September 11, 2025 • 52 mins
What happens when you combine Hollywood-level storytelling with the intimacy of audio?
Oz Woloshyn and Mangesh Hattikudur, co-founders of Kaleidoscope, have created some of the most compelling podcasts in the medium by focusing on one simple truth: people love feeling smart, especially when learning feels like an adventure.
Kaleidoscope's award-winning shows transport listeners from the Amazon rainforest in search of rare cacao to a Soviet space station where a cosmonaut finds himself stranded when his country collapses beneath him. These aren't just podcasts – they're audio journeys that hook listeners with compelling facts and emotional moments that make seemingly niche topics universally fascinating.
As we navigate the rapidly evolving landscape of AI and content creation, Oz and Mangesh offer refreshing perspectives on why human creativity remains irreplaceable. While they've embraced AI tools to streamline production (revolutionizing their workflow with transcription and editing software), they emphasize that technology serves creativity, not the other way around. The subtle imperfections of human communication (which Mangesh calls the "wobble") create emotional connections that no algorithm can replicate.
Kaleidoscope's approach to podcast creation offers valuable insights for educators and parents who are concerned about how to engage students in an era of shrinking attention spans. Whether you're a podcast creator, educator, or simply curious about the future of storytelling, this conversation will leave you rethinking how we connect with audiences in meaningful ways.
Learn more about Kaleidoscope:
aiEDU: The AI Education Project
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
Alex Kotran (aiEDU) (00:04):
Yes, welcome to AIEDU Studios, where
we're building either.
We don't know if it's a show,it could be a podcast, maybe
it's a YouTube channel.
The goal is bringinginteresting people on to talk
about the intersections of ournerdiness, talking, obviously,
about artificial intelligenceand education, but also bringing
people who are at the forefrontof different industries,
(00:26):
different spaces, places thatare being impacted by and also
driving forward innovation,artificial intelligence and sort
of help us sort of paint thispicture of what the future
really looks like, because wehear a lot about the future of
work and the age of artificialintelligence.
I think most people don'treally understand what that
means for them, depending ontheir interests, and so, to that
(00:47):
end, we have not actually hadan opportunity to invite folks
in the media space, in thecreative space, and so joining
me today are the co-founders ofKaleidoscope, mangesh Hartikadur
and Az Velashin, and we'regoing to hear more about what
they've built, what Kaleidoscopeis.
They have blazed the trail thatAIDU is just beginning, and so
(01:09):
I'm personally excited.
Oz Woloshyn (Kaleidoscope (01:13):
Maybe we'll have a little bit of a
mix.
I don't know about that.
Alex Kotran (aiEDU) (01:16):
We will certainly be cutting all the
instances where they'reproviding me feedback and
telling me all the things I'mdoing wrong.
But yeah, thank you so much forcoming on to the show.
And why don't we start withjust tell us a little bit about
you know, tell us aboutKaleidoscope and tell us about
what your life has been likeover the last couple of years,
as you've been sort of buildingthis new studio, this new
(01:38):
company.
Mangesh Hattikudur (Kaleid (01:39):
Yeah , so Kaleidoscope is a podcast
company.
We love the audio space.
We think there's something sobeautiful and transportive and
the fact that you can be insomeone's ear and that
connection that you have is someaningful.
This company we really thoughtthat it was just an incredible
(02:07):
way to tell stories from aroundthe world and almost treat it
like an engine for empathy, andthat stories were a way to just
an incredible lens forunderstanding the world.
Alex Kotran (aiEDU) (02:18):
Yeah, and so it's interesting because the
even the project of what we'redoing here, you know, along
maybe 60, maybe even 90 minutes,if we have time, um, to really
get into it it's.
It's really pushing backagainst the trend culturally
that we've seen content go interms of, you know, we went from
you know, short videos to nowliterally like three seconds of
(02:42):
just dopamine you know it was nocontext.
Context um brain rot, as thekids in the classrooms that we
teach I mean it's literally likethey're.
They're sort of likeself-indulgent in the way that
they talk about brain rot.
Well, it was the oxford, uhdictionary word of the year, I
think yeah, it was um, butpodcasts and the type of and
especially the type of podcastsyou do, which is not just sort
(03:04):
of I mean, you do some talkshows right, but you also do
long form storytelling that'sright that builds like series um
.
Folks have probably probablyremember the uh serial and sort
of like how that brought sort ofthis like sort of storytelling
format um into this space andyou've continued that work um.
Can you tell us about some ofthe stories that you've told in
(03:26):
that format?
Mangesh Hattikudur (Kaleid (03:28):
Yeah , definitely, like I said, part
of what we try to do is reallyvalue the listener's time and
really engage them and entertainthem.
There's a show about cacao andwild cacao that we did called
Wild Chocolate, but the way wepitched it was, and it's
(03:48):
ultimately an environmentalstory, but we pitched it as an
Indiana Jones of chocolate.
Or we did a story that was abouta Soviet cosmonaut who got
stuck in space because theSoviet Union disbanded beneath
him and he was stuck there for300 plus days and his only
connection to the world was thisham radio operator in Australia
(04:11):
that he managed to connect with.
Or I did a show that ended upbeing much more personal than I
was expecting, but it was abouthow astrology is seen so
differently in the East than theWest and and and the ways that
you know it's.
It's such a pseudoscience hereand and it's it's a way of life
(04:34):
in India and and and these,these stories that felt really
fun and and emotional and likebig adventures and in some ways
we were trying to trying to goagainst the grain of of, at
least in the beginning of of all.
The really long chat shows thatthat they were going on for
(04:56):
like three or four hours Likethis.
These were things that weremeant to pull you in and
entertain you and, at the end ofthe day, leave you feeling
smarter.
Alex Kotran (aiEDU) (05:07):
Yeah, we were just talking before, um
picking things off about some ofour favorite sort of youtube
shows, and there's this model,just like, bring people together
, put them on some couches ikeacouches maybe um, just banter,
and you know, I think like I'llgive some props.
It's still longer form and'sopportunities to at least, you
know, dive into a little bitmore depth of content.
(05:28):
It's not necessarily alwayseducational content, but there's
a lot more work that goes intotelling a story about cacao,
which could be interesting, butyou know, I could also see this
be I mean, like you can imaginea lecture and something that's
informational informational andpeople obviously have short
attention spans.
What are some of the ways thatyou break past?
(05:49):
Because, also, your shows areamong the most listened to shows
on Apple Podcasts and you'vewon a few awards.
Mangesh Hattikudur (Kaleid (05:56):
Yeah , and, to be clear, I actually
really value really longinterview shows and I think
different shows have differentpurposes.
Long sort of interview shows andI think different shows have
different purposes, right, andso, in the same way that sports
radio people would turn it onand keep it on in the background
, as you're like, maybe paintinga room or you know something to
keep you company, as you'redoing that the commute to work,
(06:18):
where you might have 25 minutesor 30 minutes in a car and you
want to be told something veryclearly, or you want to learn
specifically about markets orwhatever.
Right, like, every podcast hasits own flavor but also purpose,
and so I think we went in veryintentionally with the idea of
telling these really gorgeousstories and we tried to invest
(06:42):
in the sound design in a waythat other people hadn't.
We tried to look at the spacein a way that especially like,
if you think about when we'dstarted this company, it was
coming out of the pandemic andso it was people had been stuck
in their houses for a very long.
In the lockdown period peoplehad been stuck in their houses.
They'd still had this liketremendous wanderlust and to be
(07:03):
able to place someone in adifferent environment through
their earphones was somethingthat we felt was really
meaningful and how many shows doyou have out right now?
how many shows have we done?
Oz Woloshyn (Kaleidoscope (07:15):
We've done more than 10 and we have
another 10 to 12 coming out thisyear.
I think about half of theportfolio of what we do is going
to be weekly more chat stylecontent and half is going to be
weekly more chat style contentand half is going to continue to
be these highly producedcinematic stories.
Mangesh Hattikudur (Kaleidosc (07:30):
I mean, in my mind, what we're
really trying to do is alwayscreate someone's favorite show,
and when I was at iHeart we dida show called Call your
Grandmother and it was an80-year-old host and a
90-year-old host and they werelike a hoot and you wanted to
spend time with them and at theend of this half an hour with
them you're reminded how muchyou love your own grandmother
(07:55):
and for someone like me whosegrandparents have all passed, it
was just such a pleasure and tobe able to create that space.
There's so many flavors of showto provide that, I think, and
there's so many lanes.
I don't really think of otherpodcasts as competition right.
I just think there's space toreally sort of provide something
(08:18):
beautiful and let audiencesfind it and evangelize it.
I really do think good contentmakes its own luck find it and
evangelize it.
Alex Kotran (aiEDU) (08:24):
I really do think good content makes its
own luck.
I want to talk about thisformat and we're going to go a
little meta and maybe you caneven imbue some sage wisdom on
me and my team as we're tryingto figure out how to build our
own.
Again, I really don't know ifthis is a show.
I think, like right now I'mthinking about it more because
and you will be the first totell me right that, like when
you pitch an idea for a show,there's a really coherent sort
(08:46):
of like framing.
You have a sense of the formatand we're still experimenting.
And so I'm trying to give myselfa little bit of, you know,
wiggle room to really kind ofsee what works and see what
works for our guests.
Oz Woloshyn (Kaleidoscope) (08:56):
Well , that's fun.
I mean you find that you findthe format in the doing right,
so you don't have a firm launchdate.
It's quite nice because you caniterate and iterate until you
find exactly what works for youand then take that out as the
product, or you can launch itand iterate it in progress,
especially these morechat-driven formats.
You see how much they evolvefrom episode 1 to episode 10 to
episode 100.
Alex Kotran (aiEDU) (09:17):
I've only listened to three of the shows
that you've released.
The first three, wild Chocolate, skyline Drive and the Last
Soviet and yeah, those wereincredible.
Also, the Last Soviet,amazingly narrated by Lance Bass
, who I never realized heactually served in the.
It was like a Russian sort ofcosmonaut summer camp.
Mangesh Hattikudur (Kaleidos (09:38):
He was prepping to go to space and
he spent time at Star City,learned Russian, did all this
testing.
It was pretty incredible what Ithink.
Alex Kotran (aiEDU) (09:47):
What stood out to me is I never had an
interest in chocolate.
In fact, I don't really evenlike chocolate, although I did,
I do have the, the, the, thecollab that y'all did.
um, that chocolate was amazing.
Um, I actually developed anappreciation for chocolate, but
basically it was not the type ofshow I would normally have
committed to.
But there was something aboutbeing transported to, like
(10:09):
literally being in the Amazon,the interviews with people in
these remote villages, and itbrought me back to this.
Yeah, it's sort of like thisage of like much more
adventurous, like the value ofmedia was sort of like bringing
you to places that you couldn'tnecessarily go yourself.
Yeah, um, and I'm someone who I,I sort of I respect, you know,
(10:30):
the marvel universe and Irespect sort of you know, these
like big blockbuster movies, um,but it does feel like that we
have a lot less of thatculturally now.
You know, you're sort of oryou're transported to 50 or 100
different parts of the world inthe span of like two minutes on
Instagram or TikTok and youdon't, you don't really connect.
There's no, there's noopportunity to actually connect
(10:51):
and sort of hear the almost likethe tenor of this old woman's
voice and sort of like theroughness, and you can kind of
tell that she's been likeworking in, you know, in this,
in this trade, for 30 years andlike the fact that you can hear
that in her voice is sort of asubtlety to it, and I'm curious
if you feel like this is a?
I mean, do you feel like thismedium is underappreciated?
(11:13):
I mean, everything now feelslike niches.
Everybody has their, you know,there's folks spending 20 hours
a week watching people playvideo games, not playing video
games, watching people playvideo games, right, and so, and
actually my favorite, so myfavorite niche is it's a YouTube
channel called Townsend andsons, and this is this guy, john
Townsend, in Indiana, and he,uh, is a reenactor, but his,
(11:34):
specifically, he reenacts um18th century cooking and so he
has the whole gar built like ahut and he had all the materials
and he like teaches, he likelooks into old recipe books and
like cooks 17 or 18th centurycuisine, um, and he's millions,
like 2.5 million followers.
