October 9, 2025 • 44 mins
Daniel sits down with Chelsea Linder, VP of Innovation and Entrepreneurship at TechPoint, to explore the what AI innovation and impact look like on the ground. They discuss Chelsea's journey from the VC world into economic development/ innovation, the growth of an AI innovation network in Indiana (funded by the SBA), lessons learned from fostering AI communities, and how businesses are actually adapting to AI. Chelsea also shares insights from Techpoints AI workforce impact study, which explored AI related job creation and levels of AI adoption among other things.
Featuring:
Links:
Sponsors:
- Shopify – The commerce platform trusted by millions. From idea to checkout, Shopify gives you everything you need to launch and scale your business—no matter your level of experience. Build beautiful storefronts, market with built-in AI tools, and tap into the platform powering 10% of all U.S. eCommerce.
Start your one-dollar trial at shopify.com/practicalai - Fabi.ai - The all-in-one data analysis platform for modern teams. From ad hoc queries to advanced analytics, Fabi lets you explore data wherever it lives—spreadsheets, Postgres, Snowflake, Airtable and more. Built-in Python and AI assistance help you move fast, then publish interactive dashboards or automate insights delivered straight to Slack, email, spreadsheets or wherever you need to share it.
Learn more and get started for free at fabi.ai
Upcoming Events:
- Join us at the Midwest AI Summit on November 13 in Indianapolis to hear world-class speakers share how they’ve scaled AI solutions. Don’t miss the AI Engineering Lounge, where you can sit down with experts for hands-on guidance. Reserve your spot today!
- Register for upcoming webinars here!
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
Jerod (00:04):
Welcome to the Practical AI podcast, where we break down
the real world applications ofartificial intelligence and how
it's shaping the way we live,work, and create. Our goal is to
help make AI technologypractical, productive, and
accessible to everyone. Whetheryou're a developer, business
leader, or just curious aboutthe tech behind the buzz, you're
(00:24):
in the right place. Be sure toconnect with us on LinkedIn, X,
or Blue Sky to stay up to datewith episode drops, behind the
scenes content, and AI insights.You can learn more at
practicalai.fm.
Now, onto the show.
Daniel (00:49):
Welcome to another episode of Practical AI. This is
Daniel Wightnack. I am CEO atPredictionGuard, and I'm really
excited today to follow-up on anumber of topics that that Chris
and I have discussed overprevious weeks related to wider
AI innovation, what's workingacross, various industry
(01:11):
sectors, how that's actuallyhappening, how that's impacting
workforce. Because we have withus today Chelsea Linder, who is
vice president of innovation andentrepreneurship at TechPoint.
How are doing, Chelsea?
Chelsea (01:23):
I'm doing great. Excited to be with you today.
Daniel (01:26):
Yeah. Yeah. I know you and I met a little while back
because you were kind of makingyour way into, I think, what is
your current role and focus,part of which is this AI
Innovation Network, which youset up in Indiana specifically,
and that's been very successful.But how does one get to get get
(01:49):
to that point? Describe a littlebit about your background and
kind of how you end up thinkingabout these kind of wider AI
innovation network things andworkforce things.
I know we'll talk a little bitabout that later.
Chelsea (02:02):
Yeah. For sure. So my original background is actually
in user experience research. SoI did that work for a while, at
a startup here in Indiana calledAngie's List. And then I took a
career detour to work in theventure capital space.
So I was a partner at Generator,which is a startup accelerator
(02:26):
and VC firm where we investedacross the country. I got
exposure to all the most cuttingedge technology and innovation
that was happening at theearliest stages through that
impacting investing activity. Sothen when I joined the TechPoint
team, TechPoint is a nonprofitand our mission at TechPoint is
(02:49):
to drive Indiana's digitaleconomy through talent,
innovation and community. Sowhen I joined about two years
ago, we were talking a lot aboutwhat does that mean and what do
we need to drive, especiallythrough that innovation lens to
ensure Indiana's economicsuccess at our core. We're
really an economic developmentfocused organization.
(03:12):
And of course, staying up todate on the most cutting edge
use cases for AI was definitelyone of those topics. And we had
the opportunity through afederal program called the
Growth Accelerator FundCompetition that's hosted by the
SBA to apply, to do assessmentof our community and identify
(03:33):
areas of needs specificallyaround AI education. So I think
that's when I met you, we weredoing a needs assessment,
mapping out the differentstakeholders and resources we
had across the state. And thething we kept hearing over and
over again was people across thestate of Indiana wanted access
to just know what other peoplein the state were doing to share
(03:55):
best practices with each other,really get in the weeds and
learn from each other's mistakesand each other's opportunities
so that we can all, you know,rising tide lifts all boats. We
can all support each other andmaking sure that Indiana can
stay on the cutting edge of thisnew technology advancement.
So we got the grant.
Daniel (04:14):
That's great. It was successful.
Chelsea (04:15):
AI network. Yep. Yeah. It was successful. And so we
launched the AI innovationnetwork to try to help, build
that peer group and and convenethose folks to share those best
practices and keep each other upto date.
Daniel (04:27):
Yeah. And how long has that been going on now?
Chelsea (04:30):
So it's been going on for almost a year. We launched
it in January.
Daniel (04:34):
Congratulations. Yeah.
Chelsea (04:35):
Thank you. Yeah. And we have, as of today, three ninety
nine members of the AIInnovation Network.
Daniel (04:42):
Surely, as this goes out, because it's prerecorded,
you will be at 400 already. Sowe'll just say we'll just say
400. I'll I'll I'll find someoneto to register. So that's That
Chelsea (04:55):
sounds great. Yeah. So, you know, it's it's really cool
to see that the market isinterested in what we're doing
and that we are providing thevalue to our community that we
set out to provide. That'sreally cool.
Daniel (05:08):
Yeah, have you, as you were looking at this kind of
need, of course, like in aspecific geography, I mean, you
think about where we're at,we're surrounded by the
industries that are sort ofbeing transformed by AI. So
whether that's like lifesciences manufacturing or
(05:28):
logistics or all of those thingsof course are kind of
surrounding us. As you did thoseneeds assessment, as you pulled
together, and we'll talk aboutspecific things that maybe
you're doing that have workedwell or learnings that you've
had that maybe you wanna pass onto others trying to do this type
(05:49):
of work. But as you evaluatedthat, were there examples out
there across the country ofother kinds of things like this
or what kind of went into theinspiration, I guess, for the
kind of mechanics of what thiscould or should be?
Chelsea (06:06):
Yes. So we mostly drew inspiration from our other
networks that we run at TechPoint, especially the Indiana
Founders Network, which is anetwork that I launched the year
prior to this one. And we havegotten a lot of great market
success from that as well anddesigned this network to be
similar to that one. But thereare a couple of key differences.
(06:29):
One of them being we really didintentionally wanna serve the
entire state.
And that can happen at an onlyin person convening. So we
wanted this to be hybrid andwe've been navigating what that
means and how to do it well overthe past year. But it has given
us the opportunity to servepeople in the middle of the
cornfields, in the manufacturingfloor, you know, all across our
(06:51):
state. And that has been areally powerful, impact from the
network.
Daniel (06:55):
Yeah, and for those out there, maybe kind of across the
country that are maybe in anenvironment where certain
people, maybe they are in SF orsomewhere and they have an AI
meetup every night, but there'scertainly large parts of the
country that are trying tofigure out kind of how to
navigate this. And of course,we'll talk about the workforce
(07:15):
things later, but just in termsof the community and what's
needed around this, what haveyou found to be some of those,
let's say the core components ofwhat people have responded to
well and gotten value out of? Somaybe that could inspire others
in different areas that aretrying to do some of this
(07:39):
networking or maybe even spin uptheir own sorts of communities
related to this technology.
Chelsea (07:45):
Absolutely. I think the number one benefit that we have
here in Indiana is Hoosierhospitality. The way that that
shows up for the AI network isthat our members are extremely
willing to share and betransparent about their work.
And we very rarely come acrossinstances where somebody says
(08:06):
like, no, I need to protect myIP. I'm not willing to share.
Or we just don't come acrossthat kind of ego driven
unwillingness to be open. Andthat is only to the benefit of
the other members. Especially atthe beginning, I think people
were so excited to have anoutlet to talk about their
projects that we just had somany people raise their hands
(08:26):
and say, I want to share whatI'm working on. I would love to
get the feedback from otherexperts in my field. I would
love to help a new person learnhow to do this or replicate this
type of work within theirorganization.
I think it's the power of havinga strong community of helpers.
And I don't know the secret tobuilding a strong community of
(08:47):
helpers, but I do think thatthat's really the foundation
that we were able to build offof.
