Latest / The GrowthPro Edge / AI with Guardrails | The GrowthPro Edge

Transcript

AI with Guardrails | The GrowthPro Edge · The GrowthPro Edge

• 0:24
  Denise Killian: Hello, and welcome to this week's episode of The Growth Pro Edge.
• 0:29
  I'm Denise Killian with the GrowthPro Group and TrainerTainment, and I'm
• 0:32
  thrilled that you're joining us this week.
• 0:36
  Before we get started, I do wanna have the opportunity to thank our 2026 sponsors.
• 0:43
  Without them, we couldn't do what we do, which is bring you a new episode
• 0:48
  of the Growth Pro Edge each and every.
• 0:50
  Week.
• 0:52
  Well, here at the GrowthPro Group, we are all about growth, and that only
• 0:56
  happens when we take a break from the doing and spend just a little
• 1:00
  bit of time thinking, and that's where the Growth Pro Edge comes in.
• 1:04
  Each week we'll be talking about issues that impact our growth for our teams.
• 1:09
  For our businesses and for ourselves.
• 1:13
  This week we're gonna be speaking with Todd Smith about a topic that
• 1:17
  I think is so incredibly important.
• 1:21
  It's about AI and how you're using AI in your company.
• 1:27
  If you're thinking about getting started this year, you wanna watch this episode
• 1:30
  first because there's some parameters and some strategies that you can
• 1:35
  deploy right now from the beginning to create a successful transition
• 1:40
  to using AI within your business.
• 1:44
  So let's go ahead and get started and hear from Todd Smith.
• 1:50
  Todd Smith, welcome to the Growth Pro Edge.
• 1:54
  We're so excited to have you.
• 1:55
  Todd Smith: Well, thank you so much for having me.
• 1:57
  I appreciate it.
• 1:58
  It's good to be here.
• 1:59
  Denise Killian: Well thrilled to talk about this because I think it's
• 2:02
  something that is so important when we talk about AI and how we're using ai.
• 2:08
  So you're the first person I kind of thought about to have this conversation
• 2:12
  with, so I appreciate you being here.
• 2:14
  But before we get started, uh, why don't you tell us a little bit about yourself
• 2:18
  and what your connection to AI has been.
• 2:22
  Todd Smith: Sure.
• 2:22
  Uh, well, my background, uh, is in filmmaking and I've been doing
• 2:27
  that in different app capacities for the past 20 some odd years.
• 2:32
  And about two and a half, three years ago, I moved over into the AI space and
• 2:37
  since then I've kind of seen how the sausage was made, uh, on a couple of
• 2:42
  different fronts with it's, uh, working with, um, several different companies.
• 2:46
  So.
• 2:48
  Denise Killian: So when we talk about that sausage getting made, I think, you
• 2:51
  know, you had the opportunity to jump in at a very, very early stage in a
• 2:56
  formative stage with AI and working in that background of helping to cultivate.
• 3:04
  Ai, really.
• 3:06
  And so I think that this is the great conversation to have as we begin this
• 3:11
  new year and companies are thinking about their goals for the year and
• 3:15
  what they want to do this year.
• 3:17
  I can guarantee you a lot of conversations right now are
• 3:21
  around how are we going use ai?
• 3:24
  What should we do with ai?
• 3:25
  What can we do?
• 3:27
  And it's a, it's such a broad question to ask.
• 3:31
  Ask.
• 3:31
  Mm-hmm.
• 3:32
  That it can be really hard to figure out what those answers are.
• 3:35
  And so one of the things that I wanted to talk to you about today
• 3:39
  is when we look at, when you're having those conversations, now is
• 3:45
  the time to create some foundations.
• 3:49
  Of governances, if you will, of how you're going to use ai.
• 3:54
  And you and I were talking and you've got an amazing way to really help
• 4:00
  you start to frame out how you're going to use AI and beautifully.
• 4:05
  It is called ethics.
• 4:07
  Todd Smith: Sure.
• 4:08
  Yeah.
• 4:08
  Um, well, I think maybe the first step that an organization should look at
• 4:12
  when, when they're looking at integrating AI is maybe looking at it as you're
• 4:17
  bringing on a new employee, right?
• 4:19
  So you're asking this employee to do a whole lot of things for you.
• 4:23
  And so with that organization should probably ask themselves
• 4:27
  basically three different questions.
• 4:30
  Uh, you know, what is the purpose?
• 4:32
  Of your ai, what are you going to be using this for?
• 4:34
  And, and to what end?