His videos get hundreds ofthousands, sometimes millions of
views, um, and I love that.
(11:56):
I love that.
That's actually like part ofwhat the internet has done is
you don't have to have like thebest show that literally every
single person is talking about.
You have to have like the bestshow that literally every single
person is talking about.
You have to figure out how tohook you know just that right
type person, um, and and so youhave three shows that are kind
of all over.
Mangesh Hattikudur (Kaleidosc (12:15):
I mean, like you know, chocolate
space astrology, um, but I Ithink it's one of those things
that people don't know whatthey're interested in until
they're shown right.
And so one a a show aboutchocolate.
Probably you wouldn't expectthat.
It starts in the Amazon with ajournalist circling because he
can't land a plane and itfinally lands and the whole, and
(12:38):
they finally find a patch ofland to land on and the plane
comes to a stop and then fourmen with guns come out and tell
them, you know, like, this is adrug lord's land and you
shouldn't be landing here, right?
I mean, like, that isn't theexpectation of what a chocolate
show should be about.
But also it is engaging, right,and I think that I certainly,
(13:03):
like you know, netflix has showsabout glass blowing that, like
I wouldn't have.
I don't care about glass blowing, or I don't think I do, but
then I watch.
You know, like I don't reallycare that much about baking.
I watch great british bake offall the time.
You know like, and suddenly Ido care about it.
You know like, there there's somany things, so many things to
appreciate and so many ways toappreciate it, and I feel like
(13:24):
you just have to be led in inthe right way.
Oz Woloshyn (Kaleidoscope) (13:27):
I think not for nothing.
I mean, the connective threadbetween those three shows is
their quest adventures thatrelate in one way or another to
science and, in some cases,technology.
So with Wild Chocolate, youhave these people who are so
obsessed with this pre-columbian, semi-mythical cacao in the
(13:48):
amazon that can only grow if ithas enough uh canopy coverage
that it's in the dark for mostof the day.
And these people risk it alland bet their whole life savings
on trying to find these beansright.
And so you go on this adventurewith them.
In the process, you learn aboutthe amazon, you learn about the
history of cacao cultivation,you hear about his role in
pre-Columbian culture In thecase of the Last Soviet.
(14:09):
You have a guy who goes to space, is in the Mir space station.
The Soviet Union collapsesbeneath him and he's faced with
a choice Do I come home andabandon the Mir space station to
spin off into space, or do Iwait here for as long as it
takes until someone can bring mehome?
Because this mere space stationis the last redoubt and the
final pride of the of thecountry and the empire I've
(14:32):
given my life to.
And again, in that, you get tothe history of the space race,
you get to rocket technology,propulsion technology in the
case of mangashe's show, uh, theastrology show, skyline drive.
This was, in a sense, askeptic's guide to, you know, a
spiritual topic.
Alex Kotran (aiEDU) (14:50):
Because you , your family obviously believed
, believes in astrology.
You were yourself a skepticgoing into the show.
Mangesh Hattikudur (Kaleid (14:58):
Yeah , I mean, I think it's one of
these things that I don'tbelieve in astrology.
But astrology has happened tome, right, like my parents met
because they were introduced toone another because their charts
matched, you know, and it wasalmost like this dating service
where, like they could, onlytheir family only introduced
them to people who they matchedup with and they dated for a bit
(15:18):
, liked each other, you know,ended up getting married.
But I'm only here because ofastrology in this strange way
and so, and you also find,astrology, whether you believe
it or not, has affected theworld drastically in that one
search for understandingastrology led to an
understanding of astronomy.
Right, it was only through,like it's the desire to
(15:41):
understand the future that ledus to, like, investigate the
stars closer.
But also, astrology is whatconvinced Reagan for the nuclear
disarmament, to have talks withGorbachev.
(16:02):
It was his astrologer, who wasin his ear, who said now is a
great time to do this, this isthe time, this is the reason.
And he was such a believer thatif you look at the times of his
inaugurations or the times he'ssworn in, they're at
astrologically significant timesand moments.
And similarly, boris Yeltsin,mitterrand, like all these
(16:25):
people you know, it's one ofthese things that when people
need hope or they're in grief orthey need like consoling, right
, like people turn to the stars,in the same way they turn to
religion.
And so when people don't knowwhat to do, it's convenient.
(16:47):
And so there are teams in SanFrancisco baseball teams that
don't know which players to pickin drafts, and so they use
astrologers to guess which onewill have better longevity with
the team.
And we don't see this.
It's invisible.
But when we're rooting for ourfavorite teams, we don't know
(17:08):
that actually astrology hasplayed a part.
Alex Kotran (aiEDU) (17:12):
Okay, so you talked about making obscure
topics engaging.
We work with teachers every daywho this is their daily
struggle, whether they'reteaching linear algebra or the
fall of the Soviet Union, toactually come up with an example
that maybe is relevant, or theScarlet letter.
(17:32):
You know, these are some thingsthat students occasionally will
sort of develop an interest in,but for the most part kids are
looking out the window waitingfor the bell to ring and I'm
curious, like is there anythingthat teachers can learn from the
work that you do tellingstories?
Mangesh Hattikudur (Kaleidosc (17:53):
I think one of the things I
learned along the way is thatpeople love feeling smart, and I
think that part of what we dois try to infuse our
storytelling with a lot of joyand a lot of quirkiness, and the
weird facts on the edges oftenare the things that end up
(18:16):
pulling you into stories.
And so something like the LastSoviet was, I believe it was
like a one-off fact about aSoviet that had been stuck in
space that your friend told youabout, and then we were like
wait, that can't be true.
And then we looked into it andthen we dove into it and then we
realized, oh, this isincredibly rich story, and I
(18:38):
feel like, if you can figure outwhat that first fact is, that
gets someone curious, that's funto repeat, right.
Like whether it's the fact that, like I think, belgium tried to
use cats to deliver the mail touse cats to deliver the mail,
(19:04):
you know, and they equipped 37postal cats with, and the whole
experiment was a disaster.
But or that, like you know,there are pirate ships that ran
out of cannonballs and wonbattles with Edom cheese.
Or or, you know, like therethere's so many ways into
history and math and science andstorytelling and even like
Pythagoras, is more interestingwhen you know he had a cult and
(19:25):
wouldn't eat beans and, like youknow, I, I think there are so
many aspects that let you feel alittle smarter and and if you
can walk out of a room with thatsort of joy, that sort of sense
that you want to evangelize atopic that makes learning fun.
Oz Woloshyn (Kaleidoscope) (19:45):
Mary Poppins spoonful of sugar.
We had a meeting yesterday withthe former director of the
Exploratorium here in SanFrancisco and he had some
interesting insights for us.
One of the most fascinatingthings he told us is the number
one complaint about theExploratorium is it's impossible
, if you're a parent, to getyour kids to leave.
(20:06):
But he had this interestingphrase that's strategy equals
sacrifice.
And so he basically said youhave to decide where you want to
be.
Is it engagement or is itinformation that you want to
lead with?
I think, in a culture of TikTokand like dopamine highs, to be
an effective educator you haveto figure out how to meet people
(20:29):
at the place of engagement andthen, you know, educate them.
Following that, I think comingin with the assumption that
people want to know facts toimprove themselves brings you a
very narrow audience of studentsin this day and age.
And so far be it from us togive advice to teachers.
But I think thinking about notbeing allergic to the word
(20:53):
engagement even though it's alittle bit of a dirty word given
the TikTok of it all is crucialto winning any kind of audience
.
Mangesh Hattikudur (Kaleidosc (21:01):
I mean, I see it with one.
My mom was a preschool teacherand what she was so good at was
finding out what each kid wasinterested in and then tailoring
the subject to them.
And so, uh, she would.
She would make the room feellike an airplane and give
everyone passports and take themto Egypt, and take them to
(21:23):
various places, and like and youknow where you got it from Um
but.
but you know at the same time,like she would get these
preschoolers and teach them likeoh, this is, this is how you're
going to finger paint like PaulClay.
This is how you're going tofinger paint like Paul Clay.
This is how you're going tofinger paint like Van Gogh.
This is how you're going to dolike Monet.
And it was just unbelievablethat these kids would walk out
(21:46):
of these classrooms like feelingso confident and so good, and
that they could, a lot of times,like older siblings, would cry
because, like their younger kids, knew more about the solar
system than they did, but I sawit with my own
kids where this incredibleteacher brought worms into the
(22:06):
classroom as the class pet.
Everyone else has one pet, butwe've got 500 here and they
learned everything.
The places where disciplinesmix and they can talk about you
know the.
The places where disciplinesmix and they can talk about the
environment.
They can talk about biology,they can talk about like uh, uh,
you know food systems and allthese things.
Like I, I think that kids aresponges and and they they do
(22:32):
want to learn.
You just have to show, you know,you just have to engage them a
little bit.
And I realized that sounds easyor something, and obviously a
teacher's work is is is thehardest thing in the world and
and like I mean I think invarious places, like my
journalism teacher in in highschool is the one who who made
(22:55):
the intro to an investor, whoinvested in mental floss and he
invested in kaleidoscope, youknow, and like I still
communicate with my liketeachers from third grade and
sixth grade and people wholisten to Skyline Drive, and I
mean I was really lucky in thatthere was such a focus on
(23:17):
creativity in my school and thecreative applications of sort of
the inquiry-based learning wedid space I found is that when
you talk to a lot of folkswho've been doing this work and
it's, and certainly that work ofteaching is thankless and
(23:39):
underpaid and yes, you getsummers off, but it's even the
summers are not as long aspeople assume yeah, Um, and a
lot of them have these sort oflike uh, origin stories that
start with that one teacher, um,that had a transformational
impact in their life.
Alex Kotran (aiEDU) (24:00):
And yes, it is easier said than done, and
preschool teacher, elementaryschool teacher, fourth, fifth
grade there's a lot of scope tobe creative and sort of like
bringing these differentexperiences in as you go further
along and even when you hitfifth grade.
In some cases, teachers arestarting to be burdened with
standards and all the boringcontent stuff that they have to
(24:21):
teach, and so that is often thechallenge.
It's not necessarily thatteachers are not creative or
just don't realize it.
Mangesh Hattikudur (Kaleidosc (24:30):
I think it's much easier when the
kids are younger, I meanespecially as you're teaching to
tests and et cetera, butthere's still so much space.
Alex Kotran (aiEDU) (24:39):
Yeah.
So let me, let me, um, um, so II say all of that because, as
you were sort of describing someof these ideas to me, I was
trying to think about, like howcould a teacher, kind of like
tomorrow, take that and make anactionable and it might be hard
for a teacher to figure out thatweird, quirky connection to
whatever it is, the lesson thatthey're teaching.
(24:59):
We don't normally push AI as atool that teachers should be
using every single day acrossthe.
We don't believe AI should bereplacing the work of teachers,
but it occurs to me that this isactually a really interesting
use case.
Oz Woloshyn (Kaleidoscope) (25:15):
Yeah , what a great you could say
could say you know, I havestudent a who's this, these are
their passions, and student b,these are their passions.
And students see these, theirpassions.
What, what are some suggestionson the best way to engage them
on the topic at?
Alex Kotran (aiEDU) (25:28):
hand.
Oz Woloshyn (Kaleidoscope) (25:29):
I think that would be a perfect
use case.
Alex Kotran (aiEDU) (25:30):
I think it'd be a perfect use case.
Um, and, and the call out isstill that you need a teacher to
.
You know, some percent of thetime, probably a big percent of
the time, in that specificprompt it's going to hallucinate
, make it up.
So, again, not replacing thework of teaching.
You're still gonna have to do alittle bit of homework, but I
think that's the sort of thingthat could make this a lot
easier.
