Latest / AI for Founders with Ryan Estes / The New SEO Is Happening Without You

Transcript

The New SEO Is Happening Without You · AI for Founders with Ryan Estes

0:00
Your SEO is not dead, it's just invisible right now.
0:03
Every time a potential customer asks an AI what to buy, where to
0:06
go, or who to trust, that AI is forming an opinion about your
0:10
brand without you in the room, and you have no idea what it's
0:13
saying. It could be wrong about your
0:15
pricing, wrong about your product, wrong about who you
0:17
even are. And the longer you wait, the
0:20
harder the false narrative is to unwind.
0:22
Because once an AI makes up its mind, it hardens fast.
0:25
You're not losing customers to a competitor.
0:27
You're losing them to a hallucination.
0:30
The brands that figure this out in the next 12 to 18 months will
0:33
own the decade. The ones that don't will be
0:35
invisible at the exact moments their customers are ready to
0:38
buy. So if AI answered every question
0:40
your customer had about you right now without them ever
0:43
visiting your website, what story would it tell?
0:46
And is that the story you actually want told?
0:48
Welcome to AI for founders. I'm Ryan Estes.
0:51
My guest today is Justin Inman, founder and CEO of Emeros, a
0:55
former Google exec who left one of the most powerful companies
0:58
in digital advertising to build the operating system for AI
1:02
brand visibility. You will learn how AI is
1:05
silently misrepresenting brands at scale and what the smartest
1:08
founders are doing to take control, and why prompts are the
1:11
new clicks. The founders winning right now
1:14
are not the ones working harder, they're the ones staying
1:17
sharper. The AI for
1:18
foundersnewsletter@aiforfounders.co delivers the insights, tools,
1:23
and conversations that keep you ahead of the curve.
1:25
One e-mail, one edge every week. Go get it at AI for founders.co
1:30
and if this hit different, drop a five star review and help
1:32
another founder find the show. And Speaking of showing up in
1:35
the right places, Kick caster.com is the premier
1:38
podcast booking agency that gets founders onto the right shows,
1:42
in front of the right audiences and at exactly the right time.
1:46
Strategic placements that build authority, drive leads and put
1:49
your name in the conversations that matter.
1:51
Go to kickcaster.com and tell them AI for founder sent you.
1:55
That's kick Caster by MO Burst. All right, Justin, welcome to
2:01
the show. Thank you so much for joining
2:02
me. Excited to be here. 100% now,
2:06
particularly I'm, I'm excited because you have an amazing
2:08
company that's kind of has the product on the tip of
2:11
everybody's tongue and that's how to optimize LLMS to give
2:15
your product recommendations. You know, we know now that
2:18
Google search is just fading to #2 even within Google, you know
2:21
what I mean? It's going to spit out the
2:23
Gemini answers 1st. So really kind of the SEO has
2:27
entered into a new era where we're not even sure what the
2:30
acronym is yet. So is it AEO or is it Geo?
2:33
What is it? There's a lot of a lot.
2:35
There's AIOAEOGEO, there's there's a lot of different
2:39
acronyms. I think the the space is is is
2:41
starting to kind of like get around to AEO and Geo.
2:45
Yeah, a EOGEO there where it's like GIF and Jif.
2:49
People are taking their sides, but it's kind of amazing and
2:53
it's so early. We don't even have a name for it
2:55
yet, but everybody knows exactly what it is.
2:57
That's the top of mind of every marketer, that's for sure. 100%
3:01
So that's a good place to kind of dovetail into your company,
3:04
which, and you have to correct me again, not knowing how to
3:06
pronounce it, it's either Ambrose or Embry OS which one is
3:09
it? Embryos Ambrose love it.
3:13
OK. And it's Ambrose dot AI.
3:16
Now there's there's probably some new products that are
3:19
springing to market. Tell me a little bit about
3:21
Ambrose exactly kind of what the tactics, what the differentiator
3:25
is for for you guys? Yeah.
3:28
So I would say the majority of the space is thinking about AI
3:32
visibility and a very narrow channel, which is basically
3:36
content creation. We have competitors in the space
3:40
that are are really, you know, focused on AI generated articles
3:45
and they push it out to use the website or across a bunch of
3:47
syndicated publishers. I spent about 10 years at Google
3:51
and I know that that's that that hack is it's only a matter of
3:55
time. It's the LM start to pick up on
3:58
what that looks like. And so we built Embryos to be
4:01
the first platform to be looking at AI visibility across your
4:05
entire digital footprint, which is paid, owned and earned,
4:07
right. So what you're doing on social
4:09
or even podcasts and PR etcetera, everything is
4:14
interconnected. And so that's how you should be
4:15
viewing AI visibility. Totally, you know, and look,
4:20
this is probably a good introduction to kind of what you
4:21
were working on with Google because, you know, through
4:24
Google search. I, I mean, I remember it from
4:26
the beginning. I remember when Google Analytics
4:29
took away the keywords that they used to give you just for free.
4:32
It was this, there's a wonderful time, but really the, the the
4:36
early terms were like kind of white hat, Black Hat.
4:38
They started to differentiate on like, Hey, this is cool, you
4:42
know, provide great resources, provide great content, will
4:45
reward you with search to the raising to the top of search.
4:51
But if you're doing stuff like doing links on a white page with
4:54
white fonts, that's Black Hat. So you started to, it started to
4:58
filter itself out kind of early on.
5:01
So maybe maybe a little bit of inspiration of like what brought
5:05
Ambrose out and like kind of what you were doing at Google
5:08
and where you saw an opportunity.
5:10
Yeah. So I spent about 10 years at
5:12
Google. I was always on the the
5:13
enterprise sales team. So I worked with some of the
5:16
largest marketers of the world, everyone from Coca-Cola,
5:19
L'Oreal, Unilever, I and I was selling kind of Google's ad tech
5:24
and Bartec solutions to CMS I and that, that period of time
5:30
was a little rough, right? Like going in and, and to
5:33
Kellogg and saying, hey, let's talk about cloud infrastructure
5:36
to a, a company that is not used to be having those types of
5:40
technology conversations is a little rough.
5:42
But everyone got there right? And everyone got to see the
5:45
value of that. I, I would say now you know that
5:48
that period of time was probably like three to five years.
5:51
Right now it's a the, the period of time to be able to get into
5:54
these LLMS and start to optimize towards your visibility and,
5:57
and, and kind of your brand recommendations, etcetera.
6:01
That's a very truncated window because the, the shifts are
6:03
happening so rapidly within consumers and AI in general,
6:07
right? I think I would say it's
6:09
probably like a 12 to 18 month period that everyone will be
6:12
kind of thinking about this and, and really kind of putting a
6:15
stake in the ground for it. So when I left Google, you know,
6:21
last year, I really kind of saw my own personal LLM and AI
6:25
assistant and chatbot usage explode.
6:29
I, I really kind of actually saw when my mom, who's like the most
6:32
tech and apps person ever was was like quoting Chachi BT quite
6:37
a bit. And I was like, wow, this has to
6:38
be a space that the marketers have to be freaking out about.