• 4:36
  And then finally, um, as you're going through this process,
• 4:41
  for whom is this going to fail?
• 4:43
  Which is, is it going to be the organization?
• 4:45
  Is it going to be a client?
• 4:47
  Is it going to be a much greater capacity?
• 4:49
  So when you're incorporating the ethics into your, your ai, um,
• 4:55
  deployment, you know, you can start with, with the acronym of ethics and,
• 5:00
  and start with e start at the top.
• 5:01
  Start with your execs or your board members.
• 5:04
  Um, as we go in, you know, we have the, we have our technologists.
• 5:08
  There are going to be the people that are actually building this for you.
• 5:12
  Uh, after that you have your, your hr, what kind of data and security are you
• 5:17
  going to, um, incorporate into your ai?
• 5:20
  Um, obviously you're gonna wanna bring in industry experts and see
• 5:25
  how this is going to really, um.
• 5:29
  Be built effectively and more efficiently for your organization.
• 5:33
  Um, you wanna obviously incorporate your customer, your user client.
• 5:37
  Uh, and then finally, what is going to be the, um, the societal
• 5:41
  aspect of this, uh, on an overall arching, uh, question right there.
• 5:49
  Denise Killian: So I wanna start with the, what is the purpose?
• 5:52
  'cause I think that is a very important conversation to have.
• 5:57
  Just as, and I love the analogy of if you're bringing on a new employee, think
• 6:01
  of it as you're creating a new position.
• 6:04
  Mm-hmm.
• 6:04
  In your company.
• 6:05
  What are the roles.
• 6:07
  What are the specific things that this employee would be responsible for?
• 6:13
  Who do they report to?
• 6:14
  Right?
• 6:15
  All of those things that you would do if you're creating a new position and
• 6:22
  building a job description, right?
• 6:24
  You should do that for ai.
• 6:26
  I think that's such an interesting way to look at it, of really nailing down.
• 6:31
  What should it be now?
• 6:32
  Should you start small or should you just start?
• 6:35
  Huge.
• 6:36
  What would be your suggestion there?
• 6:39
  Todd Smith: Uh, well, I think an organization really has to look at their
• 6:43
  needs, you know, and, and kind of, as you say, build on a foundation of this.
• 6:48
  So maybe is it, if you're doing something that is agentic ai,
• 6:52
  you know, if you wanna look at something like that, that's kind of.
• 6:55
  The way that I've viewed this is it's like a project manager.
• 6:59
  So you have all of the agents that are responsible for maybe looking at your
• 7:04
  calendar or scheduling or summarizing emails, and then your agentic AI is your
• 7:10
  overall project manager, which is going to start deploying all of this information
• 7:15
  and maybe developing a marketing plan.
• 7:18
  So who is gonna be responsible for overseeing the project manager?
• 7:23
  You wanna also, you know, make sure that you have humans that are overseeing the
• 7:28
  end process here to double check and make sure you are on the right process.
• 7:33
  So it's all up to the organization to decide how big
• 7:36
  or how small they want to go.
• 7:39
  Denise Killian: Also another thing that, uh, I want to kind of point out too is
• 7:44
  make sure that you are using a service no matter which of the, at this point,
• 7:51
  dozens and dozens of different ais that are out there, that you're using a
• 7:57
  service that is protected, meaning you're not using open, where you're releasing
• 8:03
  all of your information out into the world that you're using the protected.
• 8:08
  Measurements, right?
• 8:10
  Todd Smith: Yeah, absolutely.
• 8:11
  Go ahead and, and pay for the service.
• 8:14
  Because if you are using for, if you are using an an open source network,
• 8:19
  then your IP is then being used to train other models and you are also.
• 8:27
  Leaving yourself open for any kind of data breaches and security, security
• 8:31
  hacks that could come down the line.
• 8:32
  So if you are using something that is proprietary, you are more able to, um,
• 8:38
  lock down your parameters and, and.
• 8:42
  Also add more value for your client and be able to tell your client, this
• 8:47
  is exactly how we are doing this.
• 8:49
  That's, that's a big part of your ethics as well, is being transparent
• 8:53
  and being able to explain in layman's terms to your client what you're
• 8:59
  doing and how you're doing it.
• 9:01
  Denise Killian: Absolutely.
• 9:02
  Uh, so you've got it locked down.
• 9:05
  You've got a service, you've picked a service, you're paying for it.
• 9:08
  So that your proprietary information, whether it's your financials,
• 9:12
  uh, or any other data that you wanna protect is protected.
