TCP084: "AI Does Not Exist" with Brandon McRill, CEO of 5Out

Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
(00:00):
You are listening to the TechChef Podcast.
This is episode number 84January 14, 2025.
This show is powered by GrowthAdvisors International Network where
travel and hospitalitycompanies come to grow.
For more information pleasevisit gainadvisors.com.

(00:21):
This is Brandon McCrell, cofounder and CEO of Five out and you're
listening to Skip on the TechChef Podcast.
Off Premise Strategy, Business Continuity.
How about a taste test ofrestaurant technology?
Drive thru or curbside mobileapps or AI?

(00:41):
It's all on the menu, cookingup for the date.
It's a recipe for success.
You're in good hands with aTech Chef.
Make a plan to be your best strategize.
Welcome back to the Tech Chef.

(01:02):
I'm Skip Kimpel, your guide tothe latest hospitality technology
insights and inspirationstraight from sunny Fort Lauderdale,
Florida.
Whether you've been with usfor a while or you're joining us
for the very first time, weare thrilled to have you here.
Your continued support andengagement motivate us to push boundaries

(01:23):
and deliver top notch techcontent that helps professionals
like you stay ahead, tacklechallenges and drive innovation.
Thank you for being aninvaluable part of our growth and
success.
Anybody that was at CES thisyear, all I have to say is wow.

(01:43):
It was an action packed showwith major announcements coming from
all industries.
Of most interest to me howeverwas the keynotes from Nvidia and
Sony.
I'm not going to spoil it foryou, but if you missed those, make
sure you jump on YouTube andcheck them out.
This week's show isinteresting as we challenge the concept

(02:04):
of AI.
Crazy, you say?
Well if you are curious youare in the right place as we explore
this topic with our specialguest, Brandon McGrill.
With more than 26 years ofexperience in the hospitality industry,
he is the co founder and CEOof five Out, a company that leverages
machine learning to provideactionable directions for food, service

(02:27):
and resort businesses.
He leads the vision, strategyand execution of five out, which
aims to revolutionize the wayhospitality operators design, manage
and optimize their operationsand customer experience.
Brandon has a very strongbackground in food and beverage conceptual
development, having foundedand co founded several successful

(02:51):
restaurants and cafes.
Buckle up and let's get alittle controversial with this week's
title episode.
AI does not Exist.
Let's go.
Brandon, thank you so much forjoining the show today.
Can you share yourprofessional journey leading up to
founding five out and reallywhat inspired you to step into the

(03:14):
restaurant and hospitalitytech industry?
First of all, thanks.
Great to be here and greattalking with you and seeing you again.
And then to answer thosequestions, I would say five out was
begotten from my career in thehospitality industry.
So I got started in therestaurant industry when I was 15
years old.
I was playing soccer at thetime, competitively, got injured.

(03:35):
I was injured enough to thepoint where I couldn't play soccer
competitively anymore, but Istill could walk and walk well.
And so my mother said, youneed to take that 15 year old energy,
get out of the house and takeit somewhere else.
So I went looking for a joband in all the places that I applied
without much experience at theripe old age of 15, the first place
that hired me was actually theGrosse Pointe Hunt Club in Grosse

(03:57):
Pointe Woods, Michigan.
And I was parking cars andwashing dishes.
So I would park cars in alittle tuxedo shirt and suit and
then I would take it off andrun inside and wash dishes and then
come back outside and put thatshirt and suit back on and unpark
people's cars at the end oftheir meal.
And I just kind of becameenamored with the industry from that
point.
I was always interested infood growing up, but I really started

(04:17):
to understand where it camefrom, literally and figuratively,
and I stuck with it from there.
So throughout the rest of highschool and college, I worked in restaurants.
I transferred from going toschool in downtown Detroit to Columbia
Chicago and accidentally fellinto working at a high end restaurant
called Alinea, which is athree Michelin star, four star restaurant
in Chicago.
And that's when I really madea material commitment to transition

(04:39):
my career to hospitality.
Previously I was studying photography.
So from there I went to workfor some pretty relatively well known
restaurant tours like one off Hospitality.
Dm Paul Khan, Richard Melmanto let us entertain you, Danny Meyer
at Union Square, Union SquareHospitality group at the MoMA, John
George Lucas, David Burke,Marcus Samuelson.