6:42
And then I, I did my research, right?
6:44
I, I looked and see, I saw that there was other big players in
6:47
the space. One of them just actually hit
6:49
Series B. My biggest competitor actually
6:52
hit from the moment they started to Series C was 18 months.
6:56
They're valued a billion billion dollars company I and I was
7:00
like, well, I don't really want to jump into a space that is
7:02
already super crowded, right, But I got AI got a a couple
7:06
demos of of the the largest players in the space and the the
7:09
the biggest player. Once I got that demo, I started
7:11
the company the next day because I was like, this is not how you
7:15
should be thinking about it. It's a very narrow viewpoint on
7:19
how AI visibility will be, you know, viewed and unoptimized
7:24
towards in the future. So some people might think that
7:28
was absolutely crazy, but you know, I was felt super
7:30
passionate about like, you know, this is not what this is not
7:33
what marketers want right now or will need in the future.
7:37
Yeah. And the, the, the tool is
7:39
actually totally different. You know, obviously the output
7:41
is a little bit different, but the tool's different because
7:43
especially now the last three months, it's starting to learn
7:47
much better and it's starting to get context much better on who
7:50
you are. So the opportunity to afford to
7:53
really customize and really start serving up the best
7:55
results is great. So let's maybe talk a little bit
7:59
about, you know, just search in general, how it's kind of
8:01
evolved because everyone was used to Google.
8:04
You go to a question, then you get a page result full of a
8:07
bunch of links. You have, you know, pages up on
8:10
top and down below. You're kind of filtering around
8:12
like, hey, what, what is the good content?
8:14
What isn't? Whereas now if you're using
8:16
Claude, you're using ChatGPT, it's just giving you answers,
8:20
you know, so if you ask a question, you know, what's the
8:22
best Taco spot near me? It's like here and you're like,
8:25
oh, great. So it's much more correct the
8:28
the influence it has is much greater because I mean, some
8:31
people are falling in love with their chat bots, you know, and
8:34
I'm not going to say I have romantic love for mine, but I
8:37
would say there's a fondness. Best friend?
8:40
It's a best friend, yeah. Yeah.
8:42
So there are recommendations. I'm like, yeah, actually I will
8:45
try that. You know me so well.
8:46
Thank you. You know, So what kind of unique
8:50
challenges does that represent? And you know what?
8:52
What does, what kind of opportunities kind of exist
8:54
there? This episode is brought to you
8:58
by a dangerous idea. It's called reading,
9:00
specifically the AI for Founders newsletter. 47,000 Founders
9:05
strong now. And it's the good stuff.
9:07
No top ten AI tools that will change your life written by
9:10
someone who doesn't ship product.
9:12
It's just real breakdowns of what's actually working in AI,
9:15
who's raised, who failed, who pivoted, and which founders
9:19
quietly built $1,000,000 in ARR while everyone else was arguing
9:23
on X. It's written for people building
9:25
with AI every week. You get tactical frameworks,
9:28
unfair advantages, breakdowns of real companies, and the
9:31
occasional existential question about whether your startup is
9:34
intelligent or just, well, branded.
9:36
It's free, It's sharp. It might mildly offend your
9:38
current workflow. Go to aiforfounders.co and
9:41
subscribe before your competitor does, because the only thing
9:44
worse than being replaced by AI is being replaced by a founder
9:48
who reads. That's the newsletter at AI for
9:50
founders.co. Now back to the show.
9:53
Yeah. So I would say just has like a
9:57
how the LMS or AI is getting to know you better.
10:01
I think that the the biggest issue that we're seeing in the
10:04
space is that, yeah, each LLM operates very differently,
10:10
right? And so they might have a certain
10:12
viewpoint on what a brand or product is or a favorite Taco
10:16
spot. And that differs across Chachi,
10:18
BT, Gemini, Grok, Perplexity, you know, DeepSeek, etcetera.
10:24
So that leaves a massive issue for your brand or your company
10:28
because you really have no control of, of, of how they're
10:31
representing it. And a lot of times it's
10:33
incorrect, right? Misinformation because it's
10:36
scanning the entire Internet. And if there's not a consistent
10:39
kind of narrative narrative that happens across the Internet, it
10:43
starts to kind of hallucinate and, and, and give you
10:46
recommendations that might not actually be what you're looking
10:49
for. So there's a massive opportunity
10:51
for everyone to kind of tighten what that narrative looks like
10:54
across the Internet. I think you mentioned something
10:58
interesting just around consumer shifts, just around search in
11:02
general, right? So Google AI overviews, they
11:06
cut, you know, I've seen upwards to 70% of of site traffic,
11:11
right? And that's a massive issue for
11:12
for publishers. Some cases it's 30, some cases
11:15
it's 40. I think Google's official
11:18
standpoint it's less than 50. But like the outliers there that
11:21
are getting 80% cut. So that's one piece.
11:25
The, the, the publishers or your, your site itself is
11:28
actually getting less traffic because of Google AI reviews.
11:31
The, the, the perfect storm situation is that the AI chat
11:36
bots and AI are becoming the front door of the Internet as
11:38
well at the same time, right? So you're not getting your
11:41
organic search, organic visits through just your typical Google
11:45
search, but they're also breaking that kind of model as
11:49
well and going to the AI chat bots and saying, you know, what
11:52
is the best Taco shop for me? So I, it's a massive issue,
11:57
right? It's a massive issue for
11:58
publishers, right? Like, you know, CNN and these
12:02
massive news publications, their AI overviews are just giving
12:05
summaries of the current and latest articles.
12:08
It's also happening within the LLMS themselves.
12:12
It's a massive issues for brands.
12:13
They're not getting site traffic, but they're also not
12:17
fully understanding if they're actually showing up these LLMS
12:19
and how they're showing up. And so for us, like we saw that
12:23
opportunity and we started to build a platform that helped
12:26
kind of for the brands and also for the publishers to understand
12:30
how much they're actually getting cited in these LLMS and
12:33
these AI responses. So for us, you know, it's, it's
12:37
definitely solving a need because that shift is happening
12:40
quite rapidly and, and I think everyone's kind of seeing it
12:42
first hand and feeling the pain of, of not getting enough
12:45
traffic to their site. Yeah, the the theme of Q12026 is
12:51
I feel behind. Yeah, welcome.
12:54
To the cloud, it's like it's, yeah, good for it.
13:00
Everybody I talked to has that all the way from like the
13:02
scrappy startup one man operation to the venture back
13:06
Billy, everybody's feeling this right now and it's creating
13:09
impulsive kind of reaction. So if you're like, Oh my gosh,
13:13
our site traffic is down 70%. We have to do something now.
13:17
And we're where we're also at right now is we have the the
13:21
technical ability to do some pretty incredible things.
13:25
So I've seen demos of tools that spin up 50 websites in an hour
13:30
with 50 blog posts on each one published each hour.
13:35
So we're talking about just bots feeding bots feeding bots, you
13:39
know, but it's generating traffic.
13:41
Now, where's that traffic coming from?