• 9:17
  You have created.
• 9:18
  A job description as it were for how we're gonna use this ai, the specifics of what
• 9:25
  you're gonna be doing with it and who is going to be the person or the people
• 9:30
  who are responsible for working with it.
• 9:33
  Todd Smith: Mm-hmm.
• 9:33
  Which
• 9:34
  Denise Killian: are great.
• 9:35
  You're not done there.
• 9:36
  And that's where I really wanna get back into the ethics
• 9:41
  that you were talking about.
• 9:42
  And the first one, of course you mentioned the e is for your
• 9:45
  executives, your, your shareholders.
• 9:48
  Bringing them to the table.
• 9:50
  What does that discussion look like?
• 9:52
  Todd Smith: That discussion really starts at high level and, and is coming
• 9:57
  up with what is our end goal for this?
• 10:00
  And so once you have that end goal in mind, then you, as the executives can
• 10:05
  maybe kind of work backwards and say, okay, let's figure out what our, what
• 10:10
  our deployment is going to be, what it's going to look like, and then start putting
• 10:14
  the people in place to make it happen.
• 10:17
  Mm-hmm.
• 10:18
  Denise Killian: Fantastic.
• 10:19
  So you've got your execs on board, you've got your, if you have
• 10:22
  shareholders, your board members, right?
• 10:24
  Uh, those people on board.
• 10:26
  And we all agree this is the path that we're moving forward with.
• 10:30
  And, and some of those parameters, uh, the, the T of course stands for tech.
• 10:35
  You gotta bring your tech people in, right?
• 10:37
  People.
• 10:38
  Todd Smith: Yeah, absolutely.
• 10:39
  And make sure that you are bringing in the right people that
• 10:42
  can do the right thing for you.
• 10:44
  Um, you know, you're.
• 10:46
  As, as quickly as AI is progressing, um, younger generations are really
• 10:53
  picking up on it quicker, uh, maybe than older generations, but at the
• 10:58
  same time, there is still institutional knowledge that the older generations
• 11:01
  have that they can bring along with this.
• 11:03
  So I guess.
• 11:05
  Make sure that you have industry professionals that are working on this and
• 11:09
  you don't necessarily want to just have a team of interns that is responsible for
• 11:15
  all of your, your data and your client's data, and building such a very important
• 11:20
  part of, of your business at this point.
• 11:22
  Mm-hmm.
• 11:25
  Denise Killian: The H of course stands for not just bring in your HR people,
• 11:30
  but really this is more of the broader kind of a human rights advocate here.
• 11:35
  Right?
• 11:36
  Todd Smith: Absolutely.
• 11:37
  You wanna make sure that your, your AI is not biased or not
• 11:41
  projecting some sort of bias.
• 11:43
  You wanna make sure I keep coming back to data and security.
• 11:46
  Making sure that your employee's data, your client's data is also secured.
• 11:51
  So making sure that there is bias avoidance in this is a real high priority.
• 11:58
  Denise Killian: And the the bias part, you know, we've all seen.
• 12:02
  Things that have happened when AI is left unchecked, and this is really
• 12:07
  what this conversation is about.
• 12:08
  How can you harness AI into a tool that's going to be beneficial to your
• 12:13
  company as opposed to doing something inadvertently that you didn't intend to
• 12:18
  do, uh, without those parameters that can easily happen where AI is presenting
• 12:24
  information that it's gathered, that is in fact bias or that might be inaccurate.
• 12:31
  Uh, or might do something that is illegal.
• 12:34
  Todd Smith: Mm-hmm.
• 12:35
  Denise Killian: God.
• 12:37
  Todd Smith: Yeah.
• 12:37
  And or also if it is, um, telling an end user to do something that is harmful
• 12:43
  either to themselves or to society.
• 12:46
  So you really want to make sure that you are, again, keeping a human in
• 12:50
  the loop that is, that is being able to oversee your AI and also being
• 12:56
  able to objectively audit yourself.
• 13:00
  Is the AI doing these sort of things and then making the
• 13:03
  fixes as quickly as possible.
• 13:05
  So you have to be flexible and you have to be nimble.
• 13:08
  Denise Killian: And this is particularly important when a company is perhaps
• 13:13
  thinking about using AI to be an automated response to customers or consumers.
• 13:19
  Mm-hmm.
• 13:20
  The chat bots or the back and forth email responses, or even if you have an an AI
• 13:25
  agent that is having those conversations potentially with customers, uh, that's
• 13:32
  where this oversight becomes really.
