Latest / The Prompt / Ep. 589 | Leading through the AI shift: Why generative AI demands new thinking, not just new tools

Transcript

Ep. 589 | Leading through the AI shift: Why generative AI demands new thinking, not just new tools · The Prompt

• 0:01
  This is on base your one stop shop for solving your most
• 0:04
  pressing B2B go to market challenges.
• 0:07
  Each week, host Chris Moody and Paul Gibson talk to sales and
• 0:09
  marketing leaders to get you in depth insights and creative
• 0:12
  strategies to set your revenue teams up for success.
• 0:18
  Well, hello everyone, and welcome back to the latest in
• 0:21
  the award-winning on Based podcast series.
• 0:23
  I'm your host, Paul Gibson. And as you know, we always look
• 0:26
  for hot topics to discuss and this is one that's the hottest
• 0:30
  of them all. And and with that, I'm excited
• 0:32
  that today I'm talking with Jim Stern.
• 0:34
  He's author, keynote speaker, and I think probably well known
• 0:38
  to a lot of people on this call. And we're going to discuss
• 0:42
  leading through the AI shift, which I think is a really
• 0:45
  important topic right now. Everyone's talking AI.
• 0:48
  So really excited, Jim, pleased to get you on the show and
• 0:51
  thanks for coming. Thank you very much.
• 0:53
  It's an honor to be here and I'm very excited about this topic in
• 0:57
  general and I'm really looking forward to your questions.
• 1:00
  So before we get into the topic, today's podcast, I always like
• 1:04
  to start with a very simple question.
• 1:06
  Jim, do you mind sharing your journey to where you are today
• 1:09
  and what particular has LED you to focus this session on leading
• 1:13
  through the AI shift? So I've always been at the
• 1:17
  forefront of helping companies use technology.
• 1:21
  I sold business computers to companies that had never owned
• 1:24
  one before. I sold software development
• 1:26
  tools to enterprise and government for COBOL and
• 1:30
  Fortran. Yes, I am that old.
• 1:33
  In 1995, wrote my first book about online marketing because
• 1:37
  the Internet was something nobody understood.
• 1:39
  I have written books on social media.
• 1:42
  I've written the last two books on artificial intelligence.
• 1:45
  I got there through web analytics.
• 1:49
  I had been advising lots of clients about their websites and
• 1:52
  how ugly they were said wait a minute, we can measure how bad
• 1:56
  your website is and we can measure how much better it is
• 1:59
  when you make changes. We can do testing.
• 2:01
  So analytics led to statistics that led to machine learning.
• 2:05
  That was my book eight years ago and then machine learning.
• 2:09
  Now we've got generative AI. So my book from 2025 is the new
• 2:14
  Science of customer Relationships.
• 2:16
  So being what a professional explainer has always been my
• 2:21
  realm and helping companies understand how the technology
• 2:25
  can help them with communicating with their customers.
• 2:29
  Yeah, makes a lot of sense. Thank you.
• 2:30
  And it's interesting that that website analytics is a is
• 2:33
  obviously a part of what our business demand based does.
• 2:36
  And it's amazing how many people ignore that.
• 2:38
  They'll throw lots of money at advertising trying to get people
• 2:41
  to their website. They don't know who's getting to
• 2:43
  their website. And if they do, they don't know
• 2:45
  what they're seeing. They don't know if it's the
• 2:46
  right quality. So yeah, I mean, it sounds like
• 2:48
  a very basic thing these days, but there's so many people that
• 2:51
  don't do it do that at all. So a very interesting start.
• 2:55
  I just started and for for this particular topic today, how do
• 2:58
  you think generative AI is redefining that relationship
• 3:02
  between technology, people and indeed the decision making
• 3:06
  inside a big enterprise? The most important thing to
• 3:09
  understand first is that this is not computing as you know it.
• 3:14
  The computers that we are used to come all the way, go all the
• 3:17
  way back to the Abacus. They add, subtract, multiply,
• 3:20
  divide. They are databases.
• 3:22
  They're search engines. They our, they are
• 3:25
  deterministic. Ask a question, get an answer if
• 3:28
  the computer said so, therefore it's right.
• 3:31
  Now we know that statistics allow you to create any reality
• 3:35
  you want. So that makes it difficult when
• 3:38
  we move from deterministic to probabilistic, which is machine
• 3:43
  learning. So machine learning says, well
• 3:46
  generally the there you've got 5 segments of customers and these
• 3:51
  customers act this way and those customers act that way.
• 3:54
  Incredibly useful. Predictive analytics, incredibly
• 3:58
  useful. But it is only useful when you
• 4:01
  have a tolerance for uncertainty.