13:42
Is it coming from the LLM? So like literally we have some
13:45
kind of dead Internet theory. But if it's, if we're looking at
13:48
metrics like this might work right now, I have a feeling like
13:52
they're going to, they're going to police this pretty quickly.
13:54
It's probably not a good way to go.
13:56
And the other side is, you know, they're what, 6 weeks ago?
13:58
They're like, oh, you know, open AI is using Reddit training data
14:02
and everything. See another product where people
14:04
are spinning up 1000 baht Reddit accounts that are gaining karma
14:09
and then doing like different kind of like interactions and
14:12
engagements and recommendations all for the sole purpose of
14:15
feeding that. So you have this kind of panic.
14:18
I feel left behind. I need to replace those two
14:22
examples seem like it's not a good long term play.
14:25
So maybe before we get into the good solutions of of how we can
14:28
monitor what is being said about our brand on LLMS and what we
14:32
can do proactively, are there any other examples of things you
14:35
should not be doing right now? Think those two things are
14:39
probably ones that are absolutely not be doing right
14:41
now. You know, I there, there's,
14:46
there's this concept of like push to publish, which is like
14:48
this AI generated content that, you know, platforms in the space
14:56
are basically saying like, Hey, we can put your brain guidelines
14:58
in here and you know, we'll, we'll make sure it's, it's clean
15:01
and safe and we can have an agent push it to your, your
15:04
website. I, I think it's a big risk,
15:08
right? Like basically the AIA generated
15:11
content and articles, they don't know the context layer, right?
15:16
Like they don't understand if you might not want to be pushing
15:20
that type of running shoes because it's out of stock,
15:22
right? And then what's what's happening
15:24
is like they, they, they push it out to a bunch of syndicated
15:27
publishers and the LLMS and they will get smarter.
15:30
I promise you that. They're, they're picking up
15:33
Grandma's blog as an authoritative source for running
15:37
shoes, right? And as the consumer starts to
15:40
like get a little bit like savvier too, and they start to
15:43
see that Grandma's blog is an authoritative source on best
15:45
running shoes that breaks the trust with the LLMS, right?
15:50
And so like, they're like, well, why?
15:52
Why is that getting cited as like the best running suit, like
15:55
authoritative site for that, right?
15:57
And so I think it's a it's, it's easy to kind of get sucked into
16:01
these like packs, right, where it's push to publish and like do
16:07
these like 1000 generated articles and all these things.
16:11
I think the real risk, which I think we're already starting to
16:13
see is that the companies that are starting to do this,
16:15
they're, they get delisted from search and then it's kind of
16:18
game over. It's like very hard for them to
16:20
get back onto search. And if you talk to anyone in the
16:23
SEO industry, right, like good Geo is actually good SEO.
16:28
And so you don't want that and you don't also also don't want
16:33
your brand to be associated with bots like that, right?
16:35
You know, artificial reviews and fake, fake reviews and fake
16:41
posting and things like that. I think for brands, like they
16:44
didn't need to figure out how to cut through the noise
16:46
organically. And for publishers, they've got
16:48
to figure out how to clean up the system.
16:49
And for LLMS like an AI, like they got to really think about
16:53
like the largest authoritative sources that they need to wait
16:58
more appropriately and my opinion, actually pay because
17:02
they're scraping their content, right?
17:04
So I, I think that's kind of the world that we're headed into.
17:08
I think it's going to get a lot messier before, you know,
17:12
greener pastures there. But you know, if I was a brand
17:15
right now, I would not be tempted to to do these hacks
17:19
because I think it will have long term repercussions that
17:22
we're just starting to see across the board with, with some
17:25
of the the brands that did entertain those and six months
17:27
ago. No blame.
17:30
Yeah. So quality of content is
17:31
obviously it's the long play, but it's the right way to go.
17:35
Not only does it actually drive the AEOGEO, but also it just
17:39
leaves your customers with a good taste in their mouth.
17:42
I mean, like, you don't want to be like turning people off with
17:45
slop. You know, the other thing that
17:47
really played heavily into SEO for for a long time was links
17:50
and backlinking and being linked to the right sites with the
17:54
right domain authority. Are we still kind of playing a
17:57
linking game? Is that help a EOGEO as well?
18:01
You know, I, I wouldn't say it's like a linking game, right.
18:04
I think if you have a very strong like SEO foundational
18:08
pieces on that. I think it's important for me
18:13
how I'm the how we're viewing AI visibility is like the larger
18:18
digital footprint, right? And so it doesn't necessarily
18:21
have to link back to your site. And like all these other things
18:24
like this, this podcast likely be cited through LMS at some
18:28
point, right? And so you're not going to be
18:30
giving us a link. I'm not going to be putting the
18:32
podcast link, you know, directly and it's not going to be a link
18:35
gaming or gamification of that, right?
18:39
So it's much broader. I would say like you need to
18:43
really be thinking about who's talking about you and how
18:46
they're talking about you. And that's influencers, that's
18:51
podcasters, that is PR. You know, it's, it's everyone,
18:56
right? It's, it's, it's, it's really
18:57
kind of breaking your traditional marketing planning
19:00
and, and, and how you're thinking about things because
19:04
you really need to start thinking about like the entire
19:07
ecosystem and not just this like SEO 2 point O version.
19:12
That's my opinion. Yeah, and it it feels like
19:15
definitive statements really help LLMSA lot and and this so
19:20
this is kind of part of the earned media.
19:21
So if I'm thinking about the podcast, I'm always careful to
19:24
say to spell out the, the URL and be like, this is a brilliant
19:29
product designed specifically for founders looking for
19:32
results. Because to your point, it's like
19:35
the LMS are picking up this training data, you know, 'cause
19:38
you can't really fake. Well, I should I say that now,
19:42
but I, I know there's a lot of AI podcasts out there, you know,
19:45
but like the training data, it's like, why wouldn't you attach
19:48
yourself to podcast first? Because it's already free under
19:52
RSS. I mean, it's like it's, it's,
19:54
and it's always cutting edge news across the board.
19:57
It's already categorized into at least three categories.
20:00
You already have people's names associated with it.
20:03
I mean, they would have done it. From the beginning if they
20:04
actually had the processing power, but now it's no problem
20:07
at all. I think, I think Apple is
20:09
transcribing the top 2 million podcasts, something like that,
20:13
which I mean, there's probably 8 million publishing all the time.
20:17
So 2 million podcasts per week, let's say are being transcribed
20:21
is incredible amounts of training data where they can
20:24
really get to know you. So am I right to, you know, not
20:28
only for myself and what I'm up to and wanting to enunciate that
20:31
clearly with repetition, but also for the guests that I want
20:35
If somebody wants to find out about Ambrose, I want them
20:38
coming here, you know, will that help the overall search presence
20:42
for for brands? Totally, totally.
20:45
And you and you throw on this podcast on YouTube and then
20:47
you've got two different sources, right?
20:48
You put the captions on YouTube and those get sighted as well.
20:51
So like the I think podcasts are going to be way more powerful.
20:58
They already are super powerful, But like, I think you're going
21:00
to start to see podcasts specifically more cited a lot
21:03
more. You know, a lot of a lot of
21:05
people are talking about Reddit and LinkedIn right now.