• 13:36
  Really essential.
• 13:39
  Todd Smith: Absolutely.
• 13:40
  Uh, and that comes down to kind of our, our last two, uh, of, of
• 13:45
  our ethics, um, uh, anagram here is, is our customer and our user.
• 13:50
  So we wanna make sure that our customer is, um, being taken care of
• 13:56
  in the most ethical way possible and is provided the correct information.
• 14:04
  And then that's.
• 14:07
  As you're talking about the chat bot, what is the overall impact on society here?
• 14:11
  Because the chat bots are, are interacting with so many users.
• 14:15
  If they're giving false information or harmful information, what
• 14:19
  is the end, end impact of that?
• 14:22
  Denise Killian: Mm-hmm.
• 14:23
  Right, and the.
• 14:25
  The clear override here is that, hey, if you're thinking about we can use ai,
• 14:32
  we can get rid of a ton of people and just let AI do all of this unchecked.
• 14:38
  That's the wrong answer,
• 14:40
  Todd Smith: correct.
• 14:42
  Yeah, absolutely.
• 14:43
  Absolutely.
• 14:44
  We want to make sure that we are on top of our ai.
• 14:47
  Uh, it can do really fabulous things for an organization, but at the end of the
• 14:52
  day, you need people to oversee your ai.
• 14:57
  Denise Killian: Right?
• 14:57
  Absolutely.
• 14:59
  So the, uh, I'm, I'm loving.
• 15:01
  It's very simple.
• 15:02
  It's clear ethics, right?
• 15:03
  That's a great way to start those conversations under that framework.
• 15:09
  Of making sure that it is a tool that is beneficial, not detrimental.
• 15:16
  And that AI does not mean replacing all people.
• 15:21
  Todd Smith: Correct?
• 15:22
  Correct.
• 15:23
  You know, it is another employee that you're bringing into your
• 15:25
  organization and organizations don't just leave employees unchecked.
• 15:32
  They're, they have supervisors.
• 15:33
  So, you know, as you're implementing AI into your pipeline and into your
• 15:38
  organization, just keep that in mind that it is not a, it, it is not a, a cure all.
• 15:44
  It's not going to do everything for you.
• 15:48
  It's a part of your organization.
• 15:51
  Denise Killian: Right?
• 15:51
  Absolutely.
• 15:53
  Uh, so in your experience of, you know, you've been for three years, you've
• 15:58
  been working on that backend part.
• 16:02
  And you mentioned before how quickly AI is changing and adapting.
• 16:09
  In your reflection of the past three years of working on that backend
• 16:13
  part, uh, I'm loving this learning of these parameters, which I think is so
• 16:18
  fantastic and applicable to anybody who's looking to use AI in their business.
• 16:24
  Uh, but talking about how quickly it progresses, what has been your biggest
• 16:30
  sort of aha or surprising experience of how quickly it has adapted?
• 16:37
  Todd Smith: Uh, well, I look at, I look at Will Smith eating spaghetti.
• 16:42
  You know, that is one of the first generative AI examples that we had,
• 16:47
  um, two and a half years ago or so.
• 16:49
  And I look at how far generative AI has, has just progressed by leaps and bounds.
• 16:56
  And I, and I really look at that, you know, from my filmmaking background.
• 17:00
  I look at what, what is going to be the impact of AI on, on,
• 17:04
  you know, the movie and, and.
• 17:06
  TV industry specifically, but I just see how far it has come in
• 17:11
  being able to, to create all these fantastic worlds and characters and
• 17:16
  being able to implement a whole lot of things together to make a story.
• 17:21
  That's what I've, that's what I have seen so much of in the past few
• 17:25
  years that I'm just blown away by.
• 17:27
  Mm-hmm.
• 17:29
  Denise Killian: And, and the projection into the future is.
• 17:34
  Kind of unknown, I guess.
• 17:36
  Right?
• 17:36
  Where is it gonna land?
• 17:39
  Todd Smith: Yeah.
• 17:40
  Uh, it really is.
• 17:41
  It, you know, there needs, I personally feel there should be
• 17:45
  guardrails put in place, uh, for the uses of AI to keep us all safe, to
• 17:51
  keep us grounded, and to be able to utilize the tools that AI has, has a.
• 17:59
  Has developed, the AI companies have developed over the past few years.
• 18:04
  Um,
• 18:06
  without guardrails and going unchecked, there is potential
• 18:10
  for very negative consequences.
• 18:13
  So yes, we do have a lot of potential, but I think that we need
• 18:17
  to have the guardrails in place to, to help guide us down the way.