(25:50):
And so when we think about howdoes ai fit into the classroom,
you know well it can makelearning more engaging and more
personalized.
It just occurred to me thatthis is maybe a really like a
tangible example of that it'ssubtle but it actually to your
point.
like the point is the hook.
You know you spend a lot oftime, I assume like significant
amount of time thinking aboutlike that first episode.
Oz Woloshyn (Kaleidoscope) (26:11):
It's always the hardest one so much
time.
Mangesh Hattikudur (Kaleido (26:13):
And but I, yeah, exactly, I don't
want to be dismissive of thework that teachers do, because I
, I saw it firsthand, right Likeand and also my mom used to
make us uh, she pulled my dadand me into like turn
refrigerator boxes into likevarious things, and she was like
, well, this one's a castle.
And I was like I, amma, we justmade a.
(26:34):
And she's like no, no, that wasa rocket ship.
Oz Woloshyn (Kaleidoscope) (26:36):
This is a castle.
Mangesh Hattikudur (Kaleid (26:37):
This is something else, but I was
constantly helping her cutthings and make things and work
sheets and whatever, andobviously that work is so
important.
But I think the benefits to thekids and the way kids walked
out of that classroom soconfident, like I think that's
(26:58):
what I was fascinated with, waslike these kids that really had
a stronger sense of self becausethey knew things and then could
go into kindergarten, where thesocialization is really hard
right, but if they've got theacademics covered then, like,
they can focus on thesocialization is really hard
right, but if they've got theacademics covered then they can
focus on the socializing.
Oz Woloshyn (Kaleidoscope) (27:20):
One of the ways I sometimes like to
prepare for interviews is toopen an LLM and say imagine you
are a Goldman Sachs analyst andyou're going to interview this
person.
What are your top threequestions?
Now imagine you're the editorof Wired magazine.
And now imagine you're KatieCouric magazine.
You're going, and now imagineyou're, you know, katie couric,
and so basically to, to, to tohave the llm like, imagine its
(27:41):
way into different interrogationor questioning approaches and
then kind of understand, usethat to piece together okay,
what is the most interesting wayto engage this person?
Mangesh Hattikudur (Kaleidosc (27:50):
I think I think that's what's
what's really fascinating is,like the, the way you get so
many options out of this.
Oz Woloshyn (Kaleidoscope) (27:56):
I won't say what should I ask this
person?
Mangesh Hattikudur (Kaleidosc (27:57):
I will say provide inspiration on
an approach and I'll choosebetween suggested approaches I
mean I we were talking aboutthis earlier that like I have a
friend who started this companycalled pseudo right, and it's to
help writers, uh, particularlywhen there's like writer's block
, to give you like options ofwhere your story could go and
that inspiration, or taking whatyou've written.
(28:18):
And I had this friend, danielalicon, and and he uh said that
he took a map and he like redrewit and tried to imagine a
landscape and like had to do allthis work to figure out, like
what this imaginary land wouldbe.
But ai can do that for you,right, it can create what your
fantasy visuals are to inspireyou and to give you even more
(28:41):
clarity on what you want to do.
And so I think there are allsorts of ways of creating
options for yourself when you'reunsure.
Alex Kotran (aiEDU) (28:50):
This is a beautiful segue, because I
wanted to have an opportunity toask you about this question
that you've already sort of ledus to, which is what are the
ways that AI either is helpingyou do your work or could help
you do your work?
And before we dive into it, Ijust want to sort of set the
stage for the type of folks thatmight be listening, in short,
(29:13):
teachers, maybe someadministrators, but also parents
, and I talked to a lot ofparents who are like my kids.
In fact, our director ofcommunications is his son loves
creating art.
Yeah, um, not like youtubevideo, like like literal, like
draw, like collages and painting, spray paint.
Um, we get questions from a lotof parents that are like my
(29:36):
kids, are really interested inart, media, design, and they're
reading the news they're seeing,maybe they're playing with Dali
or Midjourney, and you type inone sentence, in not even a
minute but like five seconds,and you get this very compelling
painting, photograph, 3drendering, and I think I've
(30:00):
actually seen some audio bookson Amazon now that are narrated
by AI Voices.
Oz Woloshyn (Kaleidoscope) (30:05):
AI Podcasts.
Alex Kotran (aiEDU) (30:08):
There's AI Podcasts, and so, to the
discerning parent, the questionthat we get is is there going to
be space for for creatives ifai is going to be able to do all
this stuff that wetraditionally associate with
creative work?
Um, is this kind of a dead endfor for my kids, and should they
(30:28):
be learning something else?
Maybe they should be learningai stuff instead of instead of
doing art.
And before I weigh in with howI respond, I'll tease it and say
that, um, I I vehemently, uh,uh, not aligned with that
perspective I think that thereis this important room for
creatives, but you're on thefront front lines of this um and
(30:50):
we're far enough along now.
Chat gbt came out at the end of22, so it's been what like two
and a half two two years andsome change.
Um, okay, is it?
Has it actually changed the wayyou work meaningfully?
Has it changed the way you hire?
Is it changing the way you'rethinking about your strategy?
Um, why don't we start withlike have.
(31:10):
Are you using ai day to day?
Did you use ai in the last week?
If so, how?
Oz Woloshyn (Kaleid (31:15):
Practically speaking, we use AI all day,
every day, in fact.
It's completely integral to ourproduction process, not in the
form of telling us what to say,but in the form of actually
efficiently editing.
And, magus, you want to explainabout Descript and how we use
it?
Mangesh Hattikudur (Kaleid (31:32):
Yeah , we use a program called
Descript and we import our filesinto it and it basically
transcribes all your content.
It can filter out, sounds forlike if there's mic bleed, it
fixes that and it allows you toedit like you're editing a text
document, so you just delete asentence and it cuts it from the
(31:55):
audio.
Oz Woloshyn (Kaleidoscope) (31:55):
You don't have to.
You know you don't have to haveany audio.
And now to have a perfectproduct, of course it's not good
enough, but to be able to gofrom a lengthy interview to an
effective rough cut, you can dothat as easily as you can edit a
Google doc, whereas even threeyears ago that would have
required.
Mangesh Hattikudur (Kaleid (32:13):
What was?
Alex Kotran (aiEDU) (32:14):
it like before you had the script.
Mangesh Hattikudur (Kale (32:15):
Before that, that would have required
specialists.
What was it like before you hadDescript?
Before that, you would have tosend the files to a transcriber.
Oz Woloshyn (Kaleidoscope) (32:21):
You would send, which cost like a
dollar a minute.
I mean, it was expensive.
Mangesh Hattikudur (Kaleid (32:23):
Yeah , it was too expensive to get
stuff.
Oz Woloshyn (Kaleidoscope) (32:26):
When I started podcasting in 2019,
it was too expensive to get allour audio transcribed.
It was too expensive to get allour audio transcribed.
Can you believe that?
So the production flow we wouldliterally decide?
Is this interview importantenough to spend $300 to get
transcribed, to figure out whatthe best moment is without
listening to it for three hours?
Mangesh Hattikudur (Kaleido (32:46):
And also it often wouldn't be you
listening to it, it would besomeone else listening to it and
you'd hope they'd make thechanges.
And you try to consolidatethose two things and try to meet
somewhere what's closer to yourvision versus.
You know, I do a show calledpart-time genius.
It's a very quick show but butI look at it, I make the tweaks
(33:11):
that I heard as I was recordingsomething.
I knew this sentence was wrong.
I knew I made a hiccup here,all these things and then I send
it to get mixed and masteredand all the other steps that
used to all the like QC, thequality control that used to
take place have been eliminated,like it's a much, much faster
and more seamless process.
Oz Woloshyn (Kaleidoscope) (33:32):
But this is a great example.
This is not like we're notgoing out and saying like or
sometimes we are, but in thiscase not.
How do I use AI?
It's more like it's not.
We don't think about it asusing ai, we think about it as
using an application.
In fact, we've forgotten howbad it was or how difficult it
was before the script.
I think you know one of theother areas, obviously creative
inspiration, and I'm super,super happy and relieved to
(33:54):
report that I think ai does agreat job of summarizing
information.
Obviously, often with halluc ithas no ability to write with
voice, and even like NotebookLMI is pretty interesting.
Like it can make a prettycompetent two-way conversation
podcast between like a malevoice and a female voice.
It's also entirely reliant onwhat you feed it.
So like.
(34:14):
If you just say like here's 10articles about like the
Superbowl, it will do a reallyboring job, whereas if you say
like okay, here's 10 articlesabout like the Superbowl in 1963
.
And I want to understand howthis social phenomenon affected
the you know the sportsenvironment, you might get
something interesting, but atthat point you've done a lot of
the creative labor yourself,right?
(34:35):
So you know this, I think.
Coming back to your questionabout technology and and what
should parents think about fortheir children, I remember so
vividly, like the 2010 era learnto code.
Like forget the humanities,learn to code.
Thank god, that was the worstadvice ever, right, um?
And similarly now, like youknow, I read jack clark, one of
the anthropic foundersnewsletter a couple of days ago
and obviously he has like avested interest, but his, his
(34:56):
main point was, like the winnersof the future will be the most
curious people, the people whokeep finding the most
interesting ways, the mostforward-thinking ways to ask um
questions.
And how do you get to be a goodquestion on asker?
By feeding your curiosity as achild, as a young person.
So I I really think that likethe future is is is not about,
(35:16):
um, you know, getting being like.
Oh well, ai can like make animage, so why would I become an
artist?
It's really.
How do I develop into the typeof person who can think
creatively about how to make themost of technology to serve me
and my community and my society?
Alex Kotran (aiEDU) (35:31):
What I'm hearing is the type of work that
you're describing.
That's being replaced.
I don't think anybody is inschool thinking I want to be a
transcriber.
I don't think anybody's inschool thinking I want to be a
transcriber.
I don't think anybody's inschool thinking I want to spend
12 hours a day in Photoshop, youknow, removing objects from the
background of videos.
Mangesh Hattikudur (Kaleido (35:51):
And it was meaningful work for
people for decades.
But yes, I agree.
I mean, like I don't think thatwas anyone's ambition, right?
But yes, I agree.
Alex Kotran (aiEDU) (36:01):
I mean, like I don't think that was
anyone's ambition, right?
Yeah, so there is a questionabout what, like where do those
dislocated workers go?
There's this sort of you know,the talking point is we're going
to empower them to do moreinteresting creative work.
The reality is, you know, thereare sort of there aren't that
many roles at companies likeKaleidoscope to do creative work
(36:22):
.
But I want to put a pin in thatconversation because I want to
actually hear more about Sorry,can I just say one other thing.
Mangesh Hattikudur (Kaleid (36:31):
Most people don't know this, but our
album art for our first sixshows was all done on an early
version of Dolly.
The art was done on it and thenwe got an artist to to draw
over it and highlight certainelements and stuff, but but the
the starting point was was aiokay.
Alex Kotran (aiEDU) (36:50):
so that is fascinating because when you
first told me that, I wasincredulous because I I sort of
pride myself at being able tospot AI art and my I like Spidey
senses didn't tingle when I waslooking at your album art.
It does for a lot of other stuffon, like Spotify and Apple.
But I understand why now,because you, you also brought in
(37:11):
that human component to take itto that next level, and I think
that's that's to me this likereally interesting takeaway,
which is the there's alwaysgoing to be this need, if
everybody has access to thisbaseline of content, this common
denominator of justAI-generated whatever, there's
(37:33):
going to be so much of that thatI think people are going to
actually seek out and identifyquality and the only way to do
something that ai can't do is totap human potential and human,
uh, creativity, um, has itimpacted hiring?
I mean, like, have you?
I guess one question is haveyou hired anybody just to like
be the ai tool user?
Because I think that's whatsome people are saying is that,
(37:54):
oh, you don't need to learn, youdon't need to don't study art,
just learn how to use ai andprompt engineer, um, and I'm
interested in that because I'mcurious if that's the type of.