Daniel (08:52):
What do you think that eagerness I mean, I've even seen
it on the podcast here withpeople that we've had on that
are really wanting to, I guessthey've felt pain maybe through
some of their AI journey andmaybe that's part of why they
want to share things. What isyour sense of, yeah, why is part
(09:14):
of that eagerness there and whatare they hoping to provide like
in the AI innovation networkthat you run when people are
sharing their stories aboutwhat's working, what's not
working? Yeah, what are theirmotivations? What are they
hoping to see?
Chelsea (09:27):
I think that, like I said, a big part of it is being
able to learn, or identifyopportunities to get better at
what you're doingprofessionally. I think that's a
big piece of it. I also thinkthat we, as a community are
we're focused on education.Indiana has like the most
universities per capita, I thinkexcept for Massachusetts. And so
(09:51):
for us, like being a lifelonglearner, being in an environment
of people who just wanna learn alot is a big piece of it.
And then thirdly, I would sayone of the key topics that our
advisory council who we havehelping us with the content for
the network, What we talkedabout at the beginning was a
feeling of imposter syndrome of,I think I'm doing a good job,
(10:14):
but am I actually? I don't havea lot of peers in this work that
I can verify with. So a littlebit of willingness to be an open
book or show your work just tocombat those feelings of
imposter syndrome. We've startedto try to kind of democratize
the opportunities within thenetwork, doing more live
(10:36):
prompting or live codingactivities, again, just to help
with that feeling of impostersyndrome and make everyone see,
you know, they're not the worstperson in the room at whatever
they're doing.
Daniel (10:47):
Yeah. It helps to be in a room with what you might
perceive as very smart people,but in reality are other
practitioners. And certainlyI've noticed that over time
where I have a very small sphereof knowledge, but others have
different sort of spheres ofknowledge. And when you, I think
(11:08):
start to realize that, that wehave a lot to learn from one
another and I may be able totalk about whatever security
related concerns with AI agents.I certainly can't talk about
other things at the core ofpharma manufacturing and how AI
(11:28):
fits into that always.
So like the, yeah, that sort ofcross pollination is really
exciting. Does that happen? Likewhat, describe for us kind of an
AI innovation network event?Like, what would what would that
typically look like? What couldpeople expect?
Yeah. How do people show upthere and what what happens?
Chelsea (11:50):
Yeah. So we have kinda three different buckets, I would
say, of events. So the first oneis just purely networking. Just
make friends, get to know peoplein your field. And so we
facilitate that throughdifferent activities or
different types of icebreakersto just help folks feel more
comfortable in the room.
Secondly, we do on a quarterlybasis, a case study where we'll
(12:13):
have a partner come in andpresent about a project that
they've done under within theirown organization or for a
client. Then we kind of take itas similarly to how you would
structure a PhD course, wherewe'll have the beginning half be
about describing the case,describing the problem, break
(12:33):
into small groups and have adiscussion about how would you
approach that problem and whattactics would you use and then
wrap it up with what theyactually did on the outcome
side. And then we follow-up witha written documentation of that
case study. We're trying againto just give that more in the
weeds knowledge to thepractitioners that are in the
(12:53):
group. And then the third one Iwould say is more of that kind
of open book sharing aboutprojects or processes that folks
are working on to be able to getmore real time feedback or just
show how an application may workinside of a different
organization.
For example, recently we had afantastic presenter from Baker
(13:13):
Hill come in to talk about howthey encouraged adoption across
their team, which is, of course,a really major issue that we're
all trying to talk about rightnow, not only adoption,
upskilling and rescaling up theworkforce, and being able to
share a shining star example ofhow it went well and what they
learned. It that's justincredibly applicable to other
(13:34):
organizations that are in theroom.
Sponsors (13:49):
Well, friends, it is time to let go of the old way of
exploring your data. It'sholding you back. Well, what
exactly is the old way? Well,I'm here with Marc Dupuy, co
founder and CEO of Fabi, acollaborative analytics platform
designed to help big explorerslike yourself. So, Mark, tell me
about this old way.
So the old way, Adam, if you're a a product manager or
(14:11):
a founder and you're trying toget insights from your data,
you're you're wrestling withyour Postgres instance or
Snowflake or your spreadsheets,or if you are you don't maybe
even have the support of a dataanalyst or data scientist to
help you with that work. Or ifyou are, for example, a data
scientist or engineer oranalyst, you're wrestling with a
bunch of different tools, localJupyter Notebooks, Google Colab,
(14:33):
or even your legacy BI to try tobuild these dashboards that
someone may or may not go andlook at. And in this new way
that we're building at ABBYY, weare creating this all in one
environment where productmanagers and founders can very
quickly go and explore dataregardless of where it is. It
can be in a spreadsheet, it canbe in Airtable, it can be in
Postgres, Snowflake. Really easyto do everything from an ad hoc
(14:56):
analysis to much more advancedanalysis if you're more
experienced.
So with Python built in rightthere, in our AI assistant, you
can move very quickly throughadvanced data analysis. And the
really cool part is that you cango from ad hoc analysis and data
science to publishing these asinteractive data apps and
dashboards, or better yet,delivering insights as automated
(15:20):
workflows to meet yourstakeholders where they are in,
say, Slack or email orspreadsheet. So if this is
something that you'reexperiencing, if you're a
founder or product managertrying to get more from your
data or for your data team todayand you're just underwater and
feel like you're wrestling withyour legacy BI tools and and
notebooks, come check out thenew way and come try out Fabi.
There you go. Well, friends, if you're trying to get
(15:41):
more insights from your data,stop resting with it. Start
exploring it the new way withFabi. Learn more and get started
for free at fabi.ai. That'sfabi.ai.
Again, fabbi.ai.
Daniel (16:00):
Well, Chelsea, it's really great to see, this
happening in my backyard and seesome of this innovation network
being formed. What lessons doyou feel for maybe, again,
others out there that arethinking about forming
communities around this topic,around AI, whether that be on
(16:24):
Discord or hybrid or via inperson events. What have been
some of the lessons you'velearned that are maybe like
minds to not step on or justlike challenges that are part of
it that you haven't totallyfigured out yet?
Chelsea (16:41):
I think the number one would be not assuming that what
works for our community willwork for your community. And
that's why we have the advisorycouncil. I am not an in the
weeds AI practitioner. That'snot my day job. And so I never
would want to pretend that I wasor lead the the content of the
(17:01):
network in that way.
So we built an advisory counciland they've been instrumental in
helping us identify where theacknowledged gaps are, how we
can help build those andchampioning the network in
general as well. So I would saystep one, start by talking to
your community and figuring outwhat they actually need. And
then I think you have to kind ofcapture and hold onto the
(17:23):
momentum as tightly as you can.So certainly at the beginning,
we had a lot of excitement,enthusiasm, momentum for the
network. With some things we dida great job capturing that and
pulling it forward and withother things we didn't.
So some of those initiativeshave kind of slowed down or
aren't happening anymore. So youkind of have to pick and choose,
(17:43):
what are the things that arereally providing the value you
wanna provide that you candouble down on? And then are
there things that maybe aren'tworking out that you don't need
to continue doing movingforward?
Daniel (17:53):
From your perspective, because you have been working on
this over this past year, haveyou seen just kind of general
trends or shifts in people'sview towards AI within companies
or people's education around thetopic or the types of things
(18:15):
that they want to talk about.Because that's something maybe
for some of us that are in theindustry or like building AI
things, we have our own view ofmaybe what's on people's mind
and that sort of thing, wherethey're at in terms of what
they're thinking about. And thatdoesn't always map to actual
(18:36):
discussions in companies, wherepeople are at on their kind of
educational journey, what theywant talk about. Like, are they
just trying to get their firstchat interface up and like have
that be a win? Are they thinkingabout, tying in AI functionality
(18:57):
to existing products or buildingcustom tools?
Are they thinking mostly about,like you said, the adoption side
of things? Have you seen thatshift, over this past year and
any anything to kind ofhighlight there?
Chelsea (19:14):
Yes, absolutely. I think, we're really uniquely
positioned as TechPoint becausewe have working relationships
with everyone from Eli Lilly,which was recently named the
number one most AI ready companyin the world.
Daniel (19:29):
Oh, I didn't know that. That's so that's awesome.
Congrats to that.
Chelsea (19:32):
Really cool. Right? It's great. All the way down to,
like, little baby startups ormore main street lifestyle type
of businesses that have perhapsbeen operating for many, many
years and have basically zerotechnology adoption. Right?
So we have the viewpoint of thatentire spectrum. And what I've
witnessed over the past year, ormaybe even more than that, is a
(19:56):
democratization of access, whichhas been really cool to see
where, you know, Lily has allthe resources and all the
capabilities and incrediblytalented and smart people to
help get them to that place.And, you know, automotive shop
down the road doesn't, butthanks to a lot of the
(20:18):
initiatives that have happenedboth within Indiana and just in
general with more education andawareness of AI, that small
business has the same access andopportunity. And that's really
cool to see because now we cansee how a small business may be
able to grow exponentially morequickly, increase their
(20:41):
productivity, etcetera, in a waythat they really never would
have been able to before. Withthat in mind, I think what's
happening is these businessesare realizing that this is a
must do.