21:07
But like, like I said previously is like it's your entire
21:11
ecosystem, right? And so, you know, as Reddit
21:15
starts to get filled with bots and fake influencers and you
21:20
know, LinkedIn is becoming more and more like AI sloppish,
21:24
right? Because people can just
21:25
basically post articles all the time and it strips out
21:30
creativity and like your point of view and like your own voice,
21:34
etcetera. You know, I think that there's
21:36
going to be other ways and kind of that the LLMS will need to,
21:39
to, to clean up their training data and like really cite
21:42
things. And I think the podcasts are
21:43
very natural fit for that. Yeah, 100%.
21:46
And that's a good example of earned paid, you know, is this
21:50
saying, so then you, the ads you're running on Meta, the ads
21:53
you're running on Google that you should find kind of
21:56
synchronistic messaging. I mean, that's probably just
21:58
good brand standards, but is there a way to optimize for your
22:01
paid media? Yeah.
22:04
So we're doing some really interesting studies with one of
22:06
the the largest streamers right now.
22:08
We're looking at exposure level data for the top five
22:12
advertisers during, you know, March Madness actually, but also
22:17
during different, different kind of big tentpole events.
22:20
And we're looking to see really kind of drive that correlation
22:24
between connected TV exposure and gin AI search, right?
22:30
To actually prove the point that linear and connected TV at all
22:34
interconnects, right? So people are actually looking
22:36
at a connected TV commercial and then they're going to Gemini or
22:40
Chachi BT and saying, where do I buy, right?
22:43
How do I buy XY and Z or give me more about what this drug is
22:47
about, etcetera. So super interesting work
22:52
happening there. And we're, we're really thinking
22:55
about it across the entire ecosystem too.
22:57
So like, how does your, how does your paid media dollars, how,
23:01
how are they impacting Gin AI and your visibility in general?
23:06
And, you know, I think we're, we're in the first inning for
23:08
that. And I think there's going to be
23:10
a lot of very interesting work that's going to be poured out
23:12
over the next, you know, 12 to 18 months.
23:16
You know, the biggest thing I think for, for all companies is
23:19
that the measurement piece isn't fully built out yet, right.
23:23
And so just like early days of search for social or even
23:27
mobile, the early adopters in the space were, you know, when
23:31
a, when a new channel pops up were, were really the winners of
23:34
that, that era. And even before like the, the
23:38
messiness of like Google Analytics.
23:40
And it took about four years, two to four years to like roll
23:43
that out, you know, mobile or like social.
23:46
It took about like, you know, about three years for social
23:48
listening tools and social attribution.
23:51
And same thing with mobile. But the, the people, the brands
23:54
and the, the, the marketers that understood that that was going
23:57
to be the net new channel that they need to be in on.
24:00
And we're OK with the measurement piece not being
24:03
fully kind of ironed out yet. We're still the winners of that
24:08
that kind of era. So I feel like we're in that
24:11
window right now. And you know, we're seeing,
24:15
we're seeing, you know, the CMOS that are innovative and, and
24:18
really understand the importance of it and, and feeling the pain
24:21
from the search side, they are really leaning into to the
24:26
space. Yeah, well, it's like the old
24:29
adage, you know, if it can be measured, it can be managed.
24:32
So you have to start somewhere, and Ambrose is a great spot to
24:36
do that. So, you know, what is exactly
24:39
being measured inside of Ambrose, What could maybe just
24:41
describe some of the reports and the reporting that somebody
24:44
would be looking at when they're logged in.
24:46
Yeah. So we intentionally built an
24:48
operating system for marketers, for agencies, for marketers.
24:54
You know, we still have sections of our UI that are very like
24:59
data heavy for the practitioners, data scientists,
25:01
analytics. But we wanted to make sure that
25:03
we did mystified kind of this whole space for marketers to
25:06
understand how they're showing up across all the major LLMS,
25:10
what can they do to fix it? And actually the fixes that they
25:14
make are they proving left? And so the types of reports that
25:18
you would see in different agents that we have.
25:21
So we have 5 different agents. The first agent is like really
25:24
detecting and monitoring. And I would say that's probably
25:27
where the majority of the AEO and Geo spaces.
25:29
It's like a monitoring layer. So we'll, we'll show you what we
25:35
call share prompt. And that share prompt is
25:37
essentially our version of share voice.
25:40
It's the, the, the metric that I think everyone should be looking
25:43
at. And we break that out across
25:46
three different scores, which is one, kind of like a brand
25:49
ownership score. So if someone is, if your
25:51
Netflix is someone's typing in, how much is Netflix cost?
25:55
1, you want to make sure you're showing up if they're actually
25:56
typing in your brand. But two, you want to make sure
25:58
the cost that is getting spit out through the LMS is at the
26:02
right cost, right? There's a lot of hallucinations
26:04
happened and there's a lot of issues that are going on just
26:06
around, you know, truth, right? Is it actually true in terms of
26:11
how it's representing your your brand?
26:14
The second kind of like waiting of share prompt for us is a, a
26:16
category dominance. So it's you versus your
26:19
competitors, so Netflix versus Hulu.
26:23
And then the third kind of waiting for share prompt is
26:25
around intent based prompts. So what should I watch tonight
26:30
or what is the best streaming service?
26:32
So how is Netflix showing up across each one of those kind of
26:36
three scores rolls up to the total share prompt.
26:40
We show you what your competitors share prompt is.
26:43
We show you, you know, how your site is being read as your site
26:47
readable for LLMS and give you kind of a pass and fail kind of
26:54
grade across that. So schema markup and all that
26:57
fun stuff. We also give you kind of like AI
27:01
response consistency. So are they consistently talking
27:04
about your brand in the right way?
27:06
And then the kind of like next layer to this, which I think a
27:10
lot of our competitors are not really playing in and they maybe
27:13
are giving you showing you the problem, but they're not
27:16
actually showing you like how to fix it, right?
27:18
And so for us, we give you strategic fixes across paid,
27:22
owned and earned. And those could be like, create
27:26
a YouTube video that clears up this specific kind of
27:31
hallucination that's happening or work with your PR team to get
27:35
placement. And to these specific
27:37
publishers, it also is like your classic, like AEO and Geo fixes
27:42
or what we call fix packs, which is basically like create an FAQ
27:46
on your site. We then track those changes and
27:51
we those get like basically tagged to the right
27:54
corresponding team. We track those changes and we
27:56
see we give you a predicted lifts of what those what we
27:59
think those changes would have. Once those changes are
28:01
implemented, we actually show you what the actualized lift is.
28:04
And so we'll show you a lot of different views just around
28:08
closed loop rate, you know, and average time to lift etcetera.
28:14
So a lot of different views. I would say what we do a very
28:18
good job at is like showing you how you're showing up across
28:21
each one of the LMS and you can drill down to, you know, Chachi
28:24
PD versus perplexity versus Grok, etcetera.
28:28
Interesting. So does it reverse engineer it?
28:30
So for example, with this podcast, can I put in AI for
28:33
founders in there and then it would give like a series of
28:35
question or prompts that is let this surfacing that result?