Is that a job posting onKaleidoscope's page?
Mangesh Hattikudur (Kaleidosc (38:09):
I don't think so.
I mean, at least from myperspective, we are looking for
people with voice and outsiderideas that are super creative,
and even when I was at I heart,a lot of people would just hire
people from the radio industryto to come in and work because,
like this is versus like wehired, um musicians who'd never
(38:32):
done podcasting, we hired, likeyou know, people who were uh in
all sorts of other fields.
There was someone who used tolike help, um uh kids who were
ill create poetry and art, but Ithink what we're looking for is
really creative, smart peoplewho have a point of view, who
(38:54):
can learn the tools and use thetools to advance our own sort of
point of view or creativity.
Alex Kotran (aiEDU) (39:01):
Yeah, I think we've had the same
experience, like we don't reallyuse AI very much for curriculum
development.
If you have a certain level ofquality, it's actually very hard
to get.
I mean, I've done, I've triedusing ChatGBD to write stuff.
Oz Woloshyn (Kaleidoscope) (39:17):
Same .
Alex Kotran (aiEDU) (39:18):
I hear people saying, oh, it can write
emails, it can write pressreleases Like I disagree.
I hear people saying, oh, itcan write emails, it can write
press releases Like I disagree.
I think that it can.
I think if you're trying towrite, let's say, a hundred blog
posts a month, then, yes,chatgbt is probably useful.
I don't know why you would wantto do that.
Oz Woloshyn (Kaleidoscope) (39:32):
If you're trying to write, well,
maybe if SEO right people don'tlike written content on the web
uh, except for from superprestige publications like I
think, or in forums like redditand stuff is like losing very
rapidly losing its authority,which I think is a big
opportunity for audio because,yes, you can make a deep fake of
(39:53):
my voice and theoretically havechat, gpt, write something,
then have the deep fake read myvoice, but, like you can hear
that pretty easily and I think,the opportunity for podcasting.
The reason we had this podcastelection in part is because it's
a very trusted medium and Ithink will retain a moat in that
respect.
People looking for guidance um,you're way less swayed by the
(40:14):
top 10 google results six monthsfrom now than you were a year
ago we're in a bit of the thewild west right now in terms of
of media.
Alex Kotran (aiEDU) (40:23):
I mean, I think that there's sort of a
definitive shift away fromlinear tv.
We're using a lot of industryspeak.
You said moat, I'm gonna uh,linear tv is sort of just like
broadcast television, cable, newcable channels, um, and you
know people were moving toNetflix.
I honestly YouTube, especiallyYouTube premium, without the ads
(40:43):
, is like totally dominated theway I consume media.
Obviously, people are onInstagram Tik TOK Um, and that's
especially pronounced in thenew generation, right Like Gen Z
, gen Alpha Um, and I'm so.
So maybe we can close with alittle bit of um, sort of meta
(41:04):
introspection about sort of whatwe are doing here at AIEDU
studios, um, and I I'd like youto just sort of I'd like your
take on this in terms of what wecould be doing with the format
Um, because one of the thingsthat I found is there's, there's
now, yes, there's, this shiftto sort of this open universe of
(41:24):
content, um, but not everyniche is actually like fully
like served Um.
So you identified long form,high quality, narrative
storytelling where they werereally.
I mean, who else?
I mean there was obviouslyImlet Media, which what are some
of their big shows?
Mangesh Hattikudur (Kal (41:45):
Startup was one.
Oz Woloshyn (Kaleidoscope (41:47):
Reply All.
Mangesh Hattikudur (41:49):
Heavyweight .
Oz Woloshyn (Kaleidoscope (41:51):
Mogul .
Alex Kotran (aiEDU) (41:53):
There was, I guess, npr, and there wasn't
much else, right.
Mangesh Hattikudur (Kaleidosc (42:00):
I think to me, what is is
partially so, so meaningfulabout all these things, right,
and and you saw it with IraGlass in this American Life was
that what made it possible forsomeone like me to podcast was
he was imperfect, he allowed hisstutters to go on air, he
(42:22):
stumbled and sounded like aperson you were actually having
a conversation with, and thatlent the show a certain
authenticity and a little bit ofroughness around the edge, a
little bit of like I know you'vetalked about this like a little
bit of wobble, right, and Ithink that when we look at
(42:45):
YouTube, when people are intheir apartments and podcasting,
when you have TikTokers intheir cars and showing you that
they're passionately talking toyou about something, but you
know that it's not as polishedit's coming, you know, sometimes
it's performative, but oftenthere's a bit of heart to it,
right?
(43:06):
And I think to me that's whatyou never get from the AI
version of your voice.
Is that sincerity, thatauthenticity, and those are
words that are overused andabused often, but I also think
when you're moved by something,it's because of that.
But I also think when you'removed by, something.
Alex Kotran (aiEDU) (43:22):
It's because of that, yeah, I think
this idea of authenticity seemslike maybe one of the biggest
undercurrents that is going toshape the way that we consume
content, because I wouldactually push back.
I think, yes, you can sort ofhear, you can tell when there's
an AI voice.
Now, I'm just going to assumethat in, let's say, five years
(43:43):
if we're going to go really farout, maybe sooner uh, it will
probably be imperceptible, um,and yet, even if it were
imperceptible, I still you can,at a certain point, ai is just,
it's very same in art.
I see this with art all the time.
It's like the reason you caneasily spot ai art is because it
actually can only do a fewdifferent sort of flavors.
(44:05):
You have like the 3d renderingversion.
You have sort of like, you know, imitating artists, um, really
bad at sketches, you know it'slike there's a lot of, and so I
I think in any kind of form ofmedia, people are eventually
going to saturate with sort oflike whatever that sort of
average output that the ai cando.
Um, and we're already seeingyour point about this.
(44:27):
Authenticity is interestingbecause even before ai sort of
like complicating it, yeah, Imean, I remember talking to um
some companies who had beeninvesting in stuff like this, uh
, in like sort of like videoseries, and they put a ton of
money on like tiktok ads andlike super like brought in like
some like these, like reallyfamous directors and you know
cinematographers, and theydidn't perform well, because
(44:50):
what they realize is that peoplelike didn't want to see
polished content.
They actually were more drawn tosomething that felt sort of
human yeah um, the other, theother, that we're the other big
idea for this is in a lot ofspaces and so we're sort of
straddling, like we're not sortof trying to reach kids with
this show.
(45:10):
Parents yes, teachersdefinitely, but also
administrators, people that workin the philanthropic sector,
that are trying, you know,decision makers within the K-12
education ecosystem.
I've tried, you know, decisionmakers within the K-12 education
ecosystem very niche veryspecific.
There certainly aren't very many, you know, like shows or
YouTube channels or like rabbitholes that you can deep dive,
and I know this because I havepursued as many of them as I can
(45:31):
possibly find.
And it's you know it's like oldlectures that you know different
universities put up.
Sometimes the audio is likecutting out half the time.
So part of this is the idea iscreate like a space for for
people to nerd out and go deepin like the specific sector and
sort of like angle and dimensionof the topic that we're
pursuing.
(45:51):
But it's also to try to breakout of, I feel like, where a lot
of the discussion that happensin like any any kind of space.
It's happening.
It's happening often atconferences, um, on panels, yeah
, and I I really believe thatthe age of like the panel is
just it's so, it's so dead.
I mean, I feel like when yousit down and just hear people
(46:12):
talk on panels and you get like10 or 15 minutes at most from
any one person, uh, and thenyou're off to the next thing and
it's all feels very scripted,um, my, my instinct is that this
format is actually a muchbetter way to like do knowledge
transfer from really interesting, smart people like give them,
like a lot of time to actuallygo deep, like, have the
(46:34):
conversation, be a little bitmore unstructured, right, um,
and allow yourself to be sort ofsurprised and and and not sort
of just try to, um, I'll justfinish it.
Yeah, allow yourself to besurprised, um, what?
So you mentioned that you'reactually working on some uh like
, is it, is it talk shows orlike can you tell me more about
the format and anything thatyou've any advice that you can
(46:58):
impart on us as we're trying tothink about, sort of like, how
to like what do with, with thisidea that we have?
Oz Woloshyn (Kaleidoscope) (47:03):
I just I just taken over a podcast
called tech stuff, which is, um, my heart's uh flagship
technology podcast, been aroundfor 16 years, I think, driven
tens, tens, if not 100 millionlifetime downloads and has a
large and loyal audience.
And so, you know, the questionhere for me and my co-host, kara
(47:24):
Price, was how do I, or how dowe serve this audience, make
sure they feel like they haven'tjust had new hosts foisted on
them, but they they feel heardand they want to stick with us,
because, obviously you know,retaining the audience is a
really important goal for us.
And so we just started a gmailaccount techstuffpodcastgmailcom
(47:44):
and constantly solicitingfeedback and we're getting some,
you know, interesting,supportive, sometimes pointed
feedback from the audience.
I find there's nothing morevaluable than that and, to your
point, like the k12 uh worldthat you're talking about is
relatively um niche in a sensethat, like you can probably you
know, there's probably fewerthan 5,000 people, or maybe
(48:05):
20,000 people, I don't knowexactly how many people, but you
could proactively pilot content, put it in front of those
audiences, get their feedbackand allow them to collaborate
with you and build with you tomake this asset as valuable as
possible for them.
What's an?
Alex Kotran (aiEDU) (48:18):
example of some critical feedback you've
gotten as possible for them,what's?
Oz Woloshyn (Kaleidoscope) (48:22):
an example of some critical
feedback.
You've gotten Good question, Ithink one of the.
I'm a former TV news producerand the history of tech stuff.
The previous iteration of techstuff was very much explainer
content that was not tied to thenews cycle and I've leaned a
little bit more into likeexplainer plus news and some
people are like I'm saturatedwith news, little bit more into
like explainer plus news, andsome people are like I'm
(48:42):
saturated with news.
I just want pure explainer.
For me it's hard to cover techin 2025 without some news.
Um, but you know that's apreference and I think there is
there is an audience who reallywant old school, you know,
explainer content, you know, butthat's not my special skill or
my passion.
So it's like well, how do Iserve that to a certain extent,
(49:04):
but match it with what drives meas a creative personality.
Mangesh Hattikudur (Kaleido (49:09):
The one, I'd say sorry, I'll say
two different things.
One is about audiences, and Ithink it's so important to both
listen to your audience andreward them, uh, and and try to
meet them where you can helpthem.
So this is from my previousexperience running a magazine
called mental floss.
We were, you know, I startedout of college had no money,
(49:34):
that, like my business partnerand I were working part-time
jobs.
We were doing this in theevenings.
We were like doing everythingwe could to get this magazine
out.
And this woman wrote us andsaid I like this, but when are
you coming out with a largeprint edition?
And my friend Will and I weresaying a large print edition, we
(49:55):
can barely get this edition out.
And then he sent her.
He was like look, we're notgoing to have a large print
edition, but we can send youword documents.
You can print them out atwhatever size you want.
And that christmas she sent alarge font size large font size
and and uh, and that christmasshe gave like 30 or 40
(50:18):
subscriptions of the magazine.
You, I'm sure with every one ofthose subscriptions she told
people.
The editors of this magazinereached out to me, wrote to me,
responded.
I think that where you cancreate connections with your
audience, where you can helpthem, that word of mouth and
evangelism is really reallymeaningful for a business, and
(50:42):
that's obviously not possiblefor everything.
The other thing I'd say you knowwe were talking previously
about how important it is tothink critically and to be
creative as we're going intothis world of AI.
I also think it's reallyimportant to learn how to edit
and to pare down and to curate,because I think that, even
though these three-hour podcastsare great, I think there are
(51:05):
lots of tangents Time is acommodity for people and the
fact that I might not have threehours but I want to know the
most important parts.