And so now I'm getting more andmore inbound questioning from
businesses that probably don'teven have a tech team at all,
have never really thought aboutthis before saying, I know I
(21:03):
need to do this, How do I getstarted? And that's a really
great position to be in to say,okay, well come to the AI
Innovation Network. And thenwe'll introduce you to some of
the people who are alreadyleading the way here and they
can share with you what worksand what doesn't work and we can
go from there. So that's been afantastic thing to see. On the
(21:25):
flip side of that, I think a bigchallenge that every business is
facing right now is with theworkforce and upskilling,
reskilling, job securityquestions, all of that.
So Tuckpoint just recentlypublished a report about AI
driven skills and the changes inworkforce demands due to AI. And
it's it's just some reallyfascinating data. Right? Like,
(21:47):
job postings for generative AIengineers are up by seven times
year over year. Fascinating datato think about.
But at the same time, we'reseeing higher levels of
unemployment, higher levels ofreduction in forces, all kind of
also attributed to AI. There's alot of contrarian data, but at
(22:10):
the end of the day, I think weall can agree that we need to
focus on upskilling andreskilling and that we need to
do it in a way that can preservethe workforce as much as
possible. And we're so lucky tohave companies like Lilly in our
backyard that are leading theway on that.
Daniel (22:24):
Yeah, this report is really fascinating. I encourage
people and we'll link this inour show notes, but this is a
report AI driven skills forIndiana's economy insights from
employers and industry trends.Certainly something like at the
heart of our country that is thereality kind of on the ground of
what's happening, which is why Ithink this is a really great way
(22:49):
to present this. So give us alittle bit of a background on
this particular report, kind ofhow it came about and the data
behind it.
Chelsea (23:01):
Yeah. So if you recall, I mentioned that TechPoint
achieves our mission throughthree pillars. So innovation,
community, and talent. And thetalent pillar has historically
been about solving the braindrain problem we had in Indiana,
helping retain and recruitworkforce, especially tech
workforce to come to our stateor stay in our state once they
graduate from one of our amazinghigher ed institutions. And over
(23:25):
the past couple of years, we'vebeen seeing that problem not be
the biggest problem foremployers anymore.
So it drove TechPoint to go backand talk to our stakeholders and
employers across the state andsay, okay, we sense that
retention and recruiting of techtalent is not your biggest
problem anymore. What is? And wekind of already knew what they
(23:47):
were gonna say, but theyabsolutely validated our
assumption that aligning theworkforce with these new in
demand skills is their numberone talent problem right now.
And so with that in mind, westarted doing a lot more of this
research and ultimately came upwith the contents of this
report. And what we found out isthat right now we needed to
(24:11):
focus on three things.
We need to focus on integratingAI into the workforce training
program. So including higher ed,as well as within an
organization upskilling, We needto focus on doing that in an
industry specific way becausethe AI skills in every industry,
as you mentioned, likemanufacturing, AI in a
manufacturing floor looks verydifferent than AI in
(24:34):
pharmaceuticals. So we need toalign that in industry specific
ways. And then we need to expandadoption past our early adopter
sectors. So we really need topush into some of those more
lagging sectors likeagriculture, etcetera.
And then thirdly, we need tocreate cross sector AI knowledge
as well. So we need to sharebest practices across all of our
(24:55):
strongest industries. And the AInetwork is one of the ways that
we can do that. But if we can dothese three things, then we
think we will be able tomaintain a competitive workforce
here in Indiana.
Daniel (25:06):
Yeah, some of these things, I encourage people to
look at the report. I mean,there's some things just
generally that are interestingkind of nationwide and some
things that are kind of a casestudy of where we live
specifically. But yeah, jobpostings for generative AI
engineers up 7X is reallyinteresting. Another thing
(25:29):
though that you mentioned is,job postings requiring
generative AI skills in other ITroles are up 35%. And I think
you have some graphs aroundthis.
So there is this kind ofgenerative AI engineers or AI
engineers specifically that is ain demand thing. It kind of
(25:51):
reminds me back when everybodywas looking for data scientists
and we were trying to figure outhow to get data science into
workforce education all of thosethings, but that is up, but even
more so or at a higher rate arethese other technology positions
that are requiring skills in AI.From your sense, is that due to
(26:17):
just the, I know you also talkabout or show kind of a graph in
the report of companies adoptionof AI, of official adoption of
AI, maybe not shadow usage ofAI, but official adoption of AI
kind of rising up to 10% orbeyond. Do you think that's just
(26:40):
a matter of them figuring outthat, yes, this is here to stay
and we need to not just have awe need to have a SIS admin that
also knows how to use these AItools. Like what do you think is
behind that?
Chelsea (27:00):
Yeah, I think it's partially because it's becoming
clear that it's a competitiveadvantage. And so companies are
doubling down on requiring thoseskills. You know? It's hard to
ignore when you hear somebodylike the CEO of Atlassian being
like, everyone has to do this.Right?
So I think that's that's a bigpart of it. And then I also
(27:22):
think that, with Gen AI, as youknow, I don't need to tell you
this, this is really just thetip of the iceberg when it comes
to truly implementing AI, notjust Gen AI within your business
and that what businesses arereally gonna see success from is
not using a chatbot, right? It'sa many layers deeper
(27:43):
application. And so I thinkhaving these roles is kind of
the preemptive steps towardsmoving into those deeper
applications where we're trulytalking about process
automation, etcetera. So I thinkit's the tip of the iceberg and
I feel like those roles aregoing to continue to require
some level of skills.
(28:04):
Similarly to now how a productowner would normally be required
to have some level of codingskills. Right? So I think we'll
continue to see that proliferateacross organizations. I recently
was, able to present to a groupof CFOs of companies that had
50,000,000 annual revenue ormore. And it was really
interesting to see the spectrumof I absolutely will never trust
(28:29):
AI to do my finance work to,like, I love it.
It does so much of the thingsthat I used to hate doing. And I
think we're just again, we'regonna see that proliferate
across every business unit orevery department within an
organization.
Sponsors (28:54):
Well, friends, when you're building and shipping AI
products at scale, there's oneconstant, complexity. Yes.
You're bringing the models, datapipelines, deployment
infrastructure, and then someonesays, let's turn this into a
business. Cue the chaos. That'swhere Shopify steps in whether
you're spinning up a storefrontfor your AI powered app or
(29:15):
launching a brand around thetools you built.
Shopify is the commerce platformtrusted by millions of
businesses and 10% of all USecommerce from names like
Mattel, Gymshark to foundersjust like you. With literally
hundreds of ready to usetemplates, powerful built in
marketing tools, and AI thatwrites product descriptions for
(29:36):
you, headlines, even polishesyour product photography.
Shopify doesn't just get youselling, it makes you look good
doing it. And we love it. We useit here at Changelog.
Check us outmerch.changelog.com. That's our
store front and it handles theheavy lifting too. Payments,
inventory, returns, shipping,even global logistics. It's like
(29:57):
having an ops team built intoyour stack to help you sell. So
if you're ready to sell, you areready for Shopify.
Sign up now for your $1 permonth trial and start selling
today atshopify.com/practicalai. Again,
that is shopify dot com slashpracticalai.
Daniel (30:22):
Chelsea, just another interesting thing that I see in
this report, which kind of stoodout to me is you talk about kind
of the skills gap, which some ofthose like highest skills gap
around AI and machine learning,data analytics, big data,
cybersecurity, but there's kindof chart of most in demand roles
(30:45):
now. And it kind of stood out tome that software developer or
engineer is still up there asthe highest in demand skill,
even higher than the AI machinelearning engineer. And so I
think there is this narrativethat we hear, right? Everything
is a kind of mixed bag. It's amessy situation, but narrative
(31:10):
on the one side of the spectrumis that, well, AI is kind of
doing all of these softwareengineering things now and we
don't need as many softwareengineers to do the same amount
of work, which means like maybeI need to be an AI engineer.
It's not enough to be a softwareengineer and maybe this is a
(31:33):
timing thing, but I don't knowif you have any thoughts on
that. That one stood out to meas kind of even in light of all
the vibe coding things that aregoing on, software development
is still and and being asoftware engineer is still a
good gig to have at least atleast here.
Chelsea (31:51):
Yeah. So this data is trailing data. So I do expect
that that number will change inthe future as we you know, we're
scraping job descriptions andthings that have already
happened. Right? So I doanticipate that those AI focused
engineering roles will take overas the number one most in
(32:13):
demand.
But I also have been doing a lotmore research on this recently.
And as an example, I spoke withan organization who had laid off
all of their junior devs. Andthey had their mid levels
working with AI to do the workthat the juniors were doing. And
at first, the mid levels lovedit. They were like, this is
(32:35):
great.