• 4:04
  When you're running payroll, no uncertainty, you're doing self
• 4:07
  driving cars, no uncertainty. When you're doing lead scoring
• 4:12
  or putting the right message in front of the right person at the
• 4:15
  right time, there's no actual answer.
• 4:17
  There's just better or worse. So it's great for that.
• 4:20
  Now we're moving off of those completely into this generative
• 4:24
  AI. We've gone from deterministic to
• 4:26
  probabilistic to artistic. It generates, that's its job is
• 4:31
  to invent, and it doesn't always tell the truth.
• 4:36
  And the technical term for that is hallucination.
• 4:38
  So we are having to relearn that the computer is about as
• 4:43
  reliable as the next human over who will make a decision and
• 4:49
  then rationalize it, maybe based on data, maybe not.
• 4:54
  It's up to you to confirm and verify.
• 4:56
  That's really hard for us to get accustomed to.
• 4:59
  That's really interesting. And we, we have lots of
• 5:01
  conversations and slightly off topic, but similar really
• 5:03
  around, you know, and AI will tell a salesperson or a sales
• 5:08
  development Rep or whatever you like to call them, go after that
• 5:12
  because we say so without anything behind it.
• 5:14
  I call it AI without the Y. And I think that's sort of what
• 5:18
  you're touching on here is that there is an element of trust
• 5:21
  going on here. But if you can get insight
• 5:23
  behind why the AI is telling you to do that as a salesperson, you
• 5:27
  certainly get a lot more confidence than just someone
• 5:30
  saying go after that because I say so, which I think is termed
• 5:33
  the black box scenario sort of thing.
• 5:36
  And that gets into some interesting nuance, because you
• 5:40
  can absolutely ask a large language model why it decided to
• 5:44
  tell you what it told you, and it will invent a reason, just
• 5:50
  like humans. So I need to buy a red sports
• 5:53
  car. Why?
• 5:55
  Well, because I earned more money this year and I deserve
• 5:58
  it, and I was really good with my physical health and so I
• 6:01
  deserve a reward. It's like, no, I just want a
• 6:03
  fast car. Stop making things up.
• 6:05
  Yeah, Yeah, absolutely. Yeah.
• 6:07
  It's an interesting and and definitely the world is changing
• 6:09
  in that, in that regard for sure.
• 6:11
  So a lot of people that will be listening to this, their
• 6:14
  organisations will be looking at, OK, how do we properly adopt
• 6:17
  this stuff and how do we start using it?
• 6:19
  So where do you see the biggest organizational barriers when it
• 6:23
  comes from embedding AI into existing workflows and the
• 6:27
  cultures as well? Yes, culture is the operative
• 6:30
  word here. It is difficulty at the top.
• 6:33
  The people who are using these tools are at the coalface.
• 6:37
  They're doing the work, and they're finding these tools to
• 6:40
  be very helpful doing the work. But they're extremely
• 6:44
  overwhelmed by mixed messages from upstairs where two years
• 6:48
  ago, Oh no, can't use it, it'll steal our data, which was true
• 6:52
  for about 6 months. But that's when everybody's
• 6:55
  policy got solidified. That's what everybody was told.
• 6:59
  No, no, you can't. We don't understand it well
• 7:02
  enough. Everybody used it anyway.
• 7:05
  The people using the tools understand the tools.
• 7:08
  The people at the very top, your average CEO, if you tell them
• 7:12
  absolutely this product AI is not going to do your job well,
• 7:16
  of course it can't do my job, we all say, but it can do tasks
• 7:21
  that can help you optimize your work.
• 7:24
  Well, ACEO doesn't have tasks. They have people who do tasks
• 7:28
  for them. I need a plane flight.
• 7:29
  I need to set a meeting and get set me up a phone call, get so
• 7:33
  and so on the phone. I don't I have people.
• 7:36
  So the upper echelon is not using the tool.
• 7:40
  So they don't understand the tool.
• 7:41
  Now let's go back to 1995 when I'm in a boardroom and they say
• 7:47
  we don't understand why you insist that we invest in this
• 7:51
  website thing. We don't understand what it's
• 7:54
  good for. And I turn to the chief
• 7:57
  financial officer and say, what do you like to do in your spare
• 8:00
  time? Just like what?
• 8:01
  What are you just humor me? Well, I, I like sailing.
• 8:06
  It's like, OK, look, sailing.com.
• 8:08
  Boom. And suddenly I've got his
• 8:10
  attention. Now I see how it relates to
• 8:12
  something I care about. The CEO.
• 8:15
  Their job is to absorb lots of information and make decisions
• 8:19
  they don't understand yet. How AI can be a thought partner,
• 8:24
  a brainstorming tool, a infinitely patient therapist,
• 8:30
  and that until they start doing and using AI, they're not able
• 8:35
  to lead by example. If they're not leading by
• 8:38
  example, the rest of the company says, are we supposed to use it?