(05:00):
And then I started opening myown restaurants.
So I was in my late 20s.
I started opening my ownrestaurants in the Lower east side
of Manhattan.
I opened Pearl Nash in 2012,Rebel in 2015 we got a Michelin star
with him.
Being only open for three anda half months at my second restaurant
put us on a national radar.
We got scooped up by apublicly traded REIT to open up a
ground floor restaurant in a$350 million mixed vertical neighborhood

(05:22):
in Philly.
Went and opened, that stayed,worked with those developers to open
up two more restaurants.
And that has been the last 26years of restaurant touring.
Wow, fast go backwards.
Yeah, that was, it was a lot.
It was A run, but it was 26years, so obviously time to get stuff
in.
And by the way, you did a lotof name dropping there.
But the only thing that stuckin my head is the picture of you

(05:44):
parking cars.
It reminds me of the.
The kid in goodfellas parkingthe cars for all the mobsters out
front.
Barely, barely, barely tallenough to look over the steering
wheel.
No driver's license.
But thankfully, the parkinglot was very close to the front door.
So fast backwards to 2015.

(06:04):
I'm in New York City.
I've just opened up my second restaurant.
And facing headwinds as I hadseen it, many other restaurants that
I had worked at before and formany other restaurant tours that
I'd worked that before, andthat's that.
Generally speaking, thehospitality industry and the restaurant
industry in particular, has afinite profit margin.

(06:25):
And with the two largestexpenses being labor and cost of
goods.
If you can understand whatyour product demand is going to be,
and therefore what you'repurchasing and expense needs to be
across those two things andunderstand them in advance, then
you can potentially be more profitable.
And the easiest way to, tokind of speak to that is if you're
running a banquet hall, let'ssay, for example, and you book a

(06:46):
wedding six months in advance,and the people who are hosting the
wedding have sent out requestsfor RSVP and ask people if they want
steak or chicken, and theyhave filled it out that they're coming
and which one they're eating.
Then you know how many peopleyou have, and you know what they're
going to, what they're goingto eat.
So you know, you only need topurchase and prep exactly for the
amount of people that you knoware coming and it's already paid

(07:07):
for and exactly what foodthey're going to be, they're going
to be having.
So you know how many staff youneed to cook that many chicken and
steak and how many drinksthey're going to have, and all these
things become predetermined.
And if a restaurant, anyrestaurant, not just a banquet hall,
but any restaurant, had thatkind of visibility, looking forward
into what was going to occurin advance, they could have a better
handle on how much staff do Ineed doing what job, where and when,

(07:31):
and how much product do Ineed, where and when.
So myself and my CFO at thetime, Mike Marion, who's now my CTO
at five out, we decided, isthere enough data out there to allow
us to create something simpleand rudimentary that can help inform
us and give us a betterunderstanding of what is likely to
occur in the future.
Utilizing Excel and a thumbdrive plugged into a Micros terminal

(07:56):
in a basement, we pulled rawdata and utilized other outside data
like foot traffic and weatherand local and national events, and
made simple algorithmshandwritten in Excel and put this
data through those algorithmsand use it to forecast what was going
to happen going forward, justby a single week, broken down by
day, and then by meal period,and then by hour, starting with revenue

(08:19):
and then items sold.
And if you know what yourhourly revenue is and your headcount
and what items are going toget sold, then you can simply use
simple arithmetic to reverseengineer into a labor budget and
a purchasing budget that ismore specific and finite than you
would on your own by guessing.

(08:40):
And in doing so, we were ableto bring $300,000 of additional net
profitability to a $3 million restaurant.
So that was a sizable materialbeneficial impact.
So we applied that to a secondrestaurant of mine and then three
more, and at five restaurantswe crashed out.
Because Mike is a human beingand not a computer, and he was spending

(09:02):
about 10 hours a week perrestaurant calculating this data,
turning it into a.
First of all pulling it, thencalculating it, turning it into a
deliverable that's actuallyutilizable by a restaurateur, which
I'm one, so I get to makecomments like that.
So it has to be simple.
And then tracking changes andfollowing up to make sure they did
it.
Those seven steps for eachrestaurant took about eight to ten
hours per week.