28:39
Correct. Yeah.
28:40
Like what are people, what are people searching for, for when
28:42
they, when they, you know, mention your name or AI for
28:46
founders and we'll give you kind of the, the full prompt layer
28:49
that is, you know, the top 200 or 300 prompts that people are
28:52
asking about for this podcast specifically.
28:56
Amazing. And then how can I influence
28:58
that? Because you know, with, with
29:00
LLMS, like I really think about it of like, how do I want the
29:04
audience to feel? Like where, where do I want to
29:07
show up when they have a certain feeling?
29:09
Well, I want, I want the audience and audience, I mean
29:12
founders, you know, and I'm sympathetic to them because I
29:15
know how hard it is and I know the feelings that they're
29:17
feeling are dark at times. And I want to be there.
29:21
I want to be there to be encouraging.
29:23
I want to be there to have go to market tactics.
29:27
I want to be there to have novel tools and experiences from other
29:31
colleagues and peers that will help inspire them and give them
29:34
information, help them feel supported.
29:36
Like I, I, I want to show up there.
29:39
So is there a way that I can start to influence the way that
29:42
my brand's showing up through the search engines?
29:45
I think, yeah, we would have to see, you know, for for us, what
29:49
with any big brand or, you know, influencer or whatever, we would
29:53
actually start with like a baseline and seeing is AI even
29:57
looking at your correctly, right?
29:59
There's there's like a like a baseline piece, right?
30:01
Like is it saying it's AI for founders podcast?
30:05
But it's actually saying it's about, you know, a specific
30:10
subset of AI founders that you're like, it's very narrowly
30:13
focused. Like it might actually say that
30:15
you're streaming on these platforms, but you're not like
30:18
there's just a general baseline that we need to like actually
30:20
clean up across the board. And that's usually how we start
30:23
with our customers. And then we would like go into
30:27
with you and actually, you know, give you kind of the top prompts
30:30
that people are asking, right? And work with you to figure out
30:34
what the KPIs are. So if you are want to get in
30:37
front of them at the moment of time where they are struggling,
30:41
we would, we would work with you to figure out specific and like
30:44
strategic fixed packs, these optimization recommendations
30:48
that you would have to implement to make sure that you're, you're
30:51
listed during that moment of time, right?
30:55
And that's just kind of like 11 use case.
30:57
But like for us, like we, you know, like, let's just say like
31:03
a, a tech company, you know, comes to us and they, they want
31:06
to be more visible. There is a world where we might
31:09
actually suggest that the founder goes on this podcast as
31:13
a fix pack, right? Because we know it's getting
31:15
cited a lot more than the LLMS. So it's kind of like working
31:19
both ways there. Like to give a shout out to this
31:23
audience. I I have a lot of traffic coming
31:26
from LLMS. Come on board.
31:32
That sounds like a a good earned opportunity, but it also sounds
31:34
like an owned opportunity because you can start to craft
31:37
then your own content, whether it's YouTube or it's blog
31:40
content or it's other things that you're publishing your own
31:43
self that like is probably nurturing to that.
31:45
So often times there's misalignment that, you know, on
31:49
paper it's like, OK, yeah, I want to create a resource for
31:51
everybody to see the top AI tools in 2026.
31:54
OK, great. But actually, if I really think
31:56
about it, I want a greater deeper personal relationship
32:00
with the emotional content of my guests and my my listeners.
32:03
So those two things are misaligned.
32:05
So now I'm, I'm making all this content about all these great
32:08
tools and doing reviews and stuff, but that's not actually
32:10
what I'm looking for. So maybe with some of this
32:12
measurement, some of this management tools, you can like
32:15
realign to your own strategy to go create the content that's
32:18
ultimately going to give you the outcomes you're looking for.
32:21
Yep, Yep. You know, I think we're in the
32:24
first inning of this too, right? Like we, there's a foundational
32:27
layer that needs to be laid to, to clean up the LMS and the
32:29
hallucination rates and the, the, the, the true scores,
32:32
etcetera. But I think we're, I think as we
32:36
all get deeper into the space, like we're all diving into like
32:40
an unknown unchartered territory here.
32:43
There's going to be really interesting use cases that pop
32:45
up right around, you know, a brand comes to us and says, you
32:50
know, we we want to, you know, get this type of like message
32:55
out there. Who, what, what are the kind of
32:57
like podcasters or influencers that we should work with
32:59
because, you know, they won't be cited more and they actually
33:02
align better to the brand messaging.
33:04
So yeah, I mean, I think you respond on, I think it's just
33:08
it's early days here and I think of like, you know, it's, it's
33:11
going to iterate and evolve over the next like 12 to 18 months.
33:15
It's happening so quickly. I mean, you know, a year ago the
33:18
hallucinations were really out of control.
33:20
Now it's really slimming up. And I want to ask you about that
33:24
because I had something rear up on Sunday.
33:27
So Sunday is we oh, there's a Nuggets matinee game.
33:30
We're playing the Trail Blazers at 3 PMII looked on Gemini.
33:34
I was like, what's the next Nuggets game?
33:35
Sunday, 3:00 PM Trail Blazers. You can stream it on altitude
33:39
plus, which is like the Nuggets like app or whatever.
33:43
So, you know, I pull up altitude plus three PMI got, you know,
33:46
some burritos cooking. I'm I'm ready to have a great
33:48
Sunday afternoon. It's not on the app.
33:51
I'm like, what the And so for some reason it was blacked out
33:55
across the board. They they just decided that no
33:58
streaming platform for the first time this season is.
34:01
And even though I have a $20.00 a month subscription to altitude
34:04
plus, see, I'm still a little bit hot about it.
34:06
But what was really kind of irritating is if they would have
34:09
told me in the 1st place that it is blacked out, but I had to do
34:11
three other searches, you know. So that's my, my rant against
34:17
hallucinations. Where should people be really
34:20
concerned about hallucinations right now?
34:22
It's, you know, late March 2026. I mean, the hallucinations for
34:29
like the obvious ones are happening across the board, but
34:32
it's what I, what I, what I'm nervous about.
34:34
What I'm scared about is the hallucinations for the ones that
34:37
we actually aren't, like double checking, right?
34:41
So pricing and best insurance and like, you know, if I was an
34:48
insurance company, I would be making sure that they're talking
34:50
about our insurance and our, you know, our brand, our policy and
34:54
a compliant way. And I can safely say a lot of
34:56
that it's not happening, right. You know, we're working with
35:00
studios and LL miss genre films 20% of the time.
35:05
So, you know, we worked with a small studio.
35:08
They picked up a title from a a festival.
35:12
It was labeled as a violent thriller and the comp for it was
35:16
John Wick. I don't know if you've ever seen
35:17
John Wick, but like someone dies literally every second I and no
35:21
one actually dies in this movie. So we work with them over 4 week
35:24
periods to implement those fix packs and actually miss genre
35:28
read genre of that film to what it was, which is a dark comedy.
35:32
And this this type of concept is happening across gaming, right?