• 8:41
  Will we get in trouble? Will I get fired?
• 8:43
  Will I train this thing to take my job?
• 8:45
  And there's just a complete disconnect between the levels in
• 8:49
  the organization. Yeah.
• 8:51
  Yeah. Some great.
• 8:51
  Points there and I think it's, it's an interesting scenario
• 8:54
  because you often hear the CEOCFO will say what's the ROI
• 8:59
  on that and it's not in many cases what we're delivering here
• 9:02
  isn't necessarily revenue generating, but the productivity
• 9:06
  savings are, are vast in this. And I think it's sometimes it's
• 9:10
  overlooked what's my financial return on this rather than what
• 9:14
  are the product is savings benefits and then obviously down
• 9:17
  the line that does have benefits to as well.
• 9:20
  And we have two big problems with the ROI.
• 9:23
  First of all, we're going through an enormous learning
• 9:25
  curve. Anybody who goes all in on AI is
• 9:29
  spending a lot more time working, trying to understand
• 9:33
  this tool. It's not like here, it's a
• 9:34
  spreadsheet. Let me give you the manual.
• 9:36
  This is not like that. We have to learn how to use it
• 9:39
  individually. It's a million YouTube videos,
• 9:42
  there's 1000 podcasts, there's thousands of blog posts, and I
• 9:47
  have spent an enormous amount of time on those in order to
• 9:51
  understand it well enough to teach others, but that's
• 9:55
  extraordinarily inefficient. The rest of the world is banging
• 9:58
  away. They ask a Google question, they
• 10:00
  get a bad answer, they go, it doesn't work.
• 10:02
  It's like, no, no, no, you're using it wrong.
• 10:04
  It works. So Part 1 is if you hand this
• 10:07
  tool to somebody, their productivity goes down because
• 10:10
  they have to learn to use the tool.
• 10:12
  Once they do and they start automating and optimizing and
• 10:17
  transforming their work, that's great, but it's going to take a
• 10:22
  year before we see the numbers show up in productivity, before
• 10:27
  we see the numbers show up in additional sales and lower
• 10:30
  costs. So we're asking the financial
• 10:34
  people, the ROI people, hey, invest a lot in this thing, but
• 10:37
  it won't pay off for 6 to 8 to 12 months.
• 10:41
  It's like, Oh no, I have to turn in a quarterly report to the
• 10:44
  board and to Wall Street. You can't ask me to to invest
• 10:47
  that much. It's like, yes, we can.
• 10:49
  Yeah, absolutely. And you touched on a really good
• 10:52
  point there. The, you know, adopting these
• 10:54
  things is is not easy. And in a in a transformation
• 10:58
  where the pace of technology seems to be outstripping human
• 11:02
  adaptation to it, or adaptation to it, should I say, what do you
• 11:07
  think the effective change management approach should be in
• 11:10
  this? Situation.
• 11:12
  It's several pieces. It starts with bringing together
• 11:15
  an AI council. So it's it is not a center of
• 11:19
  excellence. They're not going to be the
• 11:22
  rules makers. They're going to be the people
• 11:24
  who are the most interested, the most enthused, the best informed
• 11:29
  because they are listening to the podcasts every night.
• 11:32
  They do care. They are the nerds who want to
• 11:35
  know more. But we also need to bring IT
• 11:37
  because there's security issues and access issues, data issues.
• 11:42
  We need to bring in legal, obviously painfully need to
• 11:46
  bring them in. We also need to bring in the the
• 11:49
  folks who are responsible for getting the work done because
• 11:52
  they know what the jobs really are.
• 11:54
  They know what's actually happening on the factory floor,
• 11:56
  so to speak. You bring them together and
• 11:59
  their first responsibility is policy.
• 12:03
  Now they need to educate senior management so that they senior
• 12:07
  management will sign off on policy.
• 12:08
  But I have to have policy to clearly tell my people what is
• 12:13
  expected and what is allowed and what is not allowed.
• 12:16
  Then I can create rules thou shalt not.
• 12:20
  And then I can create guidelines.
• 12:22
  This works better than that. Today, check back tomorrow it'll
• 12:26
  change. So with with rules, policies and
• 12:30
  guidelines, my people now have authority to go do things on
• 12:34
  their own. And that means I have a rubric
• 12:38
  around which I can create education, getting as many
• 12:40
  people on board as possible as soon as possible.
• 12:43
  The common problem has been, oh, let's go buy, let's do a test.
• 12:47
  We'll buy 100 licenses and hand it out and see what happens.
• 12:51
  Productivity goes down. 75% of the people do a Google search
• 12:54
  and say, Nah, I don't care. Another 10% say I'm too busy to
• 12:59
  learn this thing. I'm so frantic running down the
• 13:02
  street I don't have time to jump in the car.