I'm getting basically the samequality work as a junior dev
would give me. And I don't haveto manage a human being. I don't
have to deal with the emotionalside of it. I can be as blunt as
I want to. Like, they're nevergonna get upset with me.
Right? And they loved it. Well,six months down the road, those
same mid levels are saying, canwe please hire some junior devs?
Because Oh, wow. I am so annoyedat my AI, and I don't wanna have
(32:59):
to be the one managing that.
Right? I don't wanna have to bethe one doing the prompting,
getting it to fix all of itsdumb mistakes, hallucinations,
whatever. So I think that justspeaks to this is a transition
in what those jobs look like,not a complete removal of those
jobs from our workforce.
Daniel (33:20):
Yeah, that's really interesting. It reminds me of a
discussion, I forget whatepisode it was on, but we were
talking about this kind of vibecoding work and the actual
mental burden of that work foran engineer is very different
from the kind of normal, oh, I'mtyping things in my editor, even
(33:43):
with auto complete or somethinglike that, where you have like
multiple things going on atonce. You're constantly context
switching between differentagents doing different things.
And to your point, you kind ofget into this like master
orchestrator mode. And maybe toyour point, like there are
engineers who like doingengineering and they don't want
(34:05):
to be just those orchestratortype people.
So it's almost like a different,it is a different cognitive
load. It's a different day today workflow and we haven't,
certainly haven't fullynavigated that yet and
understand, you know, what willhappen, right? Yes. When that
(34:25):
comes about. But I think atleast Chris and I's view on the
podcast here, I think we've saida few times, I don't think
software engineers are goinganywhere.
The dynamics and kind of theroles may shift, like you said,
but there's still going to be aneed for those resources. There
may additionally be a need forthese, you know, increasing need
(34:47):
for these resources around AIengineers and that sort of
thing.
Chelsea (34:51):
Yeah. I will hold fast forever that the number one
skill you need to vibe code wellis to already know how to code.
Daniel (34:59):
How to how to code. Yeah. Hey. Good good point.
That's a that's a great quote.
Yeah. I I love that. Yeah. And Ithink, I mean, even back when
people were first starting, likein the seventies and eighties
when people were programming, Ithink they were already talking
about, well, computer programsare eventually gonna be able to
(35:19):
program and we're not gonna havejobs. And that was quite a while
ago.
Now things are moving a littlebit quicker now, I think you
could argue, and maybe there'ssome differences, but I think
we're at least safe for themoment. Absolutely. Yeah. I am
interested in this graph youshow about the share of
(35:39):
businesses that are using AI.Could you talk about the kind of
range of industries that thatrepresents?
Just to give people, becauseyou're listening on audio, a
visual of this, it's essentiallyan increasing trend to more and
more, right? But this goes fromSeptember '23 up to July '25.
(36:02):
And September '23 was less than4% in US, slightly less in
Indiana, and then kind of bothconverge up to around 10% in
July 2025. So around 10% ofbusinesses using kind of using
AI. So could you help usunderstand like, what does that
(36:23):
mean using AI, what industries,etcetera?
Chelsea (36:26):
Yes. Yeah. So that data is from the US Census Bureau and
it's a survey, it's the BTOFsurvey. And forgive me for not
remembering what that acronymstands for, but it is a survey
they do every two weeks. Sothat's why we see this data.
You look at the charts, it'spretty lumpy, because they're
getting different numbers ofresponses, etcetera, from
different industries, differentstratification every two weeks.
(36:49):
But what the questions are,there's two different questions
that they ask to pull that data.They're very specific questions.
So one of them is, are you usingAI to build your product at your
organization? And then thesecond one is, are you using AI
to do business within yourorganization?
(37:09):
So it kind of touches on yourquestion, which is what exactly
are they using it for? Right?And so I didn't have access to
the data for those two separatequestions. I think that would be
really interesting to see.Assumption is that very few are
using it within their productsand very many more are using it
within their businessoperations.
(37:29):
That's where the data comesfrom. And it definitely makes me
feel positive about Indiana'strajectory to see us kind of
catching up with the rest of thenation. But then we could
compare that to the datarecently yesterday or a couple
of days ago coming out of MITtalking about how 99% of AI
projects fail or pilots fail. SoI think it's one thing to say,
(37:52):
yes, I'm doing it. And it's acompletely different thing to
say we're doing it successfully.
And I think that's the data thatwe're gonna have to dig in more
to find out more about and howwe can leverage programs like
the AI network or otherwiseacross the country to ensure the
success of our adoptionprograms.
Daniel (38:10):
Yeah, yeah. And it is interesting, like the perception
around these sorts of usenumbers. My friend Scott at Chip
dot ai, I think he sayssomething like 90% of people are
using AI or some high netfigure, 80% or whatever it is,
80% of people are using AI, but10% of businesses are using AI,
(38:33):
which if the 80% of the peoplein the businesses are using AI,
it probably just means a lot ofcompanies don't know that their
employees are using AI. Do youfind that to be a common theme
amongst these discussions ofeven just figuring out what
(38:54):
people are doing because people,it seems like are just getting
access to whatever tools theythink will make them productive
because they actually probablydo wanna be productive in their
jobs or make their jobs easieror whatever that is.
Chelsea (39:06):
Yes, absolutely. And I think that's true about
individuals as well. I would sayprobably almost 100%, at least
in The United States, are usingAI. They probably just don't
know. Right?
That tends a percent
Daniel (39:16):
of people. Fair enough.
Chelsea (39:17):
Yeah. Probably don't know. And I think that's the
same with businesses. As anexample, I my my daycare, one of
my daycare teachers startedusing Gen AI to write her daily
status updates about my daughterat daycare. Nice.
I was not cool at that for
Daniel (39:36):
probably For various reasons. Yes.
Chelsea (39:40):
And when I went to go and talk to that administrator
of that daycare, she had no cluethat that her staff was doing
that. And certainly, a daycaredoesn't have an acceptable use
policy of AI. Right? So I thinkthere's a lot of those types of
circumstances, and the bestthing that a business owner can
do is regardless of whether youthink people are using it or
(40:01):
not, you need to have a policy.You need to set up the
appropriate guide rails sopeople can do it safely.
And I think that starts withgiving approved access to
everyone.
Daniel (40:11):
Yeah, yeah. I think certainly there's companies that
try to lock down from thebeginning, but it doesn't really
stop it. It just sort of stopsthe maybe egregious use. I don't
know. Yeah.
Chelsea (40:28):
It stops them from doing it on their work computer.
Daniel (40:31):
Yeah, maybe so. That's way a to put it. But yeah, I
very much encourage people,please go and look at this
report. It's very nice. We'lllink it in the show notes and
you can find it at the TechPointwebsite.
There's a lot of cool stuffcoming up that if you are in the
(40:52):
Midwest or in Indiana,TechPoint's doing, they have
this great Muir Awards, whichI've been to, which celebrates a
lot of innovation that'shappening and people can find
out about that and nominatepeople. There's events coming
up. What are you kind of, as youlook at all the ecosystem,
Chelsea, of things that ofcourse have spun up this past
(41:16):
year, but maybe kind of wider,just what's happening and the
shifts that you're seeing interms of adoption of AI and case
studies around AI. What's mostexciting for you as you're
looking forward to the next yearof the AI Innovation Network,
maybe related to the network,but also just related to your
own view of like how thetechnology is developing?
Chelsea (41:38):
I'm really excited about, how AI is pushing people
to be even more innovative andthink about things in even more
uncomfortable ways than they didbefore. And especially right now
in Indiana, the conversationaround data centers, I think is
driving a lot of forcedinnovation where we have some of
(42:00):
the best scientists in the worldhere in Indiana, and they're
trying to figure out how we canmake data centers more efficient
and more ethical and sustainableand all of those things that
there wasn't a pressure cookerto figure out until relatively
recently. So I'm really excitedto see that because, you know,
if we can find a copperalternative that can stay cooled
(42:20):
more efficiently, that's gonnahave impacts across so many
industries. AI is just you know,data centers are just one use
case for that. Right?
So that gives me, I always,challenges drive innovation. And
so the more of these kind ofhard conversations that we have
to have, the more innovationwill stem out of it. So I'm
looking forward to seeing whatthe next crop of amazing
(42:42):
innovations are that come out ofthis really pressure full time
that we're in.
Daniel (42:47):
Yeah, that's awesome. Well, I'm sure that TechPoint
and the AI Innovation Networkwill continue to produce great
reports around this, but alsoengage the community as all of
that's going on. So really,really appreciate you joining
us, Chelsea. It's been a it'sbeen an amazing, conversation
and, of course, look lookforward to interacting with you
(43:09):
in my own community around,things that that are going on.
And, yeah, thank you so much forjoining us.
Look forward to having you backagain sometime.
Chelsea (43:17):
Thank you. It was great catching up.
