lot of leaders out
there to help the young ones.
Um, so yeah, what what would yousay to leaders uh to to be more
inspiring to individuals outthere?
SPEAKER_01 (41:20):
Don't buy into the hype that the oligarchs are
putting out there that this issome apocalyptic outcome.
Future is not happening to us,it's happening because of us.
We get to choose it.
So if you're gonna take aleadership role, inspire those
around you.
I'm a huge believer in servantleadership.
Your job there is to enablethose that you serve to be able
(41:43):
to do the best of their job.
Introduce them to thecapabilities and tools to be
able to operate in this world.
Get them to maybe change abouthow they measure their
self-worth.
That data in an Excelspreadsheet that produces that
bar chart, that's not yourvalue, right?
That is, it can't be your valueand it won't be anymore.
The output of that, thedecisions that can be made off
(42:06):
of that, is what you're workingtowards.
And you're driving agents toproduce that quicker, faster, so
those decisions can be made.
If I would ask anything ofleaders is to just stop
spreading this apocalyptic viewthat AI is wiping everything and
this universal basic incomenonsense.
I believe that we have a chancereally to put our arms around
(42:29):
this and build that future thatwe want.
It just needs to go all the wayback to elementary school.
We need to change the way we'reeducating.
We don't need to build cogs andbusiness that have repeatable
excellence.
Like you said before, you know,the reduction of people is going
to happen naturally because Igot a tractor.
Used to have 200 guys out therewith hand plows to plow an acre
to grow food to feed everyone.
(42:50):
I can now do it with two guys ina tractor.
And I think even today you cando it with just the tractor and
the will for it to go out thereand do the work.
We're doing the same thing,except we're getting out of the
repetitive service side of theworld.
That burden is now going to beAIs.
Our job is to be able to commandas many of those as we can
within the domain of expertisethat we have to make sure that
(43:11):
they're producing the output.
The work is now going to bedelegated to the AI.
I don't need to do it.
Elon Musk framed this to best,in my opinion, where he said,
you take AI out of the equation,you go back 30, 40 years, there
were skyscrapers where therewere three floors full of
business analysts withcalculators and paper crunching
(43:32):
the numbers for businesses tomake decisions on.
And today, one guy with an Excelworkbook could cook all of
those, that whole three floorsof analysts, right?
So that's just him using abetter tool.
Put AI where it belongs, a toolto be able to produce output in
orders of magnitude that we'rereally starting to play with.
(43:54):
And there's no experts outthere.
I'm not even one.
I'm learning new stuff every daythat you can do with this.
So just go in there, leap intoit, see what it can do, see what
you can get out of it.
You might surprise yourself.
SPEAKER_00 (44:09):
Well, yeah.
I mean, uh, I got a chance to goto Japan and see all the things
they do there, the height of alot of their infrastructure, um,
a lot of their customer servicethat they're able to give.
Um, I mean, we can bring a lotof that too when we travel and
you see that and you Hey, let'show do we raise the bar and
raise the experience?
So when someone puts the productin their hand, um, there's
(44:31):
there's that just greatexperience that comes with it.
And I think you're doing that.
I think your company isemulating that.
So keep up the good work.
Um, I think we got a long waysto go, and uh it's like you
said, the doom, the doom stuff,we gotta stop with that.
I may get attention for maybe acouple seconds, but then it
quickly dies off because westill got work to do.
(44:52):
So yeah.
SPEAKER_01 (44:53):
Very true.
Well, if you're interested inthe stories, grab my book on my
website.
I have all of this information.
You can get a deck for free.
So if you're wanting to bring AIinto your business, I don't
charge for it.
I'm happy to just send me a DMon LinkedIn.
I'll I'll share everything thatI have so that you can be
successful at.
SPEAKER_00 (45:12):
That's that's great.
And I think a lot of people umshould follow your lead on that,
man.
So we don't have to hold on tothese secrets or uh industry
secrets that people think, hey,this is what makes me special.
No, we can share them and knowthat people, hey, are gonna be
successful out there becausethey're also sharing that same
information and we can worktogether.
So, man, Evan, it's been a greatconversation.
Um, really appreciate you comingon, talking about what you guys
(45:34):
are doing over there.
Um, so yeah, don't don't stop,keep going.
Uh, for anyone out therelistening, wants to also
connect.
So appreciate the uh theLinkedIn.
Um, is there a handle and justyour last name?
SPEAKER_01 (45:46):
Uh it's Evan Schwartz Live, or just go to
evanjschwartz.com, all one word,and you can get to anything and
everything I'm doing, latestevents, podcasts like this, all
from that site.
SPEAKER_00 (45:57):
That's great, man.
So I really appreciate you doingthis.
Um anything else you want thelisteners to know before we go?
Um, maybe something I missed oryou wanted to touch on.
SPEAKER_01 (46:07):
Just it's a lot more approachable than you think.
It's not hard.
Just try it.
You'll be you will surpriseyourself.
SPEAKER_00 (46:16):
Man, I love it.
And you know, that's that's whywe have these shows, and yeah,
we want people to feelemboldened to do more, uh, to
take their you know, theirknowledge to another level and
to deliver that to not just uhthe packages and products, but
also the knowledge.
So thank you so much.
SPEAKER_01 (46:33):
Your experience is what makes you unique right now.
SPEAKER_00 (46:35):
Take advantage of no one can take it away from you,
you can only use it to and thenadd to it.
So, yeah, great advice, Evan.
So thanks for breaking that downfor us in a big way.
We really appreciate youconnecting with uh listeners out
there.
For anyone that wants to learnmore, please uh share this
episode.
If you like what you see andlike what you hear, uh please
share it with an individual outthere.
And uh, I think we got moregreat stories coming up for
(46:57):
individuals like yourself andmany other uh industry experts
that are trying to make it makesuccess stories out of what's
going on in this world.
So really appreciate that, man.
So if you got uh this value fromthis episode, be sure to
subscribe, share it with someonein your network, and uh that's
delivered.
Popular Podcasts
Dateline NBC
Current and classic episodes, featuring compelling true-crime mysteries, powerful documentaries and in-depth investigations. Follow now to get the latest episodes of Dateline NBC completely free, or subscribe to Dateline Premium for ad-free listening and exclusive bonus content: DatelinePremium.com
Hey Jonas!
Hey Jonas! The official Jonas Brothers podcast. Hosted by Kevin, Joe, and Nick Jonas. It’s the Jonas Brothers you know... musicians, actors, and well, yes, brothers. Now, they’re sharing another side of themselves in the playful, intimate, and irreverent way only they can. Spend time with the Jonas Brothers here and stay a little bit longer for deep conversations like never before.
Betrayal Weekly
Betrayal Weekly is back for a new season. Every Thursday, Betrayal Weekly shares first-hand accounts of broken trust, shocking deceptions, and the trail of destruction they leave behind. Hosted by Andrea Gunning, this weekly ongoing series digs into real-life stories of betrayal and the aftermath. From stories of double lives to dark discoveries, these are cautionary tales and accounts of resilience against all odds. From the producers of the critically acclaimed Betrayal series, Betrayal Weekly drops new episodes every Thursday. If you would like to share your story, you can reach out to the Betrayal Team by emailing them at betrayalpod@gmail.com and follow us on Instagram at @betrayalpod and @glasspodcasts. Please join our Substack for additional exclusive content, curated book recommendations, and community discussions. Sign up FREE by clicking this link Beyond Betrayal Substack. Join our community dedicated to truth, resilience, and healing. Your voice matters! Be a part of our Betrayal journey on Substack.