• 13:04
  And then you've got a handful, a small handful of people who are
• 13:07
  doing amazing work because of uneven education.
• 13:12
  So policy education. And then the celebration of
• 13:19
  quick wins and proof of concept gets culturally activated.
• 13:23
  All the while senior management has to be on board and talking
• 13:28
  it up and encouraging rather than demanding.
• 13:32
  The carrot works much better than the stick.
• 13:34
  I said yesterday or last week was 2 day long closed
• 13:40
  conversation and there was 1 Fortune 500 company that said
• 13:45
  first of all we require people to use it somehow.
• 13:48
  They have to tell their boss once a week some use case or it
• 13:53
  will. We will hold it against them.
• 13:55
  But on the carrot side, we award them points and the points are
• 13:59
  useful at the company store to buy tchotchkes and and swimwear
• 14:04
  and who knows? So we award them points for a
• 14:08
  unique use case, a shared story. And you get extra points if you
• 14:13
  describe a failure and what you learned because that you know
• 14:19
  the old saying of, of, of you learn this much in theory and
• 14:22
  this much in practice and this much by screwing up.
• 14:25
  And if you share that, you save everybody else the pain.
• 14:28
  Yeah, I like that carrot and stick approach.
• 14:30
  Just go back a little bit to what you, when you started this
• 14:33
  little segment, we talked around getting legal involved and
• 14:37
  things like that. We find that actually getting a
• 14:40
  yes from sales and marketing from what we're offering is
• 14:42
  normally pretty quick. The impact is obvious, the
• 14:45
  benefit is obvious, but the legal side of things can take
• 14:48
  weeks, months longer than it always used to.
• 14:51
  A lot of that seems to be around data privacy and data transfer
• 14:55
  impact assessments and all of these weird things.
• 14:58
  Do you see that more or less from an AI perspective?
• 15:02
  And you may or may not know this?
• 15:03
  I'm sure you will. Is there a difference like from
• 15:06
  the US as opposed to maybe in Europe where I'm based?
• 15:09
  Well, how much time have you got?
• 15:14
  So first of all, the legal team has to be involved because if
• 15:18
  you ignore them, you do so at your peril.
• 15:21
  The now we, we go immediate to corporate culture.
• 15:24
  The legal department is there to tell you no.
• 15:27
  That is their job. They are, are risk averse.
• 15:31
  That is their responsibility. And if they don't understand it,
• 15:35
  well then that's just a flat no. There's no, there's no
• 15:37
  recovering from that. We have them there to say what
• 15:41
  are the bright lines and then where are the fuzzy areas and
• 15:45
  the fuzzy areas? Well, that's up to management.
• 15:47
  Yes, we're willing to accept that risk.
• 15:49
  That's a corporate culture question, an industrial
• 15:52
  question, healthcare, finance, not so risk friendly research
• 15:56
  and development, sales and marketing.
• 15:58
  Yeah, we'll, we'll try stuff. So they need to be in the room,
• 16:02
  but they are not. They don't.
• 16:04
  Let's see. They're not going to lay down
• 16:06
  the law, so to speak, because there is no law yet.
• 16:10
  We're still inventing this stuff there.
• 16:12
  We have to. We're going to wait years for
• 16:14
  lawsuits to to reveal what precedents look like.
• 16:18
  In the meantime, everybody just says no.
• 16:19
  So that's Part 1. What's the difference between US
• 16:23
  and Europe? the United States has assumed
• 16:26
  the culture here assumes a lack of privacy.
• 16:29
  It's we've just been trained to click on.
• 16:33
  OK, OK, OK. No matter what, Europe has had a
• 16:37
  bad experience with a database getting into the wrong hands.
• 16:42
  Something that I fear Americans are going to discover shortly
• 16:47
  with major corporations having access to government data.
• 16:51
  With government buying private information, the US government
• 16:56
  is not allowed to spy on its own citizens.
• 16:58
  Doesn't need to, it just buys the data.
• 17:01
  So culturally, as a young country, we just gave up privacy
• 17:07
  for convenience. Sure.
• 17:09
  I, I want you to make recommendations.
• 17:10
  Yes, I want my $2 off at the till.
• 17:13
  And then there there is a fundamental legal difference
• 17:17
  going way back. The shortest way to describe it
• 17:20
  is the United States is a law by rule and Europe is law by
• 17:26
  principle. So in principle, do no harm.
• 17:30
  But the law didn't say I couldn't do that.
• 17:32
  Therefore I did not break the law.
• 17:35
  Yes, it was rude, it was crude, it was immoral, but I didn't
• 17:37
  break the law. Is the United States approach
• 17:39
  which is a moral failing. Yeah, OK.