(09:22):
So we hit bandwidth issues.
So that Mike and I said, whydon't we turn this into software?
If we can automate thisprocess, automate the data digestion,
automate the output creation,automate the updates, automate the
changes, automate thedistribution, automate the tracking,
then we can apply this to alot more people.

(09:43):
We did that.
Rolled it out of my own restaurants.
That was working, Saving time,saving money.
Then I started handing it outfor free to my friends.
They started utilizing it andtelling us they liked it a lot.
And then we said, I think wegot something here.
We should probably take thisto a broader market.
That was how five out got born.
Interesting.
Well, that leads perfectlyinto the main portion of what the

(10:05):
show is about.
It's really not a pitch forfive Out.
Even though I'm fascinated byyour product, you know, I consider
you to be one of the expertsin the industry around AI and ML
machine learning.
So.
Right.
Start heading down that routebecause everything you've explained
so far is important to theconversation because you Identified

(10:30):
very specific data points thathelp formulate the success of the
output.
How should restaurantsleverage AI and ML to optimize restaurant
operations and customer experiences?
Let's have the complicatedpart of the conversation first and
get it out of the way and thenwe'll have the simple of the conversation.
So the complicated part iswhat's AI, what's ML, what's the

(10:53):
difference between the two?
So AI doesn't really exist inour world because it is, it is a
computer learning on its own,teaching itself and creating on its
own, and having its own mindand doing what it wants to do.
And as far as we all know,besides what we've seen in the movies,
that hasn't been created andit certainly hasn't been rolled out

(11:15):
at scale.
So we're not talking AI.
So then let's talk about ML.
So machine learning islearning, but not coming up with
any ideas on its own and notdoing what it wants to do.
Therefore it's simply programming.
Higher level, more complexprogramming, but programming nonetheless.
And so what that means isgoing into the simple part of this

(11:37):
conversation is it's justautomation, which is just technology,
which is just replacing thehorse drawn carriage with an automobile.
So for all those folks thatare out there that are a little bit
skeptical or nervous or scaredor hesitant or anything that doesn't

(11:58):
get you leaned in into theconversation around AI and ML, it's
just another form oftechnology that's being introduced
to us.
Just like the printing press,or just like the Internet or like
the computer.
It's just one more thingthat's here to help us in our life.
And I think when people cometo that realization, which is going

(12:20):
to happen for everybody intheir own time, but the faster people
come to that realization, andthe more people come to that realization,
the more quickly that ourlives are all going to get a lot
simpler as a byproduct of ML.
First and foremost.
That's the lofty, broad,flowery language answer.
Well, very controversialanswer too.

(12:42):
So are you saying at thispoint in time AI is a misused term?
Yes, very much so.
I think that because of thecomplexity and the nuance of what
the actual differentiation isbetween machine learning and AI,
that a lot of people arelumping those two things together
and calling one the same andmaking them synonymous.

(13:05):
And in fact they're very, very different.
And so what I want everybodyto start doing is understanding that
starting at the top, which Ilike to call ChatGPT, that's machine
learning.
It's not AI, it's not true AIand everything, I'll call it below
that, if that's the top andthat's the gold standard, is also
machine learning.
So we are not in the age of AI yet.

(13:25):
We're actually only in the ageof machine learning.
And that's just advancedalgorithms utilizing large data sets.
It's not thinking on its own,it's not writing its own course,
it doesn't have its own thoughts.
And therefore it's not out tokill us, such as Terminator 2, which
is great because we don't need that.

(13:46):
Okay, so I will remove AI fromthis conversation.
And by the way, I think thatlast statement is probably going
to break the Internet whenthis podcast launches because we
might have some conversationsgoing on in line in online regarding
that, which is.
I love those conversations.
I would love for people topoint URI to true AI being deployed

(14:10):
currently in the restaurantspace, the hospitality space, or
any other sector, because Ihave yet to see it.
So what are some of thebiggest challenges that you faced
when integrating machinelearning into five outs platform?
The biggest challenge is thatsoftware development and machine

(14:31):
learning are two completelydifferent functions.
Almost the same as building anairplane versus building a rocket
ship.
An airplane has all of itsengineering requirements and its
engine requirements, and allof those things are well understood

(14:54):
and known by most of the samegroup of people.
But building a rocket ship isyou have the engineering side and
then you have the rocket science.
And it's a great example.
So if you want to build asoftware platform, you need engineers.
If you want to do somethingwith machine learning, you need data
scientists and other ancillaryfolks around the world of data science.