The Coen brothers when they dida director's cut of their movie
Blood Simple it's the onlydirector's cut that's eight
(51:26):
minutes shorter and it's abetter movie, you know, and so,
like I think there's value andyou know you can provide the
shorter version and extendedversion, but I think that,
thinking about what yourlistener really wants always and
how you can deliver that whereit matches your vision of a
product is is really importantall right.
Alex Kotran (aiEDU) (51:45):
Well, I guess we have to listen to our
listeners.
Um oz and gash, such a pleasurehaving you both on.
Thank you for peeling back thecurtain a little bit.
I have so many more questions Ihopefully we can cover it in
dinner.
Yes, welcome to AIEDU Studios, where
we're building either.
We don't know if it's a show,it could be a podcast, maybe
it's a YouTube channel.
The goal is bringinginteresting people on to talk
about the intersections of ournerdiness, talking, obviously,
about artificial intelligenceand education, but also bringing
people who are at the forefrontof different industries,
(00:26):
different spaces, places thatare being impacted by and also
driving forward innovation,artificial intelligence and sort
of help us sort of paint thispicture of what the future
really looks like, because wehear a lot about the future of
work and the age of artificialintelligence.
I think most people don'treally understand what that
means for them, depending ontheir interests, and so, to that
(00:47):
end, we have not actually hadan opportunity to invite folks
in the media space, in thecreative space, and so joining
me today are the co-founders ofKaleidoscope, mangesh Hartikadur
and Az Velashin, and we'regoing to hear more about what
they've built, what Kaleidoscopeis.
They have blazed the trail thatAIDU is just beginning, and so
(01:09):
I'm personally excited.
Oz Woloshyn (Kaleidoscope (01:13):
Maybe we'll have a little bit of a
mix.
I don't know about that.
Alex Kotran (aiEDU) (01:16):
We will certainly be cutting all the
instances where they'reproviding me feedback and
telling me all the things I'mdoing wrong.
But yeah, thank you so much forcoming on to the show.
And why don't we start withjust tell us a little bit about
you know, tell us aboutKaleidoscope and tell us about
what your life has been likeover the last couple of years,
as you've been sort of buildingthis new studio, this new
(01:38):
company.
Mangesh Hattikudur (Kaleid (01:39):
Yeah , so Kaleidoscope is a podcast
company.
We love the audio space.
We think there's something sobeautiful and transportive and
the fact that you can be insomeone's ear and that
connection that you have is someaningful.
This company we really thoughtthat it was just an incredible
(02:07):
way to tell stories from aroundthe world and almost treat it
like an engine for empathy, andthat stories were a way to just
an incredible lens forunderstanding the world.
Alex Kotran (aiEDU) (02:18):
Yeah, and so it's interesting because the
even the project of what we'redoing here, you know, along
maybe 60, maybe even 90 minutes,if we have time, um, to really
get into it it's.
It's really pushing backagainst the trend culturally
that we've seen content go interms of, you know, we went from
you know, short videos to nowliterally like three seconds of
(02:42):
just dopamine you know it was nocontext.
Context um brain rot, as thekids in the classrooms that we
teach I mean it's literally likethey're.
They're sort of likeself-indulgent in the way that
they talk about brain rot.
Well, it was the oxford, uhdictionary word of the year, I
think yeah, it was um, butpodcasts and the type of and
especially the type of podcastsyou do, which is not just sort
(03:04):
of I mean, you do some talkshows right, but you also do
long form storytelling that'sright that builds like series um
.
Folks have probably probablyremember the uh serial and sort
of like how that brought sort ofthis like sort of storytelling
format um into this space andyou've continued that work um.
Can you tell us about some ofthe stories that you've told in
(03:26):
that format?
Mangesh Hattikudur (Kaleid (03:28):
Yeah , definitely, like I said, part
of what we try to do is reallyvalue the listener's time and
really engage them and entertainthem.
There's a show about cacao andwild cacao that we did called
Wild Chocolate, but the way wepitched it was, and it's
(03:48):
ultimately an environmentalstory, but we pitched it as an
Indiana Jones of chocolate.
Or we did a story that was abouta Soviet cosmonaut who got
stuck in space because theSoviet Union disbanded beneath
him and he was stuck there for300 plus days and his only
connection to the world was thisham radio operator in Australia
(04:11):
that he managed to connect with.
Or I did a show that ended upbeing much more personal than I
was expecting, but it was abouthow astrology is seen so
differently in the East than theWest and and and the ways that
you know it's.
It's such a pseudoscience hereand and it's it's a way of life
(04:34):
in India and and and these,these stories that felt really
fun and and emotional and likebig adventures and in some ways
we were trying to trying to goagainst the grain of of, at
least in the beginning of of all.
The really long chat shows thatthat they were going on for
(04:56):
like three or four hours Likethis.
These were things that weremeant to pull you in and
entertain you and, at the end ofthe day, leave you feeling
smarter.
Alex Kotran (aiEDU) (05:07):
Yeah, we were just talking before, um
picking things off about some ofour favorite sort of youtube
shows, and there's this model,just like, bring people together
, put them on some couches ikeacouches maybe um, just banter,
and you know, I think like I'llgive some props.
It's still longer form and'sopportunities to at least, you
know, dive into a little bitmore depth of content.
(05:28):
It's not necessarily alwayseducational content, but there's
a lot more work that goes intotelling a story about cacao,
which could be interesting, butyou know, I could also see this
be I mean, like you can imaginea lecture and something that's
informational informational andpeople obviously have short
attention spans.
What are some of the ways thatyou break past?
(05:49):
Because, also, your shows areamong the most listened to shows
on Apple Podcasts and you'vewon a few awards.
Mangesh Hattikudur (Kaleid (05:56):
Yeah , and, to be clear, I actually
really value really longinterview shows and I think
different shows have differentpurposes.
Long sort of interview shows andI think different shows have
different purposes, right, andso, in the same way that sports
radio people would turn it onand keep it on in the background
, as you're like, maybe paintinga room or you know something to
keep you company, as you'redoing that the commute to work,
(06:18):
where you might have 25 minutesor 30 minutes in a car and you
want to be told something veryclearly, or you want to learn
specifically about markets orwhatever.
Right, like, every podcast hasits own flavor but also purpose,
and so I think we went in veryintentionally with the idea of
telling these really gorgeousstories and we tried to invest
(06:42):
in the sound design in a waythat other people hadn't.
We tried to look at the spacein a way that especially like,
if you think about when we'dstarted this company, it was
coming out of the pandemic andso it was people had been stuck
in their houses for a very long.
In the lockdown period peoplehad been stuck in their houses.
They'd still had this liketremendous wanderlust and to be
(07:03):
able to place someone in adifferent environment through
their earphones was somethingthat we felt was really
meaningful and how many shows doyou have out right now?
how many shows have we done?
Oz Woloshyn (Kaleidoscope (07:15):
We've done more than 10 and we have
another 10 to 12 coming out thisyear.
I think about half of theportfolio of what we do is going
to be weekly more chat stylecontent and half is going to be
weekly more chat style contentand half is going to continue to
be these highly producedcinematic stories.
Mangesh Hattikudur (Kaleidosc (07:30):
I mean, in my mind, what we're
really trying to do is alwayscreate someone's favorite show,
and when I was at iHeart we dida show called Call your
Grandmother and it was an80-year-old host and a
90-year-old host and they werelike a hoot and you wanted to
spend time with them and at theend of this half an hour with
them you're reminded how muchyou love your own grandmother
(07:55):
and for someone like me whosegrandparents have all passed, it
was just such a pleasure and tobe able to create that space.
There's so many flavors of showto provide that, I think, and
there's so many lanes.
I don't really think of otherpodcasts as competition right.
I just think there's space toreally sort of provide something
(08:18):
beautiful and let audiencesfind it and evangelize it.
I really do think good contentmakes its own luck find it and
evangelize it.
Alex Kotran (aiEDU) (08:24):
I really do think good content makes its
own luck.
I want to talk about thisformat and we're going to go a
little meta and maybe you caneven imbue some sage wisdom on
me and my team as we're tryingto figure out how to build our
own.
Again, I really don't know ifthis is a show.
I think, like right now I'mthinking about it more because
and you will be the first totell me right that, like when
you pitch an idea for a show,there's a really coherent sort
(08:46):
of like framing.
You have a sense of the formatand we're still experimenting.
And so I'm trying to give myselfa little bit of, you know,
wiggle room to really kind ofsee what works and see what
works for our guests.
Oz Woloshyn (Kaleidoscope) (08:56):
Well , that's fun.
I mean you find that you findthe format in the doing right,
so you don't have a firm launchdate.
It's quite nice because you caniterate and iterate until you
find exactly what works for youand then take that out as the
product, or you can launch itand iterate it in progress,
especially these morechat-driven formats.
You see how much they evolvefrom episode 1 to episode 10 to
episode 100.
Alex Kotran (aiEDU) (09:17):
I've only listened to three of the shows
that you've released.
The first three, wild Chocolate, skyline Drive and the Last
Soviet and yeah, those wereincredible.
Also, the Last Soviet,amazingly narrated by Lance Bass
, who I never realized heactually served in the.
It was like a Russian sort ofcosmonaut summer camp.
Mangesh Hattikudur (Kaleidos (09:38):
He was prepping to go to space and
he spent time at Star City,learned Russian, did all this
testing.
It was pretty incredible what Ithink.
Alex Kotran (aiEDU) (09:47):
What stood out to me is I never had an
interest in chocolate.
In fact, I don't really evenlike chocolate, although I did,
I do have the, the, the, thecollab that y'all did.
um, that chocolate was amazing.
Um, I actually developed anappreciation for chocolate, but
basically it was not the type ofshow I would normally have
committed to.
But there was something aboutbeing transported to, like
(10:09):
literally being in the Amazon,the interviews with people in
these remote villages, and itbrought me back to this.
Yeah, it's sort of like thisage of like much more
adventurous, like the value ofmedia was sort of like bringing
you to places that you couldn'tnecessarily go yourself.
Yeah, um, and I'm someone who I,I sort of I respect, you know,
(10:30):
the marvel universe and Irespect sort of you know, these
like big blockbuster movies, um,but it does feel like that we
have a lot less of thatculturally now.
You know, you're sort of oryou're transported to 50 or 100
different parts of the world inthe span of like two minutes on
Instagram or TikTok and youdon't, you don't really connect.
There's no, there's noopportunity to actually connect
(10:51):
and sort of hear the almost likethe tenor of this old woman's
voice and sort of like theroughness, and you can kind of
tell that she's been likeworking in, you know, in this,
in this trade, for 30 years andlike the fact that you can hear
that in her voice is sort of asubtlety to it, and I'm curious
if you feel like this is a?
I mean, do you feel like thismedium is underappreciated?
(11:13):
I mean, everything now feelslike niches.
Everybody has their, you know,there's folks spending 20 hours
a week watching people playvideo games, not playing video
games, watching people playvideo games, right, and so, and
actually my favorite, so myfavorite niche is it's a YouTube
channel called Townsend andsons, and this is this guy, john
Townsend, in Indiana, and he,uh, is a reenactor, but his,
(11:34):
specifically, he reenacts um18th century cooking and so he
has the whole gar built like ahut and he had all the materials
and he like teaches, he likelooks into old recipe books and
like cooks 17 or 18th centurycuisine, um, and he's millions,
like 2.5 million followers.
His videos get hundreds ofthousands, sometimes millions of
views, um, and I love that.
(11:56):
I love that.
That's actually like part ofwhat the internet has done is
you don't have to have like thebest show that literally every
single person is talking about.
You have to have like the bestshow that literally every single
person is talking about.
You have to figure out how tohook you know just that right
type person, um, and and so youhave three shows that are kind
of all over.
Mangesh Hattikudur (Kaleidosc (12:15):
I mean, like you know, chocolate
space astrology, um, but I Ithink it's one of those things
that people don't know whatthey're interested in until
they're shown right.
And so one a a show aboutchocolate.
Probably you wouldn't expectthat.