• 18:21
  Denise Killian: Yeah.
• 18:21
  And that it's an enhancement, not a replacement.
• 18:25
  Todd Smith: Exactly.
• 18:26
  It's another tool in the toolbox.
• 18:28
  Denise Killian: Yeah.
• 18:29
  Okay, Todd.
• 18:30
  Unfortunately I could talk to you about this all day.
• 18:33
  Uh, but unfortunately we are running a little bit short on time, but I am
• 18:37
  so thrilled for this conversation and I think it's a great starting point
• 18:41
  for a lot of folks as they're looking to integrate AI into their business.
• 18:46
  Uh, but before we leave, um, you know, here at the GrowthPro
• 18:49
  Group, we are all about growth.
• 18:54
  So I wanna ask you, what do you do?
• 18:59
  To grow yourself grow,
• 19:02
  Todd Smith: um, I should do a couple of things.
• 19:04
  Um, one, I feed my creativity.
• 19:07
  I do a lot of writing.
• 19:09
  I still do, um, screenplay writing, um, and I still edit
• 19:14
  videos and, and stuff like that.
• 19:15
  I, I nourish that creative aspect of me.
• 19:18
  Uh, and I've also been doing a program called Positive Intelligence.
• 19:22
  And so that is really about being in the moment, staying grounded and
• 19:27
  being appreciative for all the little things, and to have a sage mindset.
• 19:34
  Denise Killian: That is great and, and you know, positive intelligence.
• 19:37
  That is something that, uh, the mindfulness of it, it's amazing
• 19:43
  how quickly it can help shift your mindset a little bit.
• 19:47
  Todd Smith: Oh, yes, absolutely.
• 19:49
  Denise Killian: Yeah.
• 19:49
  I love that.
• 19:50
  I love that.
• 19:50
  Thank you for sharing that with us.
• 19:52
  And Todd, again, thank you so much for being here with us today, and
• 19:57
  what a great thought provoking conversation that this was.
• 20:01
  So thank you so much.
• 20:02
  Todd Smith: Thanks so much for having me, Denise.
• 20:04
  I appreciate it.
• 20:04
  It's a pleasure being here.
• 20:07
  Denise Killian: Well, I don't know about you, but I have a lot to think on from
• 20:11
  that conversation, and I really just love the foundation of creating a system.
• 20:19
  From the start with your company, with your executives, with your tech
• 20:24
  leaders, with your department heads, thinking about the customer and the
• 20:29
  impact and exactly how you're going to be using AI and using the anagram of
• 20:35
  ethics is just a perfect way to do that.
• 20:38
  So thank you so much for Todd and what his thoughts are there.
• 20:42
  And you know, again.
• 20:45
  Thinking of starting to use AI as a, you're creating a new position and the
• 20:53
  same conversations that you'd be having with your teams about what is this
• 20:58
  position, who does this person report to?
• 21:01
  What are the job descriptions?
• 21:03
  What are the requirements?
• 21:05
  I think that is a great way to think about how we're gonna use ai.
• 21:09
  So whether it's.
• 21:11
  You're looking to incorporate AI into your sales team, into your customer
• 21:15
  response departments, your customer and consumer outreach, your marketing,
• 21:21
  uh, or your financial analytics.
• 21:24
  That's a great way to kind of frame how you're gonna be using AI to
• 21:28
  start thinking about it in that way.
• 21:29
  So thank you so much for Todd and your insight and your expertise.
• 21:34
  Appreciate that.
• 21:36
  And again, before we leave, I wanna thank our 2026 digital sponsors.
• 21:40
  We appreciate what they do to help us bring you the Growth
• 21:43
  Pro Edge each and every week.
• 21:47
  If you have an idea for the Growth Pro Edge, or you'd like to be a
• 21:51
  guest, I would love to hear from you.
• 21:53
  You can reach me at.
• 21:55
  At the GrowthPro Group.com.
• 21:59
  Well, thanks for tuning into today's episode of The Growth Pro Edge, and
• 22:03
  remember that the Journey to Success is all about growth, and we at the
• 22:08
  GrowthPro Group are committed to helping you and your company thrive.
• 22:13
  You can visit the GrowthPro Group.com to learn how we can support your
• 22:17
  growth journey for yourself.
• 22:19
  For your team or your business by setting up a free growth
• 22:22
  call with one of our coaches.
• 22:24
  Until next time, keep learning, keep growing, and keep
• 22:29
  striving for the extraordinary.