Jerod (43:26):
Alright. That's our show for this week. If you haven't
checked out our website, head topracticalai.fm and be sure to
connect with us on LinkedIn, X,or Blue Sky. You'll see us
posting insights related to thelatest AI developments, and we
would love for you to join theconversation. Thanks to our
partner Prediction Guard forproviding operational support
for the show.
Check them out atpredictionguard.com. Also,
(43:49):
thanks to Breakmaster Cylinderfor the beats and to you for
listening. That's all for now,but you'll hear from us again
next week.
Welcome to the Practical AI podcast, where we break down
the real world applications ofartificial intelligence and how
it's shaping the way we live,work, and create. Our goal is to
help make AI technologypractical, productive, and
accessible to everyone. Whetheryou're a developer, business
leader, or just curious aboutthe tech behind the buzz, you're
(00:24):
in the right place. Be sure toconnect with us on LinkedIn, X,
or Blue Sky to stay up to datewith episode drops, behind the
scenes content, and AI insights.You can learn more at
practicalai.fm.
Now, onto the show.
Daniel (00:49):
Welcome to another episode of Practical AI. This is
Daniel Wightnack. I am CEO atPredictionGuard, and I'm really
excited today to follow-up on anumber of topics that that Chris
and I have discussed overprevious weeks related to wider
AI innovation, what's workingacross, various industry
(01:11):
sectors, how that's actuallyhappening, how that's impacting
workforce. Because we have withus today Chelsea Linder, who is
vice president of innovation andentrepreneurship at TechPoint.
How are doing, Chelsea?
Chelsea (01:23):
I'm doing great. Excited to be with you today.
Daniel (01:26):
Yeah. Yeah. I know you and I met a little while back
because you were kind of makingyour way into, I think, what is
your current role and focus,part of which is this AI
Innovation Network, which youset up in Indiana specifically,
and that's been very successful.But how does one get to get get
(01:49):
to that point? Describe a littlebit about your background and
kind of how you end up thinkingabout these kind of wider AI
innovation network things andworkforce things.
I know we'll talk a little bitabout that later.
Chelsea (02:02):
Yeah. For sure. So my original background is actually
in user experience research. SoI did that work for a while, at
a startup here in Indiana calledAngie's List. And then I took a
career detour to work in theventure capital space.
So I was a partner at Generator,which is a startup accelerator
(02:26):
and VC firm where we investedacross the country. I got
exposure to all the most cuttingedge technology and innovation
that was happening at theearliest stages through that
impacting investing activity. Sothen when I joined the TechPoint
team, TechPoint is a nonprofitand our mission at TechPoint is
(02:49):
to drive Indiana's digitaleconomy through talent,
innovation and community. Sowhen I joined about two years
ago, we were talking a lot aboutwhat does that mean and what do
we need to drive, especiallythrough that innovation lens to
ensure Indiana's economicsuccess at our core. We're
really an economic developmentfocused organization.
(03:12):
And of course, staying up todate on the most cutting edge
use cases for AI was definitelyone of those topics. And we had
the opportunity through afederal program called the
Growth Accelerator FundCompetition that's hosted by the
SBA to apply, to do assessmentof our community and identify
(03:33):
areas of needs specificallyaround AI education. So I think
that's when I met you, we weredoing a needs assessment,
mapping out the differentstakeholders and resources we
had across the state. And thething we kept hearing over and
over again was people across thestate of Indiana wanted access
to just know what other peoplein the state were doing to share
(03:55):
best practices with each other,really get in the weeds and
learn from each other's mistakesand each other's opportunities
so that we can all, you know,rising tide lifts all boats. We
can all support each other andmaking sure that Indiana can
stay on the cutting edge of thisnew technology advancement.
So we got the grant.
Daniel (04:14):
That's great. It was successful.
Chelsea (04:15):
AI network. Yep. Yeah. It was successful. And so we
launched the AI innovationnetwork to try to help, build
that peer group and and convenethose folks to share those best
practices and keep each other upto date.
Daniel (04:27):
Yeah. And how long has that been going on now?
Chelsea (04:30):
So it's been going on for almost a year. We launched
it in January.
Daniel (04:34):
Congratulations. Yeah.
Chelsea (04:35):
Thank you. Yeah. And we have, as of today, three ninety
nine members of the AIInnovation Network.
Daniel (04:42):
Surely, as this goes out, because it's prerecorded,
you will be at 400 already. Sowe'll just say we'll just say
400. I'll I'll I'll find someoneto to register. So that's That
Chelsea (04:55):
sounds great. Yeah. So, you know, it's it's really cool
to see that the market isinterested in what we're doing
and that we are providing thevalue to our community that we
set out to provide. That'sreally cool.
Daniel (05:08):
Yeah, have you, as you were looking at this kind of
need, of course, like in aspecific geography, I mean, you
think about where we're at,we're surrounded by the
industries that are sort ofbeing transformed by AI. So
whether that's like lifesciences manufacturing or
(05:28):
logistics or all of those thingsof course are kind of
surrounding us. As you did thoseneeds assessment, as you pulled
together, and we'll talk aboutspecific things that maybe
you're doing that have workedwell or learnings that you've
had that maybe you wanna pass onto others trying to do this type
(05:49):
of work. But as you evaluatedthat, were there examples out
there across the country ofother kinds of things like this
or what kind of went into theinspiration, I guess, for the
kind of mechanics of what thiscould or should be?
Chelsea (06:06):
Yes. So we mostly drew inspiration from our other
networks that we run at TechPoint, especially the Indiana
Founders Network, which is anetwork that I launched the year
prior to this one. And we havegotten a lot of great market
success from that as well anddesigned this network to be
similar to that one. But thereare a couple of key differences.
(06:29):
One of them being we really didintentionally wanna serve the
entire state.
And that can happen at an onlyin person convening. So we
wanted this to be hybrid andwe've been navigating what that
means and how to do it well overthe past year. But it has given
us the opportunity to servepeople in the middle of the
cornfields, in the manufacturingfloor, you know, all across our
(06:51):
state. And that has been areally powerful, impact from the
network.
Daniel (06:55):
Yeah, and for those out there, maybe kind of across the
country that are maybe in anenvironment where certain
people, maybe they are in SF orsomewhere and they have an AI
meetup every night, but there'scertainly large parts of the
country that are trying tofigure out kind of how to
navigate this. And of course,we'll talk about the workforce
(07:15):
things later, but just in termsof the community and what's
needed around this, what haveyou found to be some of those,
let's say the core components ofwhat people have responded to
well and gotten value out of? Somaybe that could inspire others
in different areas that aretrying to do some of this
(07:39):
networking or maybe even spin uptheir own sorts of communities
related to this technology.
Chelsea (07:45):
Absolutely. I think the number one benefit that we have
here in Indiana is Hoosierhospitality. The way that that
shows up for the AI network isthat our members are extremely
willing to share and betransparent about their work.
And we very rarely come acrossinstances where somebody says
(08:06):
like, no, I need to protect myIP. I'm not willing to share.
Or we just don't come acrossthat kind of ego driven
unwillingness to be open. Andthat is only to the benefit of
the other members. Especially atthe beginning, I think people
were so excited to have anoutlet to talk about their
projects that we just had somany people raise their hands
(08:26):
and say, I want to share whatI'm working on. I would love to
get the feedback from otherexperts in my field. I would
love to help a new person learnhow to do this or replicate this
type of work within theirorganization.
I think it's the power of havinga strong community of helpers.
And I don't know the secret tobuilding a strong community of
(08:47):
helpers, but I do think thatthat's really the foundation
that we were able to build offof.
Daniel (08:52):
What do you think that eagerness I mean, I've even seen
it on the podcast here withpeople that we've had on that
are really wanting to, I guessthey've felt pain maybe through
some of their AI journey andmaybe that's part of why they
want to share things. What isyour sense of, yeah, why is part
(09:14):
of that eagerness there and whatare they hoping to provide like
in the AI innovation networkthat you run when people are
sharing their stories aboutwhat's working, what's not
working? Yeah, what are theirmotivations? What are they
hoping to see?
Chelsea (09:27):
I think that, like I said, a big part of it is being
able to learn, or identifyopportunities to get better at
what you're doingprofessionally. I think that's a
big piece of it. I also thinkthat we, as a community are
we're focused on education.Indiana has like the most
universities per capita, I thinkexcept for Massachusetts. And so
(09:51):
for us, like being a lifelonglearner, being in an environment
of people who just wanna learn alot is a big piece of it.
And then thirdly, I would sayone of the key topics that our
advisory council who we havehelping us with the content for
the network, What we talkedabout at the beginning was a
feeling of imposter syndrome of,I think I'm doing a good job,
(10:14):
but am I actually? I don't havea lot of peers in this work that
I can verify with. So a littlebit of willingness to be an open
book or show your work just tocombat those feelings of
imposter syndrome. We've startedto try to kind of democratize
the opportunities within thenetwork, doing more live
(10:36):
prompting or live codingactivities, again, just to help
with that feeling of impostersyndrome and make everyone see,
you know, they're not the worstperson in the room at whatever
they're doing.