• 17:43
  It sounds like your thinking is that US will end up probably
• 17:46
  more like Europe than the other way around.
• 17:48
  Well, that's the hope. In the meantime, we have a
• 17:52
  government administration that's going to make every use of all
• 17:57
  the data they can get to forward their agenda rather than serve
• 18:01
  the public. Personal opinion, yes.
• 18:04
  Yeah, absolutely. All right.
• 18:06
  That was really interesting. But getting back, I suppose, to
• 18:08
  the topic, if we look at businesses adopting the AI side
• 18:12
  of things, how can leaders ensure that when they adopt the
• 18:17
  AI approach, they're strengthening, not signed
• 18:20
  lining, sidelining the skills, the creativity, judgement of
• 18:24
  their existing teams? Judgement excellent word.
• 18:28
  So technically when auto a large language model invent something,
• 18:34
  the technical term is hallucination, what the human
• 18:37
  brings, The technical term that has been adopted by the industry
• 18:41
  is taste, it's judgement, it's discernment, it's gut feel, it's
• 18:45
  emotional. And that's that is what we need
• 18:48
  to hang on to desperately. So the proper approach is 3
• 18:54
  points, education, optimization, transformation.
• 18:59
  We have to teach everybody. This is new, it's different.
• 19:03
  Here's how it can help you. Here's how you can use it.
• 19:05
  Here are the fun things you can do.
• 19:07
  Here are the dangerous things. And then optimization.
• 19:10
  What tasks are on your desk that you don't really need to do?
• 19:14
  Do you really need to read this report?
• 19:17
  Well, if your job is to pull the data out of the report and put
• 19:22
  it in a spreadsheet, no you don't.
• 19:25
  That's just that's just mechanical.
• 19:27
  You can automate that. If your job is to understand the
• 19:30
  import of that report and apply it in a judgmental way to make
• 19:35
  decisions, then yeah, you better read the whole thing.
• 19:38
  That's that's the so optimization is workflow.
• 19:41
  Let's make things faster. Let's connect up all of our
• 19:45
  customer data from all of those silos in ways that we've had IT
• 19:49
  try to do forever with pipelines.
• 19:52
  Now we've got a tool that has a little bit of flexibility in it
• 19:57
  to monitor and go, oh, the data coming in from this data stream
• 20:01
  looks fine. This one had a drop out, so we
• 20:03
  have to accommodate that. And those are human activities
• 20:07
  that can be automated because they have an absolute outcome.
• 20:12
  Writing code is a great example. The code runs or it doesn't.
• 20:16
  That's great. Then finally we get to
• 20:18
  transformation and that is automation computers, the
• 20:22
  Internet, AI, they help us understand, automate, they help
• 20:28
  us speed up the things that we do, but generative AI helps us
• 20:34
  do different things. Do we really need that report?
• 20:37
  Do we need that standard operating procedure?
• 20:39
  That was a procedure that was put eight put together in 1872
• 20:44
  for this particular to solve this problem that doesn't exist
• 20:47
  anymore. That is a very large ask.
• 20:50
  It's a huge challenge, but the winners will run away with the
• 20:55
  prize. People who can, I mean, think of
• 20:59
  the huge transformations of the Internet, the Airbnb's and the
• 21:02
  Ubers and then the Netflix's. They all found a different way
• 21:08
  to do not just do things differently, but to do different
• 21:11
  things. And we're with with generative
• 21:14
  AI at this point. We have discovered the surface.
• 21:17
  We found the surface and just realized we have nails and we
• 21:20
  might be able to scratch. We haven't even started yet.
• 21:23
  Yeah, yeah. I think it's very exciting and
• 21:26
  it's funny. You see all these films from
• 21:27
  from years ago, the sci-fi films, and it's all starting to
• 21:30
  very slowly come together. The only thing I haven't seen is
• 21:33
  the flying skateboard yet. From the.
• 21:36
  Back to the Future, but the rest of it seems to here's.
• 21:37
  My flying car. Exactly, exactly.
• 21:41
  Although it did say there's, there's, I think it's Uber
• 21:44
  looking at flying taxis. So that's maybe not 1,000,000
• 21:47
  miles away. So what role do you think
• 21:50
  transparency and communication play in maintaining trust during
• 21:55
  AI driven change? I mean, I mentioned that AI
• 21:57
  without the Y and sending salespeople off on wild goose
• 22:00
  chases, which kills all the trust between sales and
• 22:03
  marketing and things like that. But what are your thoughts
• 22:05
  around that? Well, I'm in the the camp of
• 22:08
  transparency and all things without information.