(15:15):
And you can't work one withoutthe other.
So if you want to build asoftware solution that utilizes machine
learning, you actually have toengage two independently skilled
teams and bring them togetherand help their work coalesce.
So I guess what I'm saying isit's a lot easier and more straightforward

(15:36):
to build software than it isto build software that utilizes ML.
And that was probably thebiggest challenge, is learning that
and then figuring out how toexist in that world where we had
to bring those two areas offocus together.
So when creating this languagemodel and this database, how do you
ensure data security andethical use of.

(15:59):
I keep wanting to use the term AI.
Now you throw me all off whenhandling sensitive restaurant and
customer data.
I mean, what does that looklike within the industry?
So I'll speak to ourexperience and then I'll speak to
other folks experience that Ican only, you know, talk about from
A from a outsider perspective,as far as we're concerned here in,

(16:24):
in our company, in oursolution, we don't utilize any personalized
data.
So we specifically, you know,block that out.
If anybody tries to give it tous, we tell them we don't want it.
Reasons are somewhat businessrelated, somewhat security like you're
talking about, but also wejust don't need it.
We're not focused on theindividualistic Persona, perspective,

(16:47):
experience, we're focused onthe backend financial data aspects.
So that's our experience,that's our walk of life when it comes
to other people.
Now that we live in a cloudbased world where data is mostly
stored with Amazon andutilizing aws, which is where a lot

(17:08):
of the government stores a lotof its data, I like to look at an
equal playing field.
So if you're able to engagewith AWS and I'm able to engage with
aws, the United Statesgovernment and IBM and JP Morgan
are able to engage with aws.
It's not exactly apples toapples, but it's at the very least
apples to oranges.

(17:28):
We all kind of have a fairshot at keeping our data secure.
Now we know starting up fromthe very top level all the way down
to the simplest of solutions,there is data breaches and it happens.
And that's the world that welive in, that we, most of us have
chosen to opt into, but not everybody.
So I think that we're allpretty secure or at the very least
as secure as most folks in theworld who focus on making it so that

(17:52):
it is we are secure, stay that way.
So that's the way I thinkabout data security and data privacy.
In terms of ethics, that's awhole other conversation and that
we are no way, shape or formin a position to have figured out.
I mean I think at the end ofthe day, going back to talking about
the gold standard of ML, whenwe're talking about open air, we're

(18:14):
talking about chat.
Machine learning is machinelearning is machine learning.
Data training is data trainingis data training.
Data sets are what's going toset people apart, no pun intended.
So whoever has access to thebest data sets to put into their
models.
And this is going to getinteresting for everybody, skip it.

(18:35):
In the next one to five plusyears it's going to be who has a
data set that no one else canget access to, who has that the report
from the United States government?
Who has that analysis fromBain, who has that Harvard study?
Who has that unpublished dataset that they can put into their
model that all of a sudden isgoing to lift it up and get it much

(18:57):
higher.
So being able to tier up intodifferent levels of data set and
who can get the best data setsand who can, you know, unfortunately
this is not a word I like touse, but obfuscate that data set
from some of their competitorsis what's gonna, is what's gonna
help separate some cream fromthe rest of the pack, so to speak.
Yeah, it's interesting you saythat because that has come to light

(19:19):
recently.
We're Magic Gate is actuallytraining a model for a very specific
industry that will be uniqueand will be competitive and will
be advantageous to Magic 8.
It becomes the proprietaryinformation of our organization and
having that unique data setreally gives us a competitive advantage.

(19:42):
So I can't wait to talk moreabout that in the future, but totally
get where you're coming from, Brandon.
We're going to take a littlebreak and when we come back, we are
going to dig into some of the,the insights that you can drive from
this data that you'recollecting, the future of AI and
hospitality and a bunch ofother great things.
So everybody stay tuned.
Are you ready to level up yourtech game?