Daniel (10:47):
Yeah. It helps to be in a room with what you might
perceive as very smart people,but in reality are other
practitioners. And certainlyI've noticed that over time
where I have a very small sphereof knowledge, but others have
different sort of spheres ofknowledge. And when you, I think
(11:08):
start to realize that, that wehave a lot to learn from one
another and I may be able totalk about whatever security
related concerns with AI agents.I certainly can't talk about
other things at the core ofpharma manufacturing and how AI
(11:28):
fits into that always.
So like the, yeah, that sort ofcross pollination is really
exciting. Does that happen? Likewhat, describe for us kind of an
AI innovation network event?Like, what would what would that
typically look like? What couldpeople expect?
Yeah. How do people show upthere and what what happens?
Chelsea (11:50):
Yeah. So we have kinda three different buckets, I would
say, of events. So the first oneis just purely networking. Just
make friends, get to know peoplein your field. And so we
facilitate that throughdifferent activities or
different types of icebreakersto just help folks feel more
comfortable in the room.
Secondly, we do on a quarterlybasis, a case study where we'll
(12:13):
have a partner come in andpresent about a project that
they've done under within theirown organization or for a
client. Then we kind of take itas similarly to how you would
structure a PhD course, wherewe'll have the beginning half be
about describing the case,describing the problem, break
(12:33):
into small groups and have adiscussion about how would you
approach that problem and whattactics would you use and then
wrap it up with what theyactually did on the outcome
side. And then we follow-up witha written documentation of that
case study. We're trying againto just give that more in the
weeds knowledge to thepractitioners that are in the
(12:53):
group. And then the third one Iwould say is more of that kind
of open book sharing aboutprojects or processes that folks
are working on to be able to getmore real time feedback or just
show how an application may workinside of a different
organization.
For example, recently we had afantastic presenter from Baker
(13:13):
Hill come in to talk about howthey encouraged adoption across
their team, which is, of course,a really major issue that we're
all trying to talk about rightnow, not only adoption,
upskilling and rescaling up theworkforce, and being able to
share a shining star example ofhow it went well and what they
learned. It that's justincredibly applicable to other
(13:34):
organizations that are in theroom.
Sponsors (13:49):
Well, friends, it is time to let go of the old way of
exploring your data. It'sholding you back. Well, what
exactly is the old way? Well,I'm here with Marc Dupuy, co
founder and CEO of Fabi, acollaborative analytics platform
designed to help big explorerslike yourself. So, Mark, tell me
about this old way.
So the old way, Adam, if you're a a product manager or
(14:11):
a founder and you're trying toget insights from your data,
you're you're wrestling withyour Postgres instance or
Snowflake or your spreadsheets,or if you are you don't maybe
even have the support of a dataanalyst or data scientist to
help you with that work. Or ifyou are, for example, a data
scientist or engineer oranalyst, you're wrestling with a
bunch of different tools, localJupyter Notebooks, Google Colab,
(14:33):
or even your legacy BI to try tobuild these dashboards that
someone may or may not go andlook at. And in this new way
that we're building at ABBYY, weare creating this all in one
environment where productmanagers and founders can very
quickly go and explore dataregardless of where it is. It
can be in a spreadsheet, it canbe in Airtable, it can be in
Postgres, Snowflake. Really easyto do everything from an ad hoc
(14:56):
analysis to much more advancedanalysis if you're more
experienced.
So with Python built in rightthere, in our AI assistant, you
can move very quickly throughadvanced data analysis. And the
really cool part is that you cango from ad hoc analysis and data
science to publishing these asinteractive data apps and
dashboards, or better yet,delivering insights as automated
(15:20):
workflows to meet yourstakeholders where they are in,
say, Slack or email orspreadsheet. So if this is
something that you'reexperiencing, if you're a
founder or product managertrying to get more from your
data or for your data team todayand you're just underwater and
feel like you're wrestling withyour legacy BI tools and and
notebooks, come check out thenew way and come try out Fabi.
There you go. Well, friends, if you're trying to get
(15:41):
more insights from your data,stop resting with it. Start
exploring it the new way withFabi. Learn more and get started
for free at fabi.ai. That'sfabi.ai.
Again, fabbi.ai.
Daniel (16:00):
Well, Chelsea, it's really great to see, this
happening in my backyard and seesome of this innovation network
being formed. What lessons doyou feel for maybe, again,
others out there that arethinking about forming
communities around this topic,around AI, whether that be on
(16:24):
Discord or hybrid or via inperson events. What have been
some of the lessons you'velearned that are maybe like
minds to not step on or justlike challenges that are part of
it that you haven't totallyfigured out yet?
Chelsea (16:41):
I think the number one would be not assuming that what
works for our community willwork for your community. And
that's why we have the advisorycouncil. I am not an in the
weeds AI practitioner. That'snot my day job. And so I never
would want to pretend that I wasor lead the the content of the
(17:01):
network in that way.
So we built an advisory counciland they've been instrumental in
helping us identify where theacknowledged gaps are, how we
can help build those andchampioning the network in
general as well. So I would saystep one, start by talking to
your community and figuring outwhat they actually need. And
then I think you have to kind ofcapture and hold onto the
(17:23):
momentum as tightly as you can.So certainly at the beginning,
we had a lot of excitement,enthusiasm, momentum for the
network. With some things we dida great job capturing that and
pulling it forward and withother things we didn't.
So some of those initiativeshave kind of slowed down or
aren't happening anymore. So youkind of have to pick and choose,
(17:43):
what are the things that arereally providing the value you
wanna provide that you candouble down on? And then are
there things that maybe aren'tworking out that you don't need
to continue doing movingforward?
Daniel (17:53):
From your perspective, because you have been working on
this over this past year, haveyou seen just kind of general
trends or shifts in people'sview towards AI within companies
or people's education around thetopic or the types of things
(18:15):
that they want to talk about.Because that's something maybe
for some of us that are in theindustry or like building AI
things, we have our own view ofmaybe what's on people's mind
and that sort of thing, wherethey're at in terms of what
they're thinking about. And thatdoesn't always map to actual
(18:36):
discussions in companies, wherepeople are at on their kind of
educational journey, what theywant talk about. Like, are they
just trying to get their firstchat interface up and like have
that be a win? Are they thinkingabout, tying in AI functionality
(18:57):
to existing products or buildingcustom tools?
Are they thinking mostly about,like you said, the adoption side
of things? Have you seen thatshift, over this past year and
any anything to kind ofhighlight there?
Chelsea (19:14):
Yes, absolutely. I think, we're really uniquely
positioned as TechPoint becausewe have working relationships
with everyone from Eli Lilly,which was recently named the
number one most AI ready companyin the world.
Daniel (19:29):
Oh, I didn't know that. That's so that's awesome.
Congrats to that.
Chelsea (19:32):
Really cool. Right? It's great. All the way down to,
like, little baby startups ormore main street lifestyle type
of businesses that have perhapsbeen operating for many, many
years and have basically zerotechnology adoption. Right?
So we have the viewpoint of thatentire spectrum. And what I've
witnessed over the past year, ormaybe even more than that, is a
(19:56):
democratization of access, whichhas been really cool to see
where, you know, Lily has allthe resources and all the
capabilities and incrediblytalented and smart people to
help get them to that place.And, you know, automotive shop
down the road doesn't, butthanks to a lot of the
(20:18):
initiatives that have happenedboth within Indiana and just in
general with more education andawareness of AI, that small
business has the same access andopportunity. And that's really
cool to see because now we cansee how a small business may be
able to grow exponentially morequickly, increase their
(20:41):
productivity, etcetera, in a waythat they really never would
have been able to before. Withthat in mind, I think what's
happening is these businessesare realizing that this is a
must do.
And so now I'm getting more andmore inbound questioning from
businesses that probably don'teven have a tech team at all,
have never really thought aboutthis before saying, I know I
(21:03):
need to do this, How do I getstarted? And that's a really
great position to be in to say,okay, well come to the AI
Innovation Network. And thenwe'll introduce you to some of
the people who are alreadyleading the way here and they
can share with you what worksand what doesn't work and we can
go from there. So that's been afantastic thing to see. On the
(21:25):
flip side of that, I think a bigchallenge that every business is
facing right now is with theworkforce and upskilling,
reskilling, job securityquestions, all of that.
So Tuckpoint just recentlypublished a report about AI
driven skills and the changes inworkforce demands due to AI. And
it's it's just some reallyfascinating data. Right? Like,
(21:47):
job postings for generative AIengineers are up by seven times
year over year. Fascinating datato think about.
But at the same time, we'reseeing higher levels of
unemployment, higher levels ofreduction in forces, all kind of
also attributed to AI. There's alot of contrarian data, but at
(22:10):
the end of the day, I think weall can agree that we need to
focus on upskilling andreskilling and that we need to
do it in a way that can preservethe workforce as much as
possible. And we're so lucky tohave companies like Lilly in our
backyard that are leading theway on that.