• 22:12
  People make things up. I'm so so let me take that to
• 22:14
  the extreme. I believe that a company's
• 22:17
  finances should be transparent to employees because if I'm in
• 22:25
  customer service and somebody in sales comes in and says, we just
• 22:28
  sold this giant account and then, well, he's making this
• 22:31
  huge Commission. All the people upstairs are
• 22:34
  going to buy mansions and boats, and I can't even get a 2% raise.
• 22:38
  If you make it transparent that, yeah, we just sold this giant
• 22:43
  contract, but here are the costs, we're going to, we're
• 22:46
  actually only going to earn this much off the top.
• 22:48
  And are you familiar with how much it costs to keep the lights
• 22:51
  on? It's like, no, nobody is because
• 22:54
  woo woo, finances are what we can't tell anybody.
• 22:57
  I'm a believer in transparency and I think with AI, because we
• 23:01
  are all scared of it, it might do something we don't want it to
• 23:06
  do. It might take my job.
• 23:08
  It might make human work irrelevant.
• 23:11
  What is senior management thinking right now?
• 23:15
  Not just what are they telling the financial world?
• 23:18
  It's like, what is the plan? Are you going to, as we've seen,
• 23:22
  fire thousands of people because you think AI can do their job?
• 23:25
  Or are you going to upskill us and train us to do our jobs
• 23:30
  better? Have that conversation open.
• 23:33
  That's the management side. On the technology side, it's
• 23:37
  much more complex. The people making these tools
• 23:41
  don't know why they are exhibiting these new
• 23:44
  capabilities. They just know if they throw
• 23:46
  more compute, if they use more data, and if they use more
• 23:51
  creative, intelligent algorithms, these things get
• 23:55
  smarter, more capable. They exhibit traits that we
• 23:59
  didn't know were possible and suddenly really it does.
• 24:01
  That now is even happening in the laboratories.
• 24:05
  So that transparency of what is to come is really important and
• 24:11
  nobody knows. Yeah, it's exciting and scary
• 24:15
  and equal measure. And I think you hit on a really
• 24:16
  good point. Now we have lots of
• 24:18
  conversations where if this stuff, if we get it right, it
• 24:21
  actually frees up marketers to be marketers rather than
• 24:25
  database owners and things like that.
• 24:27
  So I think, yeah, you're definitely right in what you're
• 24:29
  saying there. So just trying to think of a
• 24:32
  practical example, as you look back over the last few years,
• 24:35
  have you got something you can share around where generative AI
• 24:40
  maybe created some measurable business value while still
• 24:44
  enhancing that employee engagement or capability or
• 24:47
  something like that? Let me share a story of a small
• 24:50
  company. There's 25 people in this
• 24:53
  company is construction restoration.
• 24:56
  So when there's a problem with the building, they go in and fix
• 24:59
  it. It's not renovation, adding a
• 25:01
  new room or anything. It is there was a fire, there
• 25:03
  was water damage. They go in and fix it.
• 25:05
  So a year and a half ago this woman bought this company and
• 25:10
  immediately downloaded her own large language model behind the
• 25:14
  firewall to create a Co CEO help me run this company and it
• 25:19
  worked feverishly to make it help so much.
• 25:24
  Prime exam use case. She goes into a building wearing
• 25:27
  meta glasses with video and says, I see there's some water
• 25:31
  damage up there. It looks like that electrical
• 25:34
  outlet had a problem and the floor is buckling over there.
• 25:37
  And by the time she gets back to her car, it has created a
• 25:41
  statement of work of what it will take to fix it with a bid.
• 25:44
  That's how many hours have been saved by how many humans that
• 25:48
  would have to do that work. And then she sits down at her
• 25:51
  desk and uses the meta glasses to watch her work on her
• 25:55
  computer and at the end of the day says how can I be more
• 25:58
  efficient? That creative use of computing
• 26:02
  is what excites me. And how do you measure ROI?
• 26:06
  She did the work of seven people.
• 26:08
  Does she fire 7 people? No, no, no.
• 26:10
  She upscales them. She makes them more.
• 26:13
  More. There is no limit to white
• 26:17
  collar work. There's a limited number of
• 26:19
  widgets you can create in a factory because of the factory.
• 26:23
  But working at your desk, working with ideas, working with
• 26:26
  people, there's an infinite amount of work.
• 26:28
  So if I can do twice the work in half the time, does that mean I
• 26:33
  take half a day off? No, I do the other stuff that I
• 26:36
  could never have gotten to. So how do you measure ROI out of
• 26:40
  that? It's kind of the wrong question.
• 26:43
  Is this a valuable tool? Yes.
• 26:45
  Yeah, absolutely. That's a great story.