(20:03):
Understand the world of xr.
Dive into the world ofextended reality with Magic Gate's
XR Bootcamp.
Master the concepts of virtualreality, augmented reality and mixed
reality in just weeks.
No special equipment orprevious knowledge is required.
Limited seats are available tosign up now@bootcamp.magicgate.com

(20:24):
that's boot camp.magicgate.comyour future in XR starts there.
Okay, let's pick up where weleft off.
Let's talk about some of themachine learning driven insights.
How can tech companies usemachine learning to help restaurants
make better decisions aboutmenu design, staffing or marketing

(20:46):
campaigns?
Obviously I'm moving outsidethe five and out sector here, but
I think there is education tobe had for other vendors.
There's common misconceptions,especially for operators in regards
to what this technology is andwhat it isn't.
So I'm very interested to hearyour insight on that.
Yeah, so there's, there's acouple roads to think about when

(21:09):
starting to understand whatyou might want to utilize from new
technology perspective when itcomes to machine learning.
And I like to, I like tosimplify it as much as possible.
And we can say it's machinelearning, which it is, but then we
can also say it's automation,which it is, and therefore it's just
technology.
So it's really just what newautomation technology do I want to

(21:33):
bring into my restaurant.
So as an example, do I wantautomation that answers the phone
for me and indicates tosomeone whether they want to know
what my business address is orwhat my telephone number is, or what
directions to my business, orto be able to make a reservation
or maybe even to troubleshoota problematic scenario where it can

(21:57):
feed me to a pipeline of thisperson's voicemail or this person's
voicemail.
That's ML, that's automation.
That's technology.
That's one utilization.
Same type of ML can also beused to help people place orders.
So if you've been toMcDonald's lately, you might have

(22:17):
noticed that there is a voicethat answers when you pull up and
then there's a different voicethat takes your order.
So there's actually an MLthat's asking you if you want to
utilize your telephone numberto be able to be recognized, or if
you want to use the customapp, or if you've placed an order
with the custom app.

(22:37):
There's all these different ABtesting that's happening.
And then after you answer thatorder, which is being asked by ML
and answer is being logged byML and a decision tree being utilized
by ML, then a human beingcomes in.
And what does that allow for?
That allows for one humanbeing to be taking someone in order
and ML to get started on theprocess for the next person.

(22:58):
So what's that?
It's automation.
What's automation technology?
So it's ML, sure, but it'sautomation technology.
And that's simple and that's easy.
That's not anything thatanybody should be scared of or hesitant
of.
So that's another thing that Ithink that people can look to.
How do you see ML furtherrevolutionizing the restaurant and
hospitality industries overthe next five to 10 years?

(23:19):
And will we see the true AI inthat time period?
So that to me breaks down tothe definition of hospitality.
So some people think thedefinition, definition of hospitality
is the function of gettingserved, and some people think it's
the service, and then somepeople think it's both.

(23:40):
So if I go to a hotel and Ican check in and sleep in that room,
then I've gotten hospitality.
And some people think it's thelevel of service that you get, and
then some people think it'sthe marriage of those two.
That's a very good pointbecause a lot of people view it as
the touch much points andwhich with that, which adds to the
value add of the proposition.
Right?
So if I can call a restaurantand an ML can Answer the phone and

(24:05):
answer my questions and I canmake a reservation or change a reservation
or get a question answered,then that's hospitality.
But if I really enjoy thatexperience, if that ML sounds like
I want it to sound and itmakes me feel good about having done
it and I feel good leaving it,then that's an elevated level of
hospitality.
So the way I want to answerthe question is can we first hit

(24:28):
function and then can we getto form?
And if we can get form andfunction, then that's the perfect
marriage of hospitality for most.
I think some people are stillgoing to say I prefer that, that
human element and aspect.
And I'm, I'm, I want to havemy cake and eat it too.
I want to live in a worldwhere both exist for forever, hopefully,

(24:48):
knock on wood.
Some people want just one,some people just want the other.
But I think people figuringout what their druthers are on that,
on that answer is going to bean interesting thing that we're going
to see develop over the nextfive and 10 years is what does hospitality
mean to you when it comes tosomething that's been programmed
by a human?
Right, because it's not arobot you're talking to, it's a human

(25:10):
being that programmed a voice.
It's maybe a voice recordingand it's maybe a person that wrote
that code.
So underlying there is a humanat work, at play there.
So it's still going to beabout human connectivity and human
emotionality.
Well, with these advancementsin these technologies, do you see
automation in this areapotentially replacing traditional
roles in restaurants or hotels?