It starts in the Amazon with ajournalist circling because he
can't land a plane and itfinally lands and the whole, and
(12:38):
they finally find a patch ofland to land on and the plane
comes to a stop and then fourmen with guns come out and tell
them, you know, like, this is adrug lord's land and you
shouldn't be landing here, right?
I mean, like, that isn't theexpectation of what a chocolate
show should be about.
But also it is engaging, right,and I think that I certainly,
(13:03):
like you know, netflix has showsabout glass blowing that, like
I wouldn't have.
I don't care about glass blowing, or I don't think I do, but
then I watch.
You know, like I don't reallycare that much about baking.
I watch great british bake offall the time.
You know like, and suddenly Ido care about it.
You know like, there there's somany things, so many things to
appreciate and so many ways toappreciate it, and I feel like
(13:24):
you just have to be led in inthe right way.
Oz Woloshyn (Kaleidoscope) (13:27):
I think not for nothing.
I mean, the connective threadbetween those three shows is
their quest adventures thatrelate in one way or another to
science and, in some cases,technology.
So with Wild Chocolate, youhave these people who are so
obsessed with this pre-columbian, semi-mythical cacao in the
(13:48):
amazon that can only grow if ithas enough uh canopy coverage
that it's in the dark for mostof the day.
And these people risk it alland bet their whole life savings
on trying to find these beansright.
And so you go on this adventurewith them.
In the process, you learn aboutthe amazon, you learn about the
history of cacao cultivation,you hear about his role in
pre-Columbian culture In thecase of the Last Soviet.
(14:09):
You have a guy who goes to space, is in the Mir space station.
The Soviet Union collapsesbeneath him and he's faced with
a choice Do I come home andabandon the Mir space station to
spin off into space, or do Iwait here for as long as it
takes until someone can bring mehome?
Because this mere space stationis the last redoubt and the
final pride of the of thecountry and the empire I've
(14:32):
given my life to.
And again, in that, you get tothe history of the space race,
you get to rocket technology,propulsion technology in the
case of mangashe's show, uh, theastrology show, skyline drive.
This was, in a sense, askeptic's guide to, you know, a
spiritual topic.
Alex Kotran (aiEDU) (14:50):
Because you , your family obviously believed
, believes in astrology.
You were yourself a skepticgoing into the show.
Mangesh Hattikudur (Kaleid (14:58):
Yeah , I mean, I think it's one of
these things that I don'tbelieve in astrology.
But astrology has happened tome, right, like my parents met
because they were introduced toone another because their charts
matched, you know, and it wasalmost like this dating service
where, like they could, onlytheir family only introduced
them to people who they matchedup with and they dated for a bit
(15:18):
, liked each other, you know,ended up getting married.
But I'm only here because ofastrology in this strange way
and so, and you also find,astrology, whether you believe
it or not, has affected theworld drastically in that one
search for understandingastrology led to an
understanding of astronomy.
Right, it was only through,like it's the desire to
(15:41):
understand the future that ledus to, like, investigate the
stars closer.
But also, astrology is whatconvinced Reagan for the nuclear
disarmament, to have talks withGorbachev.
(16:02):
It was his astrologer, who wasin his ear, who said now is a
great time to do this, this isthe time, this is the reason.
And he was such a believer thatif you look at the times of his
inaugurations or the times he'ssworn in, they're at
astrologically significant timesand moments.
And similarly, boris Yeltsin,mitterrand, like all these
(16:25):
people you know, it's one ofthese things that when people
need hope or they're in grief orthey need like consoling, right
, like people turn to the stars,in the same way they turn to
religion.
And so when people don't knowwhat to do, it's convenient.
(16:47):
And so there are teams in SanFrancisco baseball teams that
don't know which players to pickin drafts, and so they use
astrologers to guess which onewill have better longevity with
the team.
And we don't see this.
It's invisible.
But when we're rooting for ourfavorite teams, we don't know
(17:08):
that actually astrology hasplayed a part.
Alex Kotran (aiEDU) (17:12):
Okay, so you talked about making obscure
topics engaging.
We work with teachers every daywho this is their daily
struggle, whether they'reteaching linear algebra or the
fall of the Soviet Union, toactually come up with an example
that maybe is relevant, or theScarlet letter.
(17:32):
You know, these are some thingsthat students occasionally will
sort of develop an interest in,but for the most part kids are
looking out the window waitingfor the bell to ring and I'm
curious, like is there anythingthat teachers can learn from the
work that you do tellingstories?
Mangesh Hattikudur (Kaleidosc (17:53):
I think one of the things I
learned along the way is thatpeople love feeling smart, and I
think that part of what we dois try to infuse our
storytelling with a lot of joyand a lot of quirkiness, and the
weird facts on the edges oftenare the things that end up
(18:16):
pulling you into stories.
And so something like the LastSoviet was, I believe it was
like a one-off fact about aSoviet that had been stuck in
space that your friend told youabout, and then we were like
wait, that can't be true.
And then we looked into it andthen we dove into it and then we
realized, oh, this isincredibly rich story, and I
(18:38):
feel like, if you can figure outwhat that first fact is, that
gets someone curious, that's funto repeat, right.
Like whether it's the fact that, like I think, belgium tried to
use cats to deliver the mail touse cats to deliver the mail,
(19:04):
you know, and they equipped 37postal cats with, and the whole
experiment was a disaster.
But or that, like you know,there are pirate ships that ran
out of cannonballs and wonbattles with Edom cheese.
Or or, you know, like therethere's so many ways into
history and math and science andstorytelling and even like
Pythagoras, is more interestingwhen you know he had a cult and
(19:25):
wouldn't eat beans and, like youknow, I, I think there are so
many aspects that let you feel alittle smarter and and if you
can walk out of a room with thatsort of joy, that sort of sense
that you want to evangelize atopic that makes learning fun.
Oz Woloshyn (Kaleidoscope) (19:45):
Mary Poppins spoonful of sugar.
We had a meeting yesterday withthe former director of the
Exploratorium here in SanFrancisco and he had some
interesting insights for us.
One of the most fascinatingthings he told us is the number
one complaint about theExploratorium is it's impossible
, if you're a parent, to getyour kids to leave.
(20:06):
But he had this interestingphrase that's strategy equals
sacrifice.
And so he basically said youhave to decide where you want to
be.
Is it engagement or is itinformation that you want to
lead with?
I think, in a culture of TikTokand like dopamine highs, to be
an effective educator you haveto figure out how to meet people
(20:29):
at the place of engagement andthen, you know, educate them.
Following that, I think comingin with the assumption that
people want to know facts toimprove themselves brings you a
very narrow audience of studentsin this day and age.
And so far be it from us togive advice to teachers.
But I think thinking about notbeing allergic to the word
(20:53):
engagement even though it's alittle bit of a dirty word given
the TikTok of it all is crucialto winning any kind of audience
.
Mangesh Hattikudur (Kaleidosc (21:01):
I mean, I see it with one.
My mom was a preschool teacherand what she was so good at was
finding out what each kid wasinterested in and then tailoring
the subject to them.
And so, uh, she would.
She would make the room feellike an airplane and give
everyone passports and take themto Egypt, and take them to
(21:23):
various places, and like and youknow where you got it from Um
but.
but you know at the same time,like she would get these
preschoolers and teach them likeoh, this is, this is how you're
going to finger paint like PaulClay.
This is how you're going tofinger paint like Paul Clay.
This is how you're going tofinger paint like Van Gogh.
This is how you're going to dolike Monet.
And it was just unbelievablethat these kids would walk out
(21:46):
of these classrooms like feelingso confident and so good, and
that they could, a lot of times,like older siblings, would cry
because, like their younger kids, knew more about the solar
system than they did, but I sawit with my own
kids where this incredibleteacher brought worms into the
(22:06):
classroom as the class pet.
Everyone else has one pet, butwe've got 500 here and they
learned everything.
The places where disciplinesmix and they can talk about you
know the.
The places where disciplinesmix and they can talk about the
environment.
They can talk about biology,they can talk about like uh, uh,
you know food systems and allthese things.
Like I, I think that kids aresponges and and they they do
(22:32):
want to learn.
You just have to show, you know,you just have to engage them a
little bit.
And I realized that sounds easyor something, and obviously a
teacher's work is is is thehardest thing in the world and
and like I mean I think invarious places, like my
journalism teacher in in highschool is the one who who made
(22:55):
the intro to an investor, whoinvested in mental floss and he
invested in kaleidoscope, youknow, and like I still
communicate with my liketeachers from third grade and
sixth grade and people wholisten to Skyline Drive, and I
mean I was really lucky in thatthere was such a focus on
(23:17):
creativity in my school and thecreative applications of sort of
the inquiry-based learning wedid space I found is that when
you talk to a lot of folkswho've been doing this work and
it's, and certainly that work ofteaching is thankless and
(23:39):
underpaid and yes, you getsummers off, but it's even the
summers are not as long aspeople assume yeah, Um, and a
lot of them have these sort oflike uh, origin stories that
start with that one teacher, um,that had a transformational
impact in their life.
Alex Kotran (aiEDU) (24:00):
And yes, it is easier said than done, and
preschool teacher, elementaryschool teacher, fourth, fifth
grade there's a lot of scope tobe creative and sort of like
bringing these differentexperiences in as you go further
along and even when you hitfifth grade.
In some cases, teachers arestarting to be burdened with
standards and all the boringcontent stuff that they have to
(24:21):
teach, and so that is often thechallenge.
It's not necessarily thatteachers are not creative or
just don't realize it.
Mangesh Hattikudur (Kaleidosc (24:30):
I think it's much easier when the
kids are younger, I meanespecially as you're teaching to
tests and et cetera, butthere's still so much space.
Alex Kotran (aiEDU) (24:39):
Yeah.
So let me, let me, um, um, so II say all of that because, as
you were sort of describing someof these ideas to me, I was
trying to think about, like howcould a teacher, kind of like
tomorrow, take that and make anactionable and it might be hard
for a teacher to figure out thatweird, quirky connection to
whatever it is, the lesson thatthey're teaching.
(24:59):
We don't normally push AI as atool that teachers should be
using every single day acrossthe.
We don't believe AI should bereplacing the work of teachers,
but it occurs to me that this isactually a really interesting
use case.
Oz Woloshyn (Kaleidoscope) (25:15):
Yeah , what a great you could say
could say you know, I havestudent a who's this, these are
their passions, and student b,these are their passions.
And students see these, theirpassions.
What, what are some suggestionson the best way to engage them
on the topic at?
Alex Kotran (aiEDU) (25:28):
hand.
Oz Woloshyn (Kaleidoscope) (25:29):
I think that would be a perfect
use case.
Alex Kotran (aiEDU) (25:30):
I think it'd be a perfect use case.
Um, and, and the call out isstill that you need a teacher to
.
You know, some percent of thetime, probably a big percent of
the time, in that specificprompt it's going to hallucinate
, make it up.
So, again, not replacing thework of teaching.
You're still gonna have to do alittle bit of homework, but I
think that's the sort of thingthat could make this a lot
easier.
(25:50):
And so when we think about howdoes ai fit into the classroom,
you know well it can makelearning more engaging and more
personalized.
It just occurred to me thatthis is maybe a really like a
tangible example of that it'ssubtle but it actually to your
point.
like the point is the hook.
You know you spend a lot oftime, I assume like significant
amount of time thinking aboutlike that first episode.
Oz Woloshyn (Kaleidoscope) (26:11):
It's always the hardest one so much
time.
Mangesh Hattikudur (Kaleido (26:13):
And but I, yeah, exactly, I don't
want to be dismissive of thework that teachers do, because I
, I saw it firsthand, right Likeand and also my mom used to
make us uh, she pulled my dadand me into like turn
refrigerator boxes into likevarious things, and she was like
, well, this one's a castle.
And I was like I, amma, we justmade a.
(26:34):
And she's like no, no, that wasa rocket ship.
Oz Woloshyn (Kaleidoscope) (26:36):
This is a castle.