• 26:47
  I really like that. So if we look now scaling some
• 26:50
  of these AI initiatives, are there particular leadership
• 26:54
  qualities you think will need to be there So it's able to drive
• 26:58
  both that innovation but also keep aligned with the human
• 27:01
  element? Well, the philosophy that came
• 27:04
  out of Silicon Valley that got out of hand is move fast and
• 27:08
  break things. I think that should be replaced
• 27:10
  with celebrate failure. And that's, that's not something
• 27:15
  that business is used to. You know, you missed your
• 27:17
  numbers 3/4 in a row, you're fired.
• 27:19
  What did you learn from that is the question.
• 27:23
  And if you didn't learn anything, then you're fired.
• 27:26
  You tried something and it didn't work.
• 27:28
  Tell us so that the rest of us can avoid that.
• 27:31
  That's a leadership quality that is cultural.
• 27:34
  Cultural corporate culture comes from personal psychology.
• 27:38
  So the people at the top who are fear driven, scarcity driven are
• 27:43
  going to have tough time with us.
• 27:45
  People who are all about abundance and growth, they're
• 27:49
  going to do well. Yeah, great point.
• 27:51
  And I'm a big believer in, you know, making mistakes is all
• 27:54
  part of learning. If you keep making the same
• 27:56
  mistakes over and over again, that's a different story.
• 27:58
  But yeah, definitely agree with that.
• 28:00
  Slightly more generic, but it's a trending topic.
• 28:03
  And so we're we're getting getting this a lot.
• 28:06
  So worth asking, what's the toughest feedback you've ever
• 28:09
  received and how did it shape you?
• 28:12
  Well, we're going back to the 1980s.
• 28:16
  I was selling software development tools.
• 28:18
  We had built tools for the deck vax and we just ported them over
• 28:24
  to IBM mainframe. Very exciting.
• 28:26
  And I flew from California to to Pennsylvania to give a
• 28:31
  presentation to a large railroad company.
• 28:34
  And I got about five slides and this is overhead foils into my
• 28:40
  presentation. And the guy at the head of the
• 28:43
  table said stop, just stop. You're speaking us, speaking to
• 28:48
  us in terms of mini computers where you're using files and
• 28:53
  folders. This is an IBM mainframe.
• 28:56
  We have directories and sub directories.
• 28:58
  You're not speaking our language.
• 28:59
  I want you to go back to the airport.
• 29:01
  I want you to fly home and I want you to tell your boss that
• 29:04
  I said that he did not train you well enough and it's his fault.
• 29:09
  I was as angry as I've ever been.
• 29:11
  I spent, I mean, that's a, it's a six hour flight to get home
• 29:15
  without television on the back of the screen in front of you,
• 29:18
  without a phone in your hand. I just sat there and stewed and
• 29:23
  determined that I would never give another presentation if I
• 29:25
  didn't know what I was talking about.
• 29:27
  Yeah, great learning and those things stick with you, right?
• 29:30
  It's the the making of you is when you get into a scenario
• 29:33
  like that and you actually learn from it.
• 29:34
  So I think that's that's a great one.
• 29:37
  You sound like someone, and I'm sure you are, who consumes lots
• 29:40
  of content. Is there a particular book, A
• 29:43
  blog, newsletter, podcast you'd recommend to our listeners?
• 29:48
  Let's start with LinkedIn. There is a professor from
• 29:51
  Wharton School named Ethan Malik who posts a couple of times a
• 29:56
  day, very short, very snackable, but he is very focused on what's
• 30:01
  right over the horizon, what's coming, how are these tools,
• 30:04
  what's new, what's different? How do you use them in higher
• 30:07
  education, how do you use them in business, and how to use them
• 30:09
  for some silly things. One of the things he likes to do
• 30:12
  is when it was text to image first happened, he said, OK,
• 30:18
  here's make me a picture of an Otter on an airplane using
• 30:22
  Wi-Fi. And it was a little cartoon that
• 30:24
  was humorous. Well, Fast forward to today and
• 30:28
  that command brings out a full video with audio and a script
• 30:35
  and characters interacting just Otter on an airplane using Wi-Fi
• 30:41
  and it generates it. So I follow him.
• 30:43
  I listen to the AI podcast. I wish it had a better name.
• 30:48
  It's from the smarter X dot AI is the company that does it.
• 30:53
  They're a consultancy. It is the kind of the news of
• 30:56
  the week. Also in the news of the week, I
• 30:58
  listen to Hard Fork, which is a gentleman from Platformer and a
• 31:04
  gentleman from the New York Times talking about the news,
• 31:07
  philosophy, psychology. They interview the leaders of
• 31:11
  the tech world. So there's a LinkedIn O for
• 31:15
  those in marketing. I'm going to strongly recommend
• 31:18
  LinkedIn and his blog, A Gentleman named Andy Crestadina
• 31:22
  from Orbit Studio who recommends how to use generative AI in
• 31:27
  incredible detail. Not from the technical
• 31:30
  perspective, but the how do I use it from marketing?