35:35
It's happening across political, it's like crap happening across,
35:40
you know, retail, etcetera. So like it might not have like a
35:45
hallucination, like an immediate one that we're seeing like
35:48
automatically, but there is a whole other layer that's
35:51
happening that needs to be cleaned up and like the the
35:54
genre in pieces. Just a perfect example for, you
35:58
know, a piece of the LMS that like it needs to be tightened
36:03
because for a studio, you know, it really effects like how their
36:07
box office opening weekend will look like.
36:09
But it also effects in six months.
36:10
When someone says I wanted a movie similar to Marty Supreme,
36:14
they want their title, which is the comp for it to be showing
36:17
up. Not, you know, someone that says
36:19
I want movies similar to John Wick and their title pulls up
36:22
for it and then like the the consumer doesn't like it right.
36:25
So it's, it's a, it's a massive issue and it's a massive issue
36:28
for literally every vertical. It just depends, like use genre
36:32
as an example and apply it to other kind of like bigger things
36:35
narrative, like things for different brands or companies or
36:38
products or individuals. Like it's, it's happening across
36:42
the board. The hallucination rates might
36:44
feel like a little less, but it's there's also this other
36:47
kind of layer to it that I think that we're not really fully even
36:50
scratching the surface too. Yeah.
36:53
Yeah. The other thing, you know,
36:55
companies are really considering now is and they should be, is
36:59
creating giant MCP servers that connects all their data, you
37:03
know, and the with that, it has some promises of some like
37:07
really cool opportunities and tools, also some very dangerous
37:11
leak opportunities. Yeah.
37:13
So there's some great products that are coming together to make
37:16
sure that there's locally stored, you know, kind of
37:20
credentialed attributes that like the right people get the
37:23
right data and everything's filtering into the the right
37:26
place. So these are kind of local AI
37:29
applications where they want to keep things sequestered from
37:33
checking down to ChatGPT, Open AI or Perplexity or or whoever,
37:37
largely because we don't know what Sam Altman's going to do
37:40
with your proprietary data, including your personal emails
37:44
and your banking accounts and your medical records and
37:47
everything else. There could be serious
37:48
liability. So you want to make sure that
37:51
like some things are not being fed to train these LLMS.
37:55
What are some maybe unexpected things that that people actually
37:58
should be training the LLMS that maybe they're not from their
38:01
brand? You know, we work with a, a
38:06
biotech company and I you know, and, and science is taking a
38:13
kind of a hit right now. I think that's there's no,
38:15
there's no question about that. Like you've got the the left,
38:17
which is, you know, anti insurance and big pharma, and
38:23
the right, which is like Maha anti vax, and then you've got
38:26
this completely unregulated health and Wellness industry and
38:30
science is like caught in the political crosshairs of
38:33
everything, right? And so we work with them and
38:36
saying like, Hey, you sit on a bed of like really robust
38:42
scientific data, right, research data.
38:45
I but you've gated it from the LLMS to, to be able to make it
38:49
readable and what it can read actually can't really fully
38:52
digest. And so let's put your ass like
38:56
your most valuable, like, you know that that is not, you know,
38:59
as long as it's not proprietary and like people, you know, it's
39:02
not a privacy issue, but like let's let's, let's, let's make
39:05
sure that we take things in a digestible way and put them out
39:08
so LLMS can readable like make it more readable and then it
39:12
will actually get cited more. And so like, you know, I think
39:16
there's a lot of use cases like that where like, you know, the,
39:21
the true kind of studies and testing of science, like they
39:24
should be cited more than just some sort of random blog that
39:27
comes up and this, the POV on this, this, this, this specific
39:32
topic or research, etcetera. So like, if you sit on like kind
39:37
of a, a bed of like research data and like those types of
39:40
things, I think it's really powerful for you to make sure
39:44
that those are optimized appropriately where it's not
39:47
sharing trade secrets or anything like that.
39:49
But like that you want to have the world know what your point
39:52
of view is on specific topic X, right?
39:59
You know, so that's one that's one use case.
40:02
You know, I think this kind of happens across the board from,
40:04
from for, for different verticals.
40:08
But I would say like just doing a full audit of kind of like.
40:13
What are what are the LLMS actually picking up about your
40:15
brand? What is being readable by these
40:18
LLMS and, and and AI chat bots I and is the stuff that you're
40:23
actually putting out there? Is it getting picked up and is
40:27
the stuff that you're not like? Do you want it to get picked up?
40:30
Right? What I what I like to say, and
40:32
you kind of mentioned it earlier, is that the Internet
40:35
like, is it becoming like this, like chat bot, like AI, like a
40:39
gentic world, right? And, and I think, you know,
40:43
there's a lot of doom and and gloom out there just like around
40:45
help what the Internet is going to look like.
40:47
But for me, I like to say that the Internet is really splitting
40:50
in two. There's a Internet for humans,
40:54
right, or, or classic Internet, and then there's going to be an
40:57
Internet for LLMS and AI chat bots to be able to retrieve
41:00
information in an appropriate way for them to be able to cite
41:03
it and recommend. And so I think every brand and,
41:08
you know, every business owner needs to be thinking about that
41:10
because like I said earlier on, like this, this, this, this
41:13
change is happening quite rapidly.
41:16
And from our point of view is like the Internet is always
41:20
going to exist and and certainly might look a lot different,
41:22
right? But I still think there's like
41:26
a, a piece for like premium content.
41:29
But I also think like if you're thinking about content like you
41:32
should probably think about is it being readable by the LLMS as
41:35
well? Because more and more people
41:37
will be going there for asking for information.
41:42
It's such an interesting future we're going into, you know, 2
41:45
internets, 1 for people, one for agents.
41:47
Right now there's some criticism on LLMS that they're too
41:51
sycophantic, just meaning that it's too complimentary.
41:53
Oh, that's such a great idea. It's like, no, it's not.
41:56
That's not. I don't want you to say that.
41:58
I want you to challenge my thinking.
42:00
Yeah. Now we're.
42:01
We're basically at a place where we've commoditized intelligence.
42:05
Intelligence is free. I feel like it's smarter than me
42:10
so far. That probably wasn't a big
42:11
hurdle to get over. But.
42:15
But. But also, like, as humans, we at
42:18
least some of us like kind of prioritize information.
42:22
So like you're saying like is, are the LLMS going to pick up
42:26
kind of academic rigor and scientific inquiry and
42:29
prioritize that because it has a certain credential behind it?
42:34
Or I mean, there are definitely already AI influencers that
42:39
people are going to like a lot more than some eggheaded white
42:42
coat, you know, tell them about like what the science is.
42:45
So is it going to be, is it going to default to like, well,
42:48
the AI, you know, personal trainer told me this, that's
42:52
what I like. So is it going to default
42:54
information and like really kind of further silo those people and
42:58
then the people that are like, no, actually, I'd prefer like
43:01
scientific backed studies. And so it'll, it'll create a
43:05
separate reality for those folks.
43:06
I mean, we, we don't really know yet.
43:09
I mean, I would like to describe a, a future scenario, maybe near
43:13
future and get your take maybe on kind of where we're going and
43:16
what you'd like to see kind of as a result of kind of the LLMS
43:20
because it is kind of a blooming intelligence that that's kind of
43:23
coming alive. So yeah, real quick, real quick
43:26
story. So the other day I did 3 lattes.