Mangesh Hattikudur (Kaleid (26:37):
This is something else, but I was
constantly helping her cutthings and make things and work
sheets and whatever, andobviously that work is so
important.
But I think the benefits to thekids and the way kids walked
out of that classroom soconfident, like I think that's
(26:58):
what I was fascinated with, waslike these kids that really had
a stronger sense of self becausethey knew things and then could
go into kindergarten, where thesocialization is really hard
right, but if they've got theacademics covered then, like,
they can focus on thesocialization is really hard
right, but if they've got theacademics covered then they can
focus on the socializing.
Oz Woloshyn (Kaleidoscope) (27:20):
One of the ways I sometimes like to
prepare for interviews is toopen an LLM and say imagine you
are a Goldman Sachs analyst andyou're going to interview this
person.
What are your top threequestions?
Now imagine you're the editorof Wired magazine.
And now imagine you're KatieCouric magazine.
You're going, and now imagineyou're, you know, katie couric,
and so basically to, to, to tohave the llm like, imagine its
(27:41):
way into different interrogationor questioning approaches and
then kind of understand, usethat to piece together okay,
what is the most interesting wayto engage this person?
Mangesh Hattikudur (Kaleidosc (27:50):
I think I think that's what's
what's really fascinating is,like the, the way you get so
many options out of this.
Oz Woloshyn (Kaleidoscope) (27:56):
I won't say what should I ask this
person?
Mangesh Hattikudur (Kaleidosc (27:57):
I will say provide inspiration on
an approach and I'll choosebetween suggested approaches I
mean I we were talking aboutthis earlier that like I have a
friend who started this companycalled pseudo right, and it's to
help writers, uh, particularlywhen there's like writer's block
, to give you like options ofwhere your story could go and
that inspiration, or taking whatyou've written.
(28:18):
And I had this friend, danielalicon, and and he uh said that
he took a map and he like redrewit and tried to imagine a
landscape and like had to do allthis work to figure out, like
what this imaginary land wouldbe.
But ai can do that for you,right, it can create what your
fantasy visuals are to inspireyou and to give you even more
(28:41):
clarity on what you want to do.
And so I think there are allsorts of ways of creating
options for yourself when you'reunsure.
Alex Kotran (aiEDU) (28:50):
This is a beautiful segue, because I
wanted to have an opportunity toask you about this question
that you've already sort of ledus to, which is what are the
ways that AI either is helpingyou do your work or could help
you do your work?
And before we dive into it, Ijust want to sort of set the
stage for the type of folks thatmight be listening, in short,
(29:13):
teachers, maybe someadministrators, but also parents
, and I talked to a lot ofparents who are like my kids.
In fact, our director ofcommunications is his son loves
creating art.
Yeah, um, not like youtubevideo, like like literal, like
draw, like collages and painting, spray paint.
Um, we get questions from a lotof parents that are like my
(29:36):
kids, are really interested inart, media, design, and they're
reading the news they're seeing,maybe they're playing with Dali
or Midjourney, and you type inone sentence, in not even a
minute but like five seconds,and you get this very compelling
painting, photograph, 3drendering, and I think I've
(30:00):
actually seen some audio bookson Amazon now that are narrated
by AI Voices.
Oz Woloshyn (Kaleidoscope) (30:05):
AI Podcasts.
Alex Kotran (aiEDU) (30:08):
There's AI Podcasts, and so, to the
discerning parent, the questionthat we get is is there going to
be space for for creatives ifai is going to be able to do all
this stuff that wetraditionally associate with
creative work?
Um, is this kind of a dead endfor for my kids, and should they
(30:28):
be learning something else?
Maybe they should be learningai stuff instead of instead of
doing art.
And before I weigh in with howI respond, I'll tease it and say
that, um, I I vehemently, uh,uh, not aligned with that
perspective I think that thereis this important room for
creatives, but you're on thefront front lines of this um and
(30:50):
we're far enough along now.
Chat gbt came out at the end of22, so it's been what like two
and a half two two years andsome change.
Um, okay, is it?
Has it actually changed the wayyou work meaningfully?
Has it changed the way you hire?
Is it changing the way you'rethinking about your strategy?
Um, why don't we start withlike have.
(31:10):
Are you using ai day to day?
Did you use ai in the last week?
If so, how?
Oz Woloshyn (Kaleid (31:15):
Practically speaking, we use AI all day,
every day, in fact.
It's completely integral to ourproduction process, not in the
form of telling us what to say,but in the form of actually
efficiently editing.
And, magus, you want to explainabout Descript and how we use
it?
Mangesh Hattikudur (Kaleid (31:32):
Yeah , we use a program called
Descript and we import our filesinto it and it basically
transcribes all your content.
It can filter out, sounds forlike if there's mic bleed, it
fixes that and it allows you toedit like you're editing a text
document, so you just delete asentence and it cuts it from the
(31:55):
audio.
Oz Woloshyn (Kaleidoscope) (31:55):
You don't have to.
You know you don't have to haveany audio.
And now to have a perfectproduct, of course it's not good
enough, but to be able to gofrom a lengthy interview to an
effective rough cut, you can dothat as easily as you can edit a
Google doc, whereas even threeyears ago that would have
required.
Mangesh Hattikudur (Kaleid (32:13):
What was?
Alex Kotran (aiEDU) (32:14):
it like before you had the script.
Mangesh Hattikudur (Kale (32:15):
Before that, that would have required
specialists.
What was it like before you hadDescript?
Before that, you would have tosend the files to a transcriber.
Oz Woloshyn (Kaleidoscope) (32:21):
You would send, which cost like a
dollar a minute.
I mean, it was expensive.
Mangesh Hattikudur (Kaleid (32:23):
Yeah , it was too expensive to get
stuff.
Oz Woloshyn (Kaleidoscope) (32:26):
When I started podcasting in 2019,
it was too expensive to get allour audio transcribed.
It was too expensive to get allour audio transcribed.
Can you believe that?
So the production flow we wouldliterally decide?
Is this interview importantenough to spend $300 to get
transcribed, to figure out whatthe best moment is without
listening to it for three hours?
Mangesh Hattikudur (Kaleido (32:46):
And also it often wouldn't be you
listening to it, it would besomeone else listening to it and
you'd hope they'd make thechanges.
And you try to consolidatethose two things and try to meet
somewhere what's closer to yourvision versus.
You know, I do a show calledpart-time genius.
It's a very quick show but butI look at it, I make the tweaks
(33:11):
that I heard as I was recordingsomething.
I knew this sentence was wrong.
I knew I made a hiccup here,all these things and then I send
it to get mixed and masteredand all the other steps that
used to all the like QC, thequality control that used to
take place have been eliminated,like it's a much, much faster
and more seamless process.
Oz Woloshyn (Kaleidoscope) (33:32):
But this is a great example.
This is not like we're notgoing out and saying like or
sometimes we are, but in thiscase not.
How do I use AI?
It's more like it's not.
We don't think about it asusing ai, we think about it as
using an application.
In fact, we've forgotten howbad it was or how difficult it
was before the script.
I think you know one of theother areas, obviously creative
inspiration, and I'm super,super happy and relieved to
(33:54):
report that I think ai does agreat job of summarizing
information.
Obviously, often with halluc ithas no ability to write with
voice, and even like NotebookLMI is pretty interesting.
Like it can make a prettycompetent two-way conversation
podcast between like a malevoice and a female voice.
It's also entirely reliant onwhat you feed it.
So like.
(34:14):
If you just say like here's 10articles about like the
Superbowl, it will do a reallyboring job, whereas if you say
like okay, here's 10 articlesabout like the Superbowl in 1963
.
And I want to understand howthis social phenomenon affected
the you know the sportsenvironment, you might get
something interesting, but atthat point you've done a lot of
the creative labor yourself,right?
(34:35):
So you know this, I think.
Coming back to your questionabout technology and and what
should parents think about fortheir children, I remember so
vividly, like the 2010 era learnto code.
Like forget the humanities,learn to code.
Thank god, that was the worstadvice ever, right, um?
And similarly now, like youknow, I read jack clark, one of
the anthropic foundersnewsletter a couple of days ago
and obviously he has like avested interest, but his, his
(34:56):
main point was, like the winnersof the future will be the most
curious people, the people whokeep finding the most
interesting ways, the mostforward-thinking ways to ask um
questions.
And how do you get to be a goodquestion on asker?
By feeding your curiosity as achild, as a young person.
So I I really think that likethe future is is is not about,
(35:16):
um, you know, getting being like.
Oh well, ai can like make animage, so why would I become an
artist?
It's really.
How do I develop into the typeof person who can think
creatively about how to make themost of technology to serve me
and my community and my society?
Alex Kotran (aiEDU) (35:31):
What I'm hearing is the type of work that
you're describing.
That's being replaced.
I don't think anybody is inschool thinking I want to be a
transcriber.
I don't think anybody's inschool thinking I want to be a
transcriber.
I don't think anybody's inschool thinking I want to spend
12 hours a day in Photoshop, youknow, removing objects from the
background of videos.
Mangesh Hattikudur (Kaleido (35:51):
And it was meaningful work for
people for decades.
But yes, I agree.
I mean, like I don't think thatwas anyone's ambition, right?
But yes, I agree.
Alex Kotran (aiEDU) (36:01):
I mean, like I don't think that was
anyone's ambition, right?
Yeah, so there is a questionabout what, like where do those
dislocated workers go?
There's this sort of you know,the talking point is we're going
to empower them to do moreinteresting creative work.
The reality is, you know, thereare sort of there aren't that
many roles at companies likeKaleidoscope to do creative work
(36:22):
.
But I want to put a pin in thatconversation because I want to
actually hear more about Sorry,can I just say one other thing.
Mangesh Hattikudur (Kaleid (36:31):
Most people don't know this, but our
album art for our first sixshows was all done on an early
version of Dolly.
The art was done on it and thenwe got an artist to to draw
over it and highlight certainelements and stuff, but but the
the starting point was was aiokay.
Alex Kotran (aiEDU) (36:50):
so that is fascinating because when you
first told me that, I wasincredulous because I I sort of
pride myself at being able tospot AI art and my I like Spidey
senses didn't tingle when I waslooking at your album art.
It does for a lot of other stuffon, like Spotify and Apple.
But I understand why now,because you, you also brought in
(37:11):
that human component to take itto that next level, and I think
that's that's to me this likereally interesting takeaway,
which is the there's alwaysgoing to be this need, if
everybody has access to thisbaseline of content, this common
denominator of justAI-generated whatever, there's
(37:33):
going to be so much of that thatI think people are going to
actually seek out and identifyquality and the only way to do
something that ai can't do is totap human potential and human,
uh, creativity, um, has itimpacted hiring?
I mean, like, have you?
I guess one question is haveyou hired anybody just to like
be the ai tool user?
Because I think that's whatsome people are saying is that,
(37:54):
oh, you don't need to learn, youdon't need to don't study art,
just learn how to use ai andprompt engineer, um, and I'm
interested in that because I'mcurious if that's the type of.
Is that a job posting onKaleidoscope's page?
Mangesh Hattikudur (Kaleidosc (38:09):
I don't think so.
I mean, at least from myperspective, we are looking for
people with voice and outsiderideas that are super creative,
and even when I was at I heart,a lot of people would just hire
people from the radio industryto to come in and work because,
like this is versus like wehired, um musicians who'd never
(38:32):
done podcasting, we hired, likeyou know, people who were uh in
all sorts of other fields.
There was someone who used tolike help, um uh kids who were
ill create poetry and art, but Ithink what we're looking for is
really creative, smart peoplewho have a point of view, who
(38:54):
can learn the tools and use thetools to advance our own sort of
point of view or creativity.
Alex Kotran (aiEDU) (39:01):
Yeah, I think we've had the same
experience, like we don't reallyuse AI very much for curriculum
development.
If you have a certain level ofquality, it's actually very hard
to get.
I mean, I've done, I've triedusing ChatGBD to write stuff.