Daniel (22:24):
Yeah, this report is really fascinating. I encourage
people and we'll link this inour show notes, but this is a
report AI driven skills forIndiana's economy insights from
employers and industry trends.Certainly something like at the
heart of our country that is thereality kind of on the ground of
what's happening, which is why Ithink this is a really great way
(22:49):
to present this. So give us alittle bit of a background on
this particular report, kind ofhow it came about and the data
behind it.
Chelsea (23:01):
Yeah. So if you recall, I mentioned that TechPoint
achieves our mission throughthree pillars. So innovation,
community, and talent. And thetalent pillar has historically
been about solving the braindrain problem we had in Indiana,
helping retain and recruitworkforce, especially tech
workforce to come to our stateor stay in our state once they
graduate from one of our amazinghigher ed institutions. And over
(23:25):
the past couple of years, we'vebeen seeing that problem not be
the biggest problem foremployers anymore.
So it drove TechPoint to go backand talk to our stakeholders and
employers across the state andsay, okay, we sense that
retention and recruiting of techtalent is not your biggest
problem anymore. What is? And wekind of already knew what they
(23:47):
were gonna say, but theyabsolutely validated our
assumption that aligning theworkforce with these new in
demand skills is their numberone talent problem right now.
And so with that in mind, westarted doing a lot more of this
research and ultimately came upwith the contents of this
report. And what we found out isthat right now we needed to
(24:11):
focus on three things.
We need to focus on integratingAI into the workforce training
program. So including higher ed,as well as within an
organization upskilling, We needto focus on doing that in an
industry specific way becausethe AI skills in every industry,
as you mentioned, likemanufacturing, AI in a
manufacturing floor looks verydifferent than AI in
(24:34):
pharmaceuticals. So we need toalign that in industry specific
ways. And then we need to expandadoption past our early adopter
sectors. So we really need topush into some of those more
lagging sectors likeagriculture, etcetera.
And then thirdly, we need tocreate cross sector AI knowledge
as well. So we need to sharebest practices across all of our
(24:55):
strongest industries. And the AInetwork is one of the ways that
we can do that. But if we can dothese three things, then we
think we will be able tomaintain a competitive workforce
here in Indiana.
Daniel (25:06):
Yeah, some of these things, I encourage people to
look at the report. I mean,there's some things just
generally that are interestingkind of nationwide and some
things that are kind of a casestudy of where we live
specifically. But yeah, jobpostings for generative AI
engineers up 7X is reallyinteresting. Another thing
(25:29):
though that you mentioned is,job postings requiring
generative AI skills in other ITroles are up 35%. And I think
you have some graphs aroundthis.
So there is this kind ofgenerative AI engineers or AI
engineers specifically that is ain demand thing. It kind of
(25:51):
reminds me back when everybodywas looking for data scientists
and we were trying to figure outhow to get data science into
workforce education all of thosethings, but that is up, but even
more so or at a higher rate arethese other technology positions
that are requiring skills in AI.From your sense, is that due to
(26:17):
just the, I know you also talkabout or show kind of a graph in
the report of companies adoptionof AI, of official adoption of
AI, maybe not shadow usage ofAI, but official adoption of AI
kind of rising up to 10% orbeyond. Do you think that's just
(26:40):
a matter of them figuring outthat, yes, this is here to stay
and we need to not just have awe need to have a SIS admin that
also knows how to use these AItools. Like what do you think is
behind that?
Chelsea (27:00):
Yeah, I think it's partially because it's becoming
clear that it's a competitiveadvantage. And so companies are
doubling down on requiring thoseskills. You know? It's hard to
ignore when you hear somebodylike the CEO of Atlassian being
like, everyone has to do this.Right?
So I think that's that's a bigpart of it. And then I also
(27:22):
think that, with Gen AI, as youknow, I don't need to tell you
this, this is really just thetip of the iceberg when it comes
to truly implementing AI, notjust Gen AI within your business
and that what businesses arereally gonna see success from is
not using a chatbot, right? It'sa many layers deeper
(27:43):
application. And so I thinkhaving these roles is kind of
the preemptive steps towardsmoving into those deeper
applications where we're trulytalking about process
automation, etcetera. So I thinkit's the tip of the iceberg and
I feel like those roles aregoing to continue to require
some level of skills.
(28:04):
Similarly to now how a productowner would normally be required
to have some level of codingskills. Right? So I think we'll
continue to see that proliferateacross organizations. I recently
was, able to present to a groupof CFOs of companies that had
50,000,000 annual revenue ormore. And it was really
interesting to see the spectrumof I absolutely will never trust
(28:29):
AI to do my finance work to,like, I love it.
It does so much of the thingsthat I used to hate doing. And I
think we're just again, we'regonna see that proliferate
across every business unit orevery department within an
organization.
Sponsors (28:54):
Well, friends, when you're building and shipping AI
products at scale, there's oneconstant, complexity. Yes.
You're bringing the models, datapipelines, deployment
infrastructure, and then someonesays, let's turn this into a
business. Cue the chaos. That'swhere Shopify steps in whether
you're spinning up a storefrontfor your AI powered app or
(29:15):
launching a brand around thetools you built.
Shopify is the commerce platformtrusted by millions of
businesses and 10% of all USecommerce from names like
Mattel, Gymshark to foundersjust like you. With literally
hundreds of ready to usetemplates, powerful built in
marketing tools, and AI thatwrites product descriptions for
(29:36):
you, headlines, even polishesyour product photography.
Shopify doesn't just get youselling, it makes you look good
doing it. And we love it. We useit here at Changelog.
Check us outmerch.changelog.com. That's our
store front and it handles theheavy lifting too. Payments,
inventory, returns, shipping,even global logistics. It's like
(29:57):
having an ops team built intoyour stack to help you sell. So
if you're ready to sell, you areready for Shopify.
Sign up now for your $1 permonth trial and start selling
today atshopify.com/practicalai. Again,
that is shopify dot com slashpracticalai.
Daniel (30:22):
Chelsea, just another interesting thing that I see in
this report, which kind of stoodout to me is you talk about kind
of the skills gap, which some ofthose like highest skills gap
around AI and machine learning,data analytics, big data,
cybersecurity, but there's kindof chart of most in demand roles
(30:45):
now. And it kind of stood out tome that software developer or
engineer is still up there asthe highest in demand skill,
even higher than the AI machinelearning engineer. And so I
think there is this narrativethat we hear, right? Everything
is a kind of mixed bag. It's amessy situation, but narrative
(31:10):
on the one side of the spectrumis that, well, AI is kind of
doing all of these softwareengineering things now and we
don't need as many softwareengineers to do the same amount
of work, which means like maybeI need to be an AI engineer.
It's not enough to be a softwareengineer and maybe this is a
(31:33):
timing thing, but I don't knowif you have any thoughts on
that. That one stood out to meas kind of even in light of all
the vibe coding things that aregoing on, software development
is still and and being asoftware engineer is still a
good gig to have at least atleast here.
Chelsea (31:51):
Yeah. So this data is trailing data. So I do expect
that that number will change inthe future as we you know, we're
scraping job descriptions andthings that have already
happened. Right? So I doanticipate that those AI focused
engineering roles will take overas the number one most in
(32:13):
demand.
But I also have been doing a lotmore research on this recently.
And as an example, I spoke withan organization who had laid off
all of their junior devs. Andthey had their mid levels
working with AI to do the workthat the juniors were doing. And
at first, the mid levels lovedit. They were like, this is
(32:35):
great.
I'm getting basically the samequality work as a junior dev
would give me. And I don't haveto manage a human being. I don't
have to deal with the emotionalside of it. I can be as blunt as
I want to. Like, they're nevergonna get upset with me.
Right? And they loved it. Well,six months down the road, those
same mid levels are saying, canwe please hire some junior devs?
Because Oh, wow. I am so annoyedat my AI, and I don't wanna have
(32:59):
to be the one managing that.
Right? I don't wanna have to bethe one doing the prompting,
getting it to fix all of itsdumb mistakes, hallucinations,
whatever. So I think that justspeaks to this is a transition
in what those jobs look like,not a complete removal of those
jobs from our workforce.
Daniel (33:20):
Yeah, that's really interesting. It reminds me of a
discussion, I forget whatepisode it was on, but we were
talking about this kind of vibecoding work and the actual
mental burden of that work foran engineer is very different
from the kind of normal, oh, I'mtyping things in my editor, even
(33:43):
with auto complete or somethinglike that, where you have like
multiple things going on atonce. You're constantly context
switching between differentagents doing different things.
And to your point, you kind ofget into this like master
orchestrator mode. And maybe toyour point, like there are
engineers who like doingengineering and they don't want
(34:05):
to be just those orchestratortype people.