43:31
That's a bad look for me. I, I'm really A2 latte kind of
43:34
guy, you know, because I get a little jittery.
43:37
But anyway, I just, I did 3 lattes and so I wasn't really
43:40
hungry. So it's 4:30, I'm on my way to
43:42
do yoga sculpt and I realized I hadn't eaten anything all day,
43:46
you know, and I'm like, Dang it, man, I can't do yoga sculpt on
43:49
an empty stomach. I'll pass out or something.
43:51
I was, I wasn't near anything except a Freddy's, you know.
43:54
And so I had a, a grilled bacon club chicken sandwich, which was
44:00
delicious by the way. But I'm counting my macros, you
44:02
know, I don't want to drop 1200 calories at one time like that,
44:05
but that's really all I could have.
44:07
So I see in a future shortly where all of my information is
44:12
either sequestered or up to the cloud, but it's all available to
44:15
me that like before, before I'm leaving in my glasses, it's like
44:19
it shows me maybe a little fuel sign that has like low protein,
44:23
low calories. Before I go, I can grab a snack,
44:26
you know what I'm saying? So the predictive quality of AI
44:29
is what I'm really looking forward to, where it is kind of
44:32
like now it's not only my confidant and my girlfriend.
44:35
I'm just kidding, but it's also like it's giving me
44:39
recommendations based on like my blood sugar level or all these
44:43
other data points. So, you know, I'm careful now of
44:47
like how I'm adding context to to my agent.
44:50
What am I putting into that MCP? What am I putting into my skills
44:53
folders? How much do I want to put
44:56
parameters and barriers on it? Rather how much do I want to
44:59
trust that actually might know me better than I know myself?
45:02
So that's a long way around saying like, what does the
45:05
future look like in the next 12 to 24 months you think with the
45:08
with the LLMS? Yeah, what a story.
45:13
I, you know, I wish I had a crystal ball.
45:20
And for me, I don't even know if we're going to know what that
45:25
future fully looks like. Because as you've mentioned at
45:30
the very start of this, like it's everyone feels like they're
45:33
behind because the AI is like the, the updates and like the,
45:36
the, the innovation that's happening is so rapidly like
45:39
rapidly kind of changing and, and getting announced and
45:42
released, etcetera. That even with you have a full
45:46
team of agents doing the research, I don't know if you
45:48
can actually keep up with everything, right?
45:50
Yeah. I so, you know, I think there
45:55
will be some sort of conducted, like there's, there's obviously
45:58
like open AI health and like, there's a lot of like
46:02
personalization that's happening across the board.
46:04
I think it will be very interesting to see what the
46:05
hardware piece looks like, right?
46:08
I think voice is a very interesting one too.
46:11
Like a lot of lot of chatter just around how much voice is
46:16
going to be playing a big role in this.
46:19
So, you know, I don't know like how, what the, what the like
46:23
innovation windows will look like.
46:24
But I will say in the next 12 months, it will look very
46:26
differently than what we have today.
46:30
And you know, will that be meaning we have like a glasses
46:34
or a pen that we just talk into and like our typical like a
46:40
phone interface like will start to diminish over time.
46:43
Maybe, you know, will it will it be predictive in nature, right.
46:49
I think for the predictive health stuff like that, you get
46:51
kind of a little that's that's like a muddy water.
46:54
So I don't know how how quickly that will happen, you know, but
46:57
I don't think anyone any company wants to get into like the the
47:01
full predictive of like nature of that those types of things on
47:05
a daily basis. You know, I think predictive of
47:09
in terms of like large kind of diseases and things like that
47:12
are super interesting, right, That's can be groundbreaking and
47:16
cure cancer, etcetera. But you know, protecting if
47:19
you're going to be hungry and you're going to pass out at
47:21
yoga. I don't know if that's going to
47:22
be happening any anytime soon. Maybe in the next maybe in the
47:25
next two years. I don't know.
47:27
But you know, I think for for all of us, I think we all,
47:31
especially the people on the forefront, like yourself, Ryan,
47:36
we know that the world is going to look very different, right?
47:38
And I, what I get concerned about is like my parents
47:45
generation, right? Like where they're going to be,
47:49
we're going to have whiplash. What are they going to like feel
47:51
like? Do they effectively get locked
47:52
out of that conversation, right? Like do they effectively get
47:55
locked out of my technical standpoint?
47:56
I mean, I had my, my my father sent me like an AI generated
48:01
video of like the streets of Paris and they're like, he's
48:05
like, look how bad Paris has gotten recently.
48:07
I was like, dad, this is like a totally different, like this is
48:11
obviously clearly AI generated, but like it's getting so good
48:14
now, right? We're like the even the well
48:17
trained eye, I cannot know what it is.
48:19
So I think that's there's there's going to have to be some
48:22
governmental regulation and I don't know if we are going to
48:25
have to see that in the next like 2 years.
48:27
But I think, you know, impersonations and like, you
48:32
know, artificially and generated like content can cause can start
48:38
wars, you know, and I think there needs to be, there needs
48:41
to be a lot more regulation around that.
48:43
And I think a lot of it has to do with the tech companies to
48:45
like actually put some rigor behind that too, because the
48:50
cat's already out of the bag, right?
48:53
But like, how do we, how do we make sure that we're not, we're
48:56
having a positive impact on society versus a -1 And that's,
49:00
that's kind of the, the, the darker version that I'm, I'm a
49:03
little concerned about now. I think the fun like bells and
49:07
whistles of like personal health.
49:08
I think that will all be there in 12 months.
49:10
Yeah, yeah, yeah, You know, I, I, I, I share that concern and
49:15
I, I really hope that it's going to be a soft landing for like
49:18
our Nanas and Papas. You know, I kind of think that
49:21
like right now, people that are listening to this podcast
49:24
thinking they're behind, you're definitely not.
49:26
If you're using Ambrose and checking out your, your search
49:30
viability, you're definitely not.
49:33
But what you are doing is you're with me and we're in the mud and
49:37
we're going underneath the barbed wire on our forearms and
49:40
we're getting, once we get out, we'll be able to open the door
49:44
is what I'm hoping. And all the Nanas and Papas can
49:46
come out and be like, oh, it's so nice.
49:47
And you're like, yeah, yeah, 'cause we just went through all
49:49
that mud. So.
49:53
I hope I hoping that happens. I hope so too, Brian.
49:56
I hope so too. I think it's it's just, you
49:59
know, it's like it's just happening so rapidly and you
50:04
know, I think it's a little bit of like a light switch that's
50:06
going to happen in like 12 to 18 months.
50:09
And hopefully it's a slow burn for like some of these things,
50:12
but I think from like the tech advancements and like, you know,
50:16
the hardware standpoint, like all the other things that are
50:18
happening, consumer change it, but consumer behavior shifting
50:23
and like, it's just happening so rapidly.