Oz Woloshyn (Kaleidoscope) (39:17):
Same .
Alex Kotran (aiEDU) (39:18):
I hear people saying, oh, it can write
emails, it can write pressreleases Like I disagree.
I hear people saying, oh, itcan write emails, it can write
press releases Like I disagree.
I think that it can.
I think if you're trying towrite, let's say, a hundred blog
posts a month, then, yes,chatgbt is probably useful.
I don't know why you would wantto do that.
Oz Woloshyn (Kaleidoscope) (39:32):
If you're trying to write, well,
maybe if SEO right people don'tlike written content on the web
uh, except for from superprestige publications like I
think, or in forums like redditand stuff is like losing very
rapidly losing its authority,which I think is a big
opportunity for audio because,yes, you can make a deep fake of
(39:53):
my voice and theoretically havechat, gpt, write something,
then have the deep fake read myvoice, but, like you can hear
that pretty easily and I think,the opportunity for podcasting.
The reason we had this podcastelection in part is because it's
a very trusted medium and Ithink will retain a moat in that
respect.
People looking for guidance um,you're way less swayed by the
(40:14):
top 10 google results six monthsfrom now than you were a year
ago we're in a bit of the thewild west right now in terms of
of media.
Alex Kotran (aiEDU) (40:23):
I mean, I think that there's sort of a
definitive shift away fromlinear tv.
We're using a lot of industryspeak.
You said moat, I'm gonna uh,linear tv is sort of just like
broadcast television, cable, newcable channels, um, and you
know people were moving toNetflix.
I honestly YouTube, especiallyYouTube premium, without the ads
(40:43):
, is like totally dominated theway I consume media.
Obviously, people are onInstagram Tik TOK Um, and that's
especially pronounced in thenew generation, right Like Gen Z
, gen Alpha Um, and I'm so.
So maybe we can close with alittle bit of um, sort of meta
(41:04):
introspection about sort of whatwe are doing here at AIEDU
studios, um, and I I'd like youto just sort of I'd like your
take on this in terms of what wecould be doing with the format
Um, because one of the thingsthat I found is there's, there's
now, yes, there's, this shiftto sort of this open universe of
(41:24):
content, um, but not everyniche is actually like fully
like served Um.
So you identified long form,high quality, narrative
storytelling where they werereally.
I mean, who else?
I mean there was obviouslyImlet Media, which what are some
of their big shows?
Mangesh Hattikudur (Kal (41:45):
Startup was one.
Oz Woloshyn (Kaleidoscope (41:47):
Reply All.
Mangesh Hattikudur (41:49):
Heavyweight .
Oz Woloshyn (Kaleidoscope (41:51):
Mogul .
Alex Kotran (aiEDU) (41:53):
There was, I guess, npr, and there wasn't
much else, right.
Mangesh Hattikudur (Kaleidosc (42:00):
I think to me, what is is
partially so, so meaningfulabout all these things, right,
and and you saw it with IraGlass in this American Life was
that what made it possible forsomeone like me to podcast was
he was imperfect, he allowed hisstutters to go on air, he
(42:22):
stumbled and sounded like aperson you were actually having
a conversation with, and thatlent the show a certain
authenticity and a little bit ofroughness around the edge, a
little bit of like I know you'vetalked about this like a little
bit of wobble, right, and Ithink that when we look at
(42:45):
YouTube, when people are intheir apartments and podcasting,
when you have TikTokers intheir cars and showing you that
they're passionately talking toyou about something, but you
know that it's not as polishedit's coming, you know, sometimes
it's performative, but oftenthere's a bit of heart to it,
right?
(43:06):
And I think to me that's whatyou never get from the AI
version of your voice.
Is that sincerity, thatauthenticity, and those are
words that are overused andabused often, but I also think
when you're moved by something,it's because of that.
But I also think when you'removed by, something.
Alex Kotran (aiEDU) (43:22):
It's because of that, yeah, I think
this idea of authenticity seemslike maybe one of the biggest
undercurrents that is going toshape the way that we consume
content, because I wouldactually push back.
I think, yes, you can sort ofhear, you can tell when there's
an AI voice.
Now, I'm just going to assumethat in, let's say, five years
(43:43):
if we're going to go really farout, maybe sooner uh, it will
probably be imperceptible, um,and yet, even if it were
imperceptible, I still you can,at a certain point, ai is just,
it's very same in art.
I see this with art all the time.
It's like the reason you caneasily spot ai art is because it
actually can only do a fewdifferent sort of flavors.
(44:05):
You have like the 3d renderingversion.
You have sort of like, you know, imitating artists, um, really
bad at sketches, you know it'slike there's a lot of, and so I
I think in any kind of form ofmedia, people are eventually
going to saturate with sort oflike whatever that sort of
average output that the ai cando.
Um, and we're already seeingyour point about this.
(44:27):
Authenticity is interestingbecause even before ai sort of
like complicating it, yeah, Imean, I remember talking to um
some companies who had beeninvesting in stuff like this, uh
, in like sort of like videoseries, and they put a ton of
money on like tiktok ads andlike super like brought in like
some like these, like reallyfamous directors and you know
cinematographers, and theydidn't perform well, because
(44:50):
what they realize is that peoplelike didn't want to see
polished content.
They actually were more drawn tosomething that felt sort of
human yeah um, the other, theother, that we're the other big
idea for this is in a lot ofspaces and so we're sort of
straddling, like we're not sortof trying to reach kids with
this show.
(45:10):
Parents yes, teachersdefinitely, but also
administrators, people that workin the philanthropic sector,
that are trying, you know,decision makers within the K-12
education ecosystem.
I've tried, you know, decisionmakers within the K-12 education
ecosystem very niche veryspecific.
There certainly aren't very many, you know, like shows or
YouTube channels or like rabbitholes that you can deep dive,
and I know this because I havepursued as many of them as I can
(45:31):
possibly find.
And it's you know it's like oldlectures that you know different
universities put up.
Sometimes the audio is likecutting out half the time.
So part of this is the idea iscreate like a space for for
people to nerd out and go deepin like the specific sector and
sort of like angle and dimensionof the topic that we're
pursuing.
(45:51):
But it's also to try to breakout of, I feel like, where a lot
of the discussion that happensin like any any kind of space.
It's happening.
It's happening often atconferences, um, on panels, yeah
, and I I really believe thatthe age of like the panel is
just it's so, it's so dead.
I mean, I feel like when yousit down and just hear people
(46:12):
talk on panels and you get like10 or 15 minutes at most from
any one person, uh, and thenyou're off to the next thing and
it's all feels very scripted,um, my, my instinct is that this
format is actually a muchbetter way to like do knowledge
transfer from really interesting, smart people like give them,
like a lot of time to actuallygo deep, like, have the
(46:34):
conversation, be a little bitmore unstructured, right, um,
and allow yourself to be sort ofsurprised and and and not sort
of just try to, um, I'll justfinish it.
Yeah, allow yourself to besurprised, um, what?
So you mentioned that you'reactually working on some uh like
, is it, is it talk shows orlike can you tell me more about
the format and anything thatyou've any advice that you can
(46:58):
impart on us as we're trying tothink about, sort of like, how
to like what do with, with thisidea that we have?
Oz Woloshyn (Kaleidoscope) (47:03):
I just I just taken over a podcast
called tech stuff, which is, um, my heart's uh flagship
technology podcast, been aroundfor 16 years, I think, driven
tens, tens, if not 100 millionlifetime downloads and has a
large and loyal audience.
And so, you know, the questionhere for me and my co-host, kara
(47:24):
Price, was how do I, or how dowe serve this audience, make
sure they feel like they haven'tjust had new hosts foisted on
them, but they they feel heardand they want to stick with us,
because, obviously you know,retaining the audience is a
really important goal for us.
And so we just started a gmailaccount techstuffpodcastgmailcom
(47:44):
and constantly solicitingfeedback and we're getting some,
you know, interesting,supportive, sometimes pointed
feedback from the audience.
I find there's nothing morevaluable than that and, to your
point, like the k12 uh worldthat you're talking about is
relatively um niche in a sensethat, like you can probably you
know, there's probably fewerthan 5,000 people, or maybe
(48:05):
20,000 people, I don't knowexactly how many people, but you
could proactively pilot content, put it in front of those
audiences, get their feedbackand allow them to collaborate
with you and build with you tomake this asset as valuable as
possible for them.
What's an?
Alex Kotran (aiEDU) (48:18):
example of some critical feedback you've
gotten as possible for them,what's?
Oz Woloshyn (Kaleidoscope) (48:22):
an example of some critical
feedback.
You've gotten Good question, Ithink one of the.
I'm a former TV news producerand the history of tech stuff.
The previous iteration of techstuff was very much explainer
content that was not tied to thenews cycle and I've leaned a
little bit more into likeexplainer plus news and some
people are like I'm saturatedwith news, little bit more into
like explainer plus news, andsome people are like I'm
(48:42):
saturated with news.
I just want pure explainer.
For me it's hard to cover techin 2025 without some news.
Um, but you know that's apreference and I think there is
there is an audience who reallywant old school, you know,
explainer content, you know, butthat's not my special skill or
my passion.
So it's like well, how do Iserve that to a certain extent,
(49:04):
but match it with what drives meas a creative personality.
Mangesh Hattikudur (Kaleido (49:09):
The one, I'd say sorry, I'll say
two different things.
One is about audiences, and Ithink it's so important to both
listen to your audience andreward them, uh, and and try to
meet them where you can helpthem.
So this is from my previousexperience running a magazine
called mental floss.
We were, you know, I startedout of college had no money,
(49:34):
that, like my business partnerand I were working part-time
jobs.
We were doing this in theevenings.
We were like doing everythingwe could to get this magazine
out.
And this woman wrote us andsaid I like this, but when are
you coming out with a largeprint edition?
And my friend Will and I weresaying a large print edition, we
(49:55):
can barely get this edition out.
And then he sent her.
He was like look, we're notgoing to have a large print
edition, but we can send youword documents.
You can print them out atwhatever size you want.
And that christmas she sent alarge font size large font size
and and uh, and that christmasshe gave like 30 or 40
(50:18):
subscriptions of the magazine.
You, I'm sure with every one ofthose subscriptions she told
people.
The editors of this magazinereached out to me, wrote to me,
responded.
I think that where you cancreate connections with your
audience, where you can helpthem, that word of mouth and
evangelism is really reallymeaningful for a business, and
(50:42):
that's obviously not possiblefor everything.
The other thing I'd say you knowwe were talking previously
about how important it is tothink critically and to be
creative as we're going intothis world of AI.
I also think it's reallyimportant to learn how to edit
and to pare down and to curate,because I think that, even
though these three-hour podcastsare great, I think there are
(51:05):
lots of tangents Time is acommodity for people and the
fact that I might not have threehours but I want to know the
most important parts.
The Coen brothers when they dida director's cut of their movie
Blood Simple it's the onlydirector's cut that's eight
(51:26):
minutes shorter and it's abetter movie, you know, and so,
like I think there's value andyou know you can provide the
shorter version and extendedversion, but I think that,
thinking about what yourlistener really wants always and
how you can deliver that whereit matches your vision of a
product is is really importantall right.
Alex Kotran (aiEDU) (51:45):
Well, I guess we have to listen to our
listeners.
Um oz and gash, such a pleasurehaving you both on.
Thank you for peeling back thecurtain a little bit.
I have so many more questions Ihopefully we can cover it in
dinner.
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.
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
CrimeLess: Hillbilly Heist
It’s 1996 in rural North Carolina, and an oddball crew makes history when they pull off America’s third largest cash heist. But it’s all downhill from there. Join host Johnny Knoxville as he unspools a wild and woolly tale about a group of regular ‘ol folks who risked it all for a chance at a better life. CrimeLess: Hillbilly Heist answers the question: what would you do with 17.3 million dollars? The answer includes diamond rings, mansions, velvet Elvis paintings, plus a run for the border, murder-for-hire-plots, and FBI busts.