• 31:33
  Here is a 17 page long blog post on how to create a ideal
• 31:38
  customer profile persona. Here is how to measure the
• 31:42
  success of your blog post using AI to to be your thought leader.
• 31:47
  Really solid stuff and about 20 other people that I follow.
• 31:52
  Yeah, they're all, they're all names that are new to me, but I
• 31:56
  suppose that's my weakness Spirit and across the pond.
• 31:58
  But they're definitely ones I'll be looking into after this for
• 32:01
  sure. That's always a big thing for
• 32:03
  me. You've been an awesome guest,
• 32:04
  Jim. I have no doubt there's an
• 32:06
  argument for a follow up because we haven't even scratched the
• 32:09
  surface of the surface yet. And so I think there's an
• 32:11
  opportunity to do some more going forward.
• 32:13
  But right now, this is a cheeky ask, I know, but could you share
• 32:17
  the names of maybe two or three inspirational people in the B to
• 32:21
  B space that you recommend we should also bring on this show?
• 32:25
  Oh, well, I'm, you know, I don't know if you can get Seth Godin,
• 32:28
  but he's always inspirational. Over on the creative side,
• 32:35
  storytelling, which is becoming even more important now that we
• 32:39
  have AI, there's a gentleman named Mark Sylvester who has
• 32:44
  been running a local Santa Barbara Ted X events for 15
• 32:49
  years and has the last year and a half.
• 32:53
  He's a podcaster and he said, oh, I'm going to create a couple
• 32:57
  of agents to help me with my podcast and my writing.
• 33:01
  And he now has something like 43 individual assistants who are
• 33:06
  responsible for some portion to the point where he can be in his
• 33:10
  car and say, hey, Walter, I have an idea about blah, blah, blah.
• 33:15
  Interview me about that and he has 1/2 an hour conversation.
• 33:19
  Walter then hands it off to a whole series.
• 33:21
  It's a whole workflow and the result out the other end is
• 33:25
  here's a podcast script, here's a Facebook posters LinkedIn
• 33:29
  post, here is medium post and this is for your sub stack.
• 33:34
  It creates it. Now he's not a technologist, but
• 33:39
  he had he dove in and would be fascinating to talk to.
• 33:43
  I would also recommend well the guy who runs the podcast, the AI
• 33:49
  podcast and smarter X dot AI is Paul Reitzer ROETZER.
• 33:56
  He runs a marketing AI conference every year I get to
• 34:01
  be a speaker at he is right on top of things and paying
• 34:04
  attention. Who else would I recommend right
• 34:07
  off the top of my head. If your audience is more
• 34:09
  interested on the data side, the co-author of my last book is Tom
• 34:14
  Davenport, who got acclaim for writing the article that data
• 34:19
  scientists will be the sexiest job of the 21st century.
• 34:23
  So he's been in analytics as long as there's been analytics
• 34:27
  and is extraordinarily insightful.
• 34:29
  Awesome. Again, great names.
• 34:31
  I know I'm, I'm pretty confident we have hadn't had any on our
• 34:34
  podcast yet, so I'll be working hard to get them on.
• 34:37
  So really. One of them you tell them I
• 34:39
  recommended them. Absolutely.
• 34:41
  We always, we always would do that for sure.
• 34:43
  Now, I'm pretty confident you've opened up lots of potential
• 34:47
  conversations going forward. So how would someone listening
• 34:50
  to this show, if you're happy for them to get in touch with
• 34:53
  you after this podcast? I am on LinkedIn.
• 34:56
  I'm very happy to, to connect on LinkedIn.
• 34:58
  I have had the same e-mail address for 35 years.
• 35:01
  So it's extremely easy to find me that way.
• 35:04
  In between, you know, there's, there's Slack and Reddit and
• 35:07
  Twitter and, and there's just so many channels, e-mail, LinkedIn.
• 35:12
  That's me. Awesome, brilliant.
• 35:14
  Wow, what a great session. Jim.
• 35:16
  Thanks so much for your time today.
• 35:18
  I I hope you enjoyed it and I hope everyone listening enjoyed
• 35:21
  the episode. And don't forget there's nearly
• 35:24
  600 other episodes in the series, so make sure you check
• 35:27
  some of those out as well. In the meantime, thanks again,
• 35:30
  Jim. Thanks everyone for listening.
• 35:32
  Until next time, Cheerio. Thanks for joining us for On
• 35:37
  Base. If you enjoyed this episode,
• 35:38
  please be sure to share, rate, review and subscribe on Spotify,
• 35:42
  Apple, YouTube, or wherever you get your podcasts.
• 35:45
  This show is brought to you by Demand Base, the only account
• 35:48
  based experience platform built for sales and marketing
• 35:51
  alignment.