(25:32):
Or is it more about augmentingthe human decision making process?
That's the million dollarquestion right there.
It might be the trilliondollar question.
Yeah, exactly.
I was listening to aninterview by Mark Andreessen at Andreessen
Horowitz this morning and hethat the, the new ML that's going

(25:53):
to be coming out is going tobe able to decide what is the next
best solution to code on itsown and then we'll be able to execute
on that.
And in that occurring, thatwhole process that you just talked
about where there is no humanintervention at a certain point is

(26:15):
coming the next six to 12 months.
And as that's the case, I'mnot certain who's going to outperform.
I want to say that the humanbeing is still going to be able to
outperform and is still goingto be able to have more human empathy
and more human sympathy justby simple definition.
But I'm not, I'm not rulingout the option or the possibility

(26:40):
that the computer is going tobe able to confuse the human and
we're going to lose, we'regoing to lose an ability to be able
to differentiate between the two.
And then when we hit there, ofcourse we're right back up to the
initial part of this conversation.
True AI when we can no longer tell.
Right.
That's the touring test.
So many restaurant operatorsmay not be tech savvy and it sounds
like you've built a lot ofthis technology that we just talked

(27:02):
about into five outs offeringwhich, you know, how do you make,
how do you make these toolsintuitive and user friendly to the
operator?
I think there's nothing moreimportant than that, frankly.
Simplicity is key.
And I also think that insightsare a bad word and I think that helping

(27:29):
is a bad word.
And I think that those thingsare currently where a lot of people
are focused on and we're past that.
I think everybody should bepast that to a point of giving something
in its final state or nearfinal state, it has to be delivered
as an answer and we need toactually take it beyond that and

(27:53):
get to what we internally calla closed loop system.
So as an example, you mightsend an email to someone and say,
hey, I want to do this thingat work and I'm putting you on notice
on Monday that if I don't getan answer back from you about Wednesday,
I'm going to do that thing.

(28:13):
And so what we would like todo in our areas of focus and what
I think other people are goingto do is start creating similar closed
loop systems where I'mactually going to create a schedule,
a labor schedule for arestaurant and I'm going to send
it to an operator, let's sayon a Monday and say I'm giving you
until Wednesday to edit,change, opine, pull back, what have

(28:39):
you.
But after that, if you haven'tmade any of those changes or edits,
I'm actually going todistribute that to your team and
then those folks are the folksthat are going to actually show up.
And we're doing the same thingwith purchase orders and prep lists.
So I want to create anautomated prep list which is how
much food do I need to makefor tomorrow?
I want to email that to theteam in the morning, first thing

(29:00):
and prior to I want to send itto whoever's the decision maker at
the night time and say, here'sthe prep list the teams are going
to get tomorrow, make edits or changes.
And if you don't by this timeat 6am they're going to have those
in their inbox.
Same thing for the vendor.
I'm going to put together yourmeat order, your dairy order, your
fish order, your beverageorder and I'm going to send it to
you and I'm going to say youhave this amount of time to apply
and if you don't reply at achange or otherwise, I'm going to

(29:22):
send it to your vendor, whichmeans the next day that's the food
and that's the beverage that'sgoing to show up at your front door.
So that's where we need to getto, is not only well past insights
and well past help andguidance, we got to get to a delivered
answer, a finite, explicitdelivered answer and we actually
need to ship that answer tothe next person so that we're not

(29:43):
bottlenecking decision making,we're not bottlenecking on execution
and that we're moving at speed.
Well, that was really my nextquestion, but I think that's your,
you've answered it and it wasreally around, you know, how do you
balance the complexity ofthese data driven insights with the
simplicity that restaurantoperators often need in their day
to day tools?
And it's really just handingthem the results of what they're

(30:06):
looking for, what they need,not necessarily what they're looking
for.
It's absolutely similar asopening up a new computer.
So if you go get a new Applelaptop, you open the box, you take
it out of the plastic, youopen it, you turn it on and that's
it and it runs from there.
There are so many complexsystems running underneath the surface