So it's almost like a different,it is a different cognitive
load. It's a different day today workflow and we haven't,
certainly haven't fullynavigated that yet and
understand, you know, what willhappen, right? Yes. When that
(34:25):
comes about. But I think atleast Chris and I's view on the
podcast here, I think we've saida few times, I don't think
software engineers are goinganywhere.
The dynamics and kind of theroles may shift, like you said,
but there's still going to be aneed for those resources. There
may additionally be a need forthese, you know, increasing need
(34:47):
for these resources around AIengineers and that sort of
thing.
Chelsea (34:51):
Yeah. I will hold fast forever that the number one
skill you need to vibe code wellis to already know how to code.
Daniel (34:59):
How to how to code. Yeah. Hey. Good good point.
That's a that's a great quote.
Yeah. I I love that. Yeah. And Ithink, I mean, even back when
people were first starting, likein the seventies and eighties
when people were programming, Ithink they were already talking
about, well, computer programsare eventually gonna be able to
(35:19):
program and we're not gonna havejobs. And that was quite a while
ago.
Now things are moving a littlebit quicker now, I think you
could argue, and maybe there'ssome differences, but I think
we're at least safe for themoment. Absolutely. Yeah. I am
interested in this graph youshow about the share of
(35:39):
businesses that are using AI.Could you talk about the kind of
range of industries that thatrepresents?
Just to give people, becauseyou're listening on audio, a
visual of this, it's essentiallyan increasing trend to more and
more, right? But this goes fromSeptember '23 up to July '25.
(36:02):
And September '23 was less than4% in US, slightly less in
Indiana, and then kind of bothconverge up to around 10% in
July 2025. So around 10% ofbusinesses using kind of using
AI. So could you help usunderstand like, what does that
(36:23):
mean using AI, what industries,etcetera?
Chelsea (36:26):
Yes. Yeah. So that data is from the US Census Bureau and
it's a survey, it's the BTOFsurvey. And forgive me for not
remembering what that acronymstands for, but it is a survey
they do every two weeks. Sothat's why we see this data.
You look at the charts, it'spretty lumpy, because they're
getting different numbers ofresponses, etcetera, from
different industries, differentstratification every two weeks.
(36:49):
But what the questions are,there's two different questions
that they ask to pull that data.They're very specific questions.
So one of them is, are you usingAI to build your product at your
organization? And then thesecond one is, are you using AI
to do business within yourorganization?
(37:09):
So it kind of touches on yourquestion, which is what exactly
are they using it for? Right?And so I didn't have access to
the data for those two separatequestions. I think that would be
really interesting to see.Assumption is that very few are
using it within their productsand very many more are using it
within their businessoperations.
(37:29):
That's where the data comesfrom. And it definitely makes me
feel positive about Indiana'strajectory to see us kind of
catching up with the rest of thenation. But then we could
compare that to the datarecently yesterday or a couple
of days ago coming out of MITtalking about how 99% of AI
projects fail or pilots fail. SoI think it's one thing to say,
(37:52):
yes, I'm doing it. And it's acompletely different thing to
say we're doing it successfully.
And I think that's the data thatwe're gonna have to dig in more
to find out more about and howwe can leverage programs like
the AI network or otherwiseacross the country to ensure the
success of our adoptionprograms.
Daniel (38:10):
Yeah, yeah. And it is interesting, like the perception
around these sorts of usenumbers. My friend Scott at Chip
dot ai, I think he sayssomething like 90% of people are
using AI or some high netfigure, 80% or whatever it is,
80% of people are using AI, but10% of businesses are using AI,
(38:33):
which if the 80% of the peoplein the businesses are using AI,
it probably just means a lot ofcompanies don't know that their
employees are using AI. Do youfind that to be a common theme
amongst these discussions ofeven just figuring out what
(38:54):
people are doing because people,it seems like are just getting
access to whatever tools theythink will make them productive
because they actually probablydo wanna be productive in their
jobs or make their jobs easieror whatever that is.
Chelsea (39:06):
Yes, absolutely. And I think that's true about
individuals as well. I would sayprobably almost 100%, at least
in The United States, are usingAI. They probably just don't
know. Right?
That tends a percent
Daniel (39:16):
of people. Fair enough.
Chelsea (39:17):
Yeah. Probably don't know. And I think that's the
same with businesses. As anexample, I my my daycare, one of
my daycare teachers startedusing Gen AI to write her daily
status updates about my daughterat daycare. Nice.
I was not cool at that for
Daniel (39:36):
probably For various reasons. Yes.
Chelsea (39:40):
And when I went to go and talk to that administrator
of that daycare, she had no cluethat that her staff was doing
that. And certainly, a daycaredoesn't have an acceptable use
policy of AI. Right? So I thinkthere's a lot of those types of
circumstances, and the bestthing that a business owner can
do is regardless of whether youthink people are using it or
(40:01):
not, you need to have a policy.You need to set up the
appropriate guide rails sopeople can do it safely.
And I think that starts withgiving approved access to
everyone.
Daniel (40:11):
Yeah, yeah. I think certainly there's companies that
try to lock down from thebeginning, but it doesn't really
stop it. It just sort of stopsthe maybe egregious use. I don't
know. Yeah.
Chelsea (40:28):
It stops them from doing it on their work computer.
Daniel (40:31):
Yeah, maybe so. That's way a to put it. But yeah, I
very much encourage people,please go and look at this
report. It's very nice. We'lllink it in the show notes and
you can find it at the TechPointwebsite.
There's a lot of cool stuffcoming up that if you are in the
(40:52):
Midwest or in Indiana,TechPoint's doing, they have
this great Muir Awards, whichI've been to, which celebrates a
lot of innovation that'shappening and people can find
out about that and nominatepeople. There's events coming
up. What are you kind of, as youlook at all the ecosystem,
Chelsea, of things that ofcourse have spun up this past
(41:16):
year, but maybe kind of wider,just what's happening and the
shifts that you're seeing interms of adoption of AI and case
studies around AI. What's mostexciting for you as you're
looking forward to the next yearof the AI Innovation Network,
maybe related to the network,but also just related to your
own view of like how thetechnology is developing?
Chelsea (41:38):
I'm really excited about, how AI is pushing people
to be even more innovative andthink about things in even more
uncomfortable ways than they didbefore. And especially right now
in Indiana, the conversationaround data centers, I think is
driving a lot of forcedinnovation where we have some of
(42:00):
the best scientists in the worldhere in Indiana, and they're
trying to figure out how we canmake data centers more efficient
and more ethical and sustainableand all of those things that
there wasn't a pressure cookerto figure out until relatively
recently. So I'm really excitedto see that because, you know,
if we can find a copperalternative that can stay cooled
(42:20):
more efficiently, that's gonnahave impacts across so many
industries. AI is just you know,data centers are just one use
case for that. Right?
So that gives me, I always,challenges drive innovation. And
so the more of these kind ofhard conversations that we have
to have, the more innovationwill stem out of it. So I'm
looking forward to seeing whatthe next crop of amazing
(42:42):
innovations are that come out ofthis really pressure full time
that we're in.
Daniel (42:47):
Yeah, that's awesome. Well, I'm sure that TechPoint
and the AI Innovation Networkwill continue to produce great
reports around this, but alsoengage the community as all of
that's going on. So really,really appreciate you joining
us, Chelsea. It's been a it'sbeen an amazing, conversation
and, of course, look lookforward to interacting with you
(43:09):
in my own community around,things that that are going on.
And, yeah, thank you so much forjoining us.
Look forward to having you backagain sometime.
Chelsea (43:17):
Thank you. It was great catching up.
Jerod (43:26):
Alright. That's our show for this week. If you haven't
checked out our website, head topracticalai.fm and be sure to
connect with us on LinkedIn, X,or Blue Sky. You'll see us
posting insights related to thelatest AI developments, and we
would love for you to join theconversation. Thanks to our
partner Prediction Guard forproviding operational support
for the show.
Check them out atpredictionguard.com. Also,
(43:49):
thanks to Breakmaster Cylinderfor the beats and to you for
listening. That's all for now,but you'll hear from us again
next week.
Popular Podcasts
Stuff You Should Know
If you've ever wanted to know about champagne, satanism, the Stonewall Uprising, chaos theory, LSD, El Nino, true crime and Rosa Parks, then look no further. Josh and Chuck have you covered.
Las Culturistas with Matt Rogers and Bowen Yang
Ding dong! Join your culture consultants, Matt Rogers and Bowen Yang, on an unforgettable journey into the beating heart of CULTURE. Alongside sizzling special guests, they GET INTO the hottest pop-culture moments of the day and the formative cultural experiences that turned them into Culturistas. Produced by the Big Money Players Network and iHeartRadio.
Dateline NBC
Current and classic episodes, featuring compelling true-crime mysteries, powerful documentaries and in-depth investigations. Follow now to get the latest episodes of Dateline NBC completely free, or subscribe to Dateline Premium for ad-free listening and exclusive bonus content: DatelinePremium.com