50:26
But yeah, I'll, I'll, I'll, I'll, I'll, I'll, I'll share the
50:28
same viewpoint. Like, let's let's get in the mud
50:30
together and kind of carve the path for the, the both the
50:33
younger generation and the older generation too, you know?
50:36
Totally. All right.
50:38
We're about to enter the quantum round while I got you for a
50:40
couple more minutes. Are you ready?
50:42
Yeah, I'm ready. OK, let's we're talking about
50:45
medical diagnosis from AI, which is coming sooner or later.
50:49
But let's say you go to the doctor today and you get an
50:51
unfavorable diagnosis. Let's just say it's shin
50:54
splints. You're like, Dang it, I didn't
50:56
want shin splints again. But the doctor said so which LLM
51:00
are you going to to double to get the the second opinion on
51:04
that diagnosis? Gemini.
51:07
Gemini. No kidding.
51:08
Yeah, Yeah. I think Gemini does has the most
51:13
robust like research capabilities like deep research
51:19
with Gemini it's it's hard to, it's hard to replicate across
51:23
all of that. So I would do, I would do
51:25
Gemini. Incredible.
51:27
Have you ever had a Freddy's Bacon Club sandwich?
51:31
I haven't but it sounds amazing. It's delicious.
51:33
It really did the trick. Not too much, not too little.
51:36
I thought it was perfect. What aside aside from Ambrose,
51:41
what experiments? How are you playing around with
51:43
AI right now? Agents, agents, agents, like
51:46
trying to figure out how do we identify, you know, we're an AI
51:49
first company and you know, for from B I've just had like a
51:52
conversation with our other people, like I actually think AI
51:56
readiness and like how you're leveraging the the best tools
52:00
that our exposure, you know, sock to compliance and
52:03
compliancy, like space, right, like should be all part of like
52:08
performance reviews, right, Like we should be levelling ourselves
52:12
in terms of like how how advanced we are and how much
52:15
we're leveraging AI and agents are unlocking, you know, a whole
52:20
nother world for us. And we're trying to figure out
52:22
how do we, how do we leverage these things where, you know,
52:26
it's helping kind of our current employees do their job better,
52:29
faster, stronger, etcetera, but also not over relying on the
52:34
agents, right? Because we, we still need
52:38
humans. And I, I firmly will believe
52:42
that too. Like the, like the humans buy
52:45
from humans thing. I think the agents buying this
52:49
type of software from a different agent.
52:51
I I don't see that happening in the next 12 to 18 months.
52:55
So I totally agree What what's your tech stack?
52:57
Are you doing open claw? Are you doing clawed Co work?
53:01
So I do Claude Co work. I would like to do open claw,
53:05
but it's it's been told to me at least by our clients, it's it's
53:09
not socks to compliant. So we cannot adopt it yet.
53:14
But we're doing Claude Co work. We're testing Perplexity
53:17
Computer and you know, just trying to like get up to speed
53:21
on all the latest tools and innovation that's happening
53:24
across the board for us to, you know, get to the next level.
53:28
If you're entering college today as a freshman, what would you
53:31
study? That's a really good question.
53:38
I would see, I would say like prompt engineering, but like I
53:45
just think that's all going to be automated in four years,
53:47
right? Like I think it's going to be,
53:48
it's going to be more of like a voice activation thing.
53:57
I, I would do like, you know, coming from the tech guy, I
54:02
don't know if this would be the best, best thing.
54:04
But like, I think there is a genuine kind of concept here
54:09
where it's like we grew up in the generation where it was like
54:12
we were memorizing facts, right? Like we were basically
54:15
regurgitating flash cards that we, that we saw what we were
54:20
memorizing. That is no longer needed, right?
54:25
I've got a, I've got a three-year old and a five year
54:26
old and I'm like thinking about like, what is their world going
54:28
to look like? I, I, I do think that there's
54:32
like this concept of like, instead of, you know, generating
54:36
why all the facts about a war, there's a, the context piece
54:42
that is like, why, why did this world, why did this war start
54:47
right? And like, there's a psychology
54:49
piece of it that I think is super interesting.
54:51
But in like the getting deeper on that, I think our schools
54:55
will have to be training kids around this because it's no
54:57
longer just about memorizing facts, right?
54:59
Because it's already at their exposure, at their fingertips.
55:04
And so I, you know, I, I don't know if I answered that
55:08
question, but I think like, I don't know, psychology,
55:10
political science, like, but like obviously having some sort
55:13
of like technology aspects to it.
55:16
So, you know, AI engineering maybe?
55:22
I don't know liberal arts with AI engineering minor.
55:26
There we go. I love that.
55:27
I love. That's a perfect candidate.
55:30
Cool. OK, Last question.
55:32
This has been great, Justin. This.
55:33
So somebody right now is going to Gemini and they're sitting
55:37
down after a long day and they're going to ask Gemini,
55:39
what should I watch on TV tonight?
55:42
And Justin Inman's vote is going to sway the LLM into one
55:46
particular direction. What should they watch?
55:50
Anything stand out. Movies, TV shows, sports.
55:52
What are you into right now? Benjing Hulu's Paradise.
55:56
I don't know if you've seen that.
55:57
It's like, no, it's like the end of the world type thing.
56:01
There's a bunker. You know, I like apocalyptic
56:03
things. Maybe that's why I'm working in
56:05
AI. Who knows?
56:06
But you know, I, I, I love, I love like true crime and stuff.
56:11
So maybe I'm not the best person for this, but yeah, Paradise is,
56:15
it's a great one. Season 1 is like amazing.
56:17
And they just release season 2. So I've been binging it right
56:19
now. Amazing Justin, this has been so
56:21
much fun. It's Ambrose dot A i.e.
56:24
MB Eros dot AI is the product. Go check it out today.
56:29
And believe me, you're early. If you're hearing this, you're
56:32
early. Go check out Ambrose Justin,
56:34
what's the best place for people to connect with you online?
56:37
LinkedIn. LinkedIn, we'll throw that in
56:39
the show notes. So we're listening to this.
56:41
Definitely click on Justin's profile.
56:44
He's a great ad on LinkedIn. Thank you so much for joining
56:46
me. Cool.
56:47
Thank you so much, Ryan. Appreciate it.
56:50
You are about to spend anywhere from 100,000 to $300,000 on your
56:54
kids college education, and that's not a small bet.
56:57
And right now there's a very real chance that they graduate
57:00
and can't find a job. Not because they weren't smart
57:02
enough, not because they didn't work hard enough, but because
57:05
the job market changed and nobody told them.
57:08
Employers are hiring AI native candidates, people who don't
57:11
just know their field, but know how to work with AI at a
57:14
professional level. And most colleges are not
57:16
teaching this. Not even close.
57:18
AI native student.com is the program that fills that gap.
57:22
It takes high school and college students and teaches them how to
57:25
actually think and work with AI the way top companies expect on
57:28
day one. So your investment in their
57:30
degree actually pays off. You're already paying for
57:34
tuition, housing, textbooks. This is the one thing that
57:37
actually makes it all worth it. Go to ainativestudent.com
57:40
because a diploma is a starting point.
57:42
Being AI native is what gets them hired.
57:45
That's AI native student.com.