(30:30):
that none of it, well, most Idon't and a lot of people don't understand.
There are many out there thatdo, but most don't.
But it just works just like acar, which I also don't understand
how a combustion engine works,but it works and I go in and I press
the on button and the carworks and it runs.
That's the level that I thinkthat the restaurant industry and
probably other people as well,hospitality inclusive, need to be

(30:54):
thinking about machine learning.
It has to be.
The end result has to beautomated, deliverable action.
And anything else I think is afailure of our own ability to recognize
what works as a solution inthe hospitality industry.
I don't think insights work, Idon't think data works, I don't think

(31:17):
help works, I don't thinkguidance works, I think delivered
actions are it.
And that's anything else Ithink is short sighted and it's it's
doesn't have legs.
Brandon, in closing, if youcould give one piece of advice to
restaurant operators,navigating, and I'm going to say
the word the AI landscape,what would it be and what would you

(31:39):
caution them against?
Now, obviously there'squestions they should be asking based
upon our conversation today.
You know, I've said for years,a lot of smoke and mirrors out there
when it comes to AI, a lot ofwizard of Ozing.
And I think you have providedsome very interesting insight today.
Like I said, it's going tocause some controversy and I can't

(32:00):
wait to see the post on thiswhen we, when we go live with the
show.
But what is your advice forthose operators looking to do go
down this path?
I would, I would advise peopleto ask anyone that they're speaking
to who is offering up AI andML solutions to show them first and

(32:22):
foremost what the deliverableis, not to start at the beginning
of the story and say, why didyou create this?
What does it do, how does itwork, who does it work for, et cetera.
I would go to the very end ofthe line and say, what does the deliverable
look like?
Show me the deliverable.
And then if the deliverablelooks like it's something that your

(32:42):
team can use and that youcould put it into your work process
and you can digest it and thenit's going to be helpful for you,
then I would start working backwards.
Where did this informationcome from?
How did you get to it?
What is the, what is the process?
Who else have you used it for?
That's how I would ask peopleto go about evaluating a solution.

(33:04):
Show me the end deliverableand look at it and say, could my,
and would my team use this toexecute in their restaurants?
And if the answer is yes, thenstart working your way backwards
to the rest of that journey.
Brandon, I am so glad Ibrought you on the show.
I know we had a littlescheduling issues, but this was a

(33:26):
very worthwhile conversation.
I love your insight, I loveyour perspective and I am so appreciative
that you came on the show totalk to our audience here today all
about what is not AI, but moreor less it's machine learning.
Great to do it.
Happy to be here and thank youfor the invitation.

(33:48):
Boy, that was interesting.
I bet you might have a fewquestions or a few comments and I
would love to hear yourthoughts on this and continue the
conversation.
If you'd like to do so, youcan leave a message on my LinkedIn
profile.
You can always reach out to mevia Everything Social, Skip Kimple
or everything magicgatetech.

(34:09):
This includes X, Facebook,Instagram and LinkedIn.
You can always go to thewebsite skipkimple.com for all of
the archived shows includingthe show notes which are posted there
as well.
You can also hear these newepisodes on the Magic gate website@magicgate.com
and of course you can email meat any point in time@Skipagicgate.com

(34:32):
next week on the Tech Chef KenGarvanovits joins us to talk about
transformative growth andleadership for high impact hospitality
executives.
He is all about drivinginnovation, creating scalable growth
and creating team excellence.
Known for his Amazon bestselling book entitled Business Breakthrough

(34:53):
3.0 and his skills at publicspeaking, Ken shows us some eye opening
opportunities to fine tuneyour leadership style.
Well, here is to wishing you agood productive week ahead as we
push forward to conquerJanuary together.
So until next week, stay safe,stay healthy and stay hungry.

(35:17):
My friends.

All Episodes

Takeaways:

Links Referenced in This Episode:

How To Contact Me

Episode Transcript

Popular Podcasts

Stuff You Should Know

Dateline NBC

On Purpose with Jay Shetty

.css-15opob5{left:0;position:absolute;top:0.8rem;} All Episodes

.css-14f5ked{margin:0;word-break:break-word;display:-webkit-box;-webkit-box-orient:vertical;box-orient:vertical;-webkit-line-clamp:2;overflow:hidden;}TCP084: "AI Does Not Exist" with Brandon McRill, CEO of 5Out