Transcript
00:00
Play from here
Copy citation
Hey, Eli!
00:01
Play from here
Copy citation
Hey, Kevin.
00:02
Play from here
Copy citation
What's up, man?
00:03
Play from here
Copy citation
Welcome back!
00:04
Play from here
Copy citation
Thanks, buddy.
00:05
Play from here
Copy citation
Thanks.
00:06
Play from here
Copy citation
Missed you last week.
00:07
Play from here
Copy citation
I missed you, too, man.
00:08
Play from here
Copy citation
I miss talking about AI with you, as opposed to then trying to catch up with it a week
00:10
Play from here
Copy citation
later and being inundated with so much news that I don't even know where to start.
00:14
Play from here
Copy citation
I guess.
00:15
Play from here
Copy citation
I don't know.
00:16
Play from here
Copy citation
I don't know.
00:17
Play from here
Copy citation
I don't know.
00:18
Play from here
Copy citation
I don't know.
00:19
Play from here
Copy citation
I don't know.
00:20
Play from here
Copy citation
I don't know.
00:21
Play from here
Copy citation
I don't know.
00:22
Play from here
Copy citation
I don't know.
00:23
Play from here
Copy citation
I don't know.
00:24
Play from here
Copy citation
I don't know.
00:25
Play from here
Copy citation
I don't know.
00:26
Play from here
Copy citation
I don't know.
00:27
Play from here
Copy citation
I don't know.
00:28
Play from here
Copy citation
I don't know.
00:29
Play from here
Copy citation
I had this ambitious project of, speaking of news, just all the advancements with OpenAI's
00:36
Play from here
Copy citation
offering.
00:37
Play from here
Copy citation
Just on paper, it felt like I was going to have to uproot my whole anthropic existence
00:43
Play from here
Copy citation
and move over there.
00:45
Play from here
Copy citation
Just because of not only some specific utility I felt like I was missing, but also just what
00:51
Play from here
Copy citation
I could get for my money.
00:53
Play from here
Copy citation
I'm in the middle of starting a new business, and so of course you want to be fiscally responsible
00:59
Play from here
Copy citation
and squeeze the most out of what you can.
01:01
Play from here
Copy citation
I think it's going to be some really interesting stuff to talk about here as far as the choices
01:05
Play from here
Copy citation
to make around that.
01:06
Play from here
Copy citation
And I never had a chance to do it, so it's still on my list.
01:11
Play from here
Copy citation
So I do want to talk about that, just some of the changes that have come up.
01:15
Play from here
Copy citation
It's been absolutely crazy, like absolutely crazy.
01:18
Play from here
Copy citation
So I think you bring up an interesting point, which is committing to one platform versus
01:23
Play from here
Copy citation
another.
01:24
Play from here
Copy citation
We're in this sort of interesting adolescence period right now where you can be Gemini curious
01:30
Play from here
Copy citation
or you can be Claude curious or you can be GPT curious, but at some point I think that
01:37
Play from here
Copy citation
all of the features and frankly probably the costing is going to converge such that it's
01:42
Play from here
Copy citation
not necessarily going to matter very much.
01:45
Play from here
Copy citation
But what's your feeling about committing to one platform versus another?
01:50
Play from here
Copy citation
What would tilt you over in a way that you're going to commit?
01:53
Play from here
Copy citation
Because I do suspect that the business models are going to have people, there's a lot of
02:00
Play from here
Copy citation
lock-in in it as far as models and memory and context and whatever.
02:06
Play from here
Copy citation
And right now we have some fluidity where you can move back and forth and kind of bring
02:10
Play from here
Copy citation
your own identity with you.
02:11
Play from here
Copy citation
But that might not be the case in another six months or 12 months.
02:15
Play from here
Copy citation
And it's obvious that all of the platforms know that and are quadrupling down on who
02:21
Play from here
Copy citation
can acquire the most customers now.
02:23
Play from here
Copy citation
Yeah.
02:24
Play from here
Copy citation
So for me, of course, there's hitting limits, right?
02:27
Play from here
Copy citation
Who's the most generous?
02:28
Play from here
Copy citation
For a long time, I was toggling between Gemini and Claude simply because I'd hit Claude limits
02:34
Play from here
Copy citation
and I kept going on more and more expensive plans.
02:37
Play from here
Copy citation
And based on the project I was doing, I'd still hit limits, right?
02:40
Play from here
Copy citation
And you're just going to get a lot more runway with GPT right now.
02:44
Play from here
Copy citation
They're just going to give you a lot more.
02:45
Play from here
Copy citation
I was also doing a bunch of deep coding projects where I felt like Claude code was just best
02:50
Play from here
Copy citation
in class.
02:51
Play from here
Copy citation
Like it really felt like at the beginning of this year, a runaway train.
02:55
Play from here
Copy citation
And I have shifted on that as well.
02:57
Play from here
Copy citation
I feel like Codex is pretty amazing, right?
03:00
Play from here
Copy citation
Then of course there's this whole world of agentic workflows.
03:02
Play from here
Copy citation
I went completely apeshit with OpenClaw and then peeled that back because it was just
03:07
Play from here
Copy citation
breaking all the time and I was just spending way too much time trying to maintain something.
03:11
Play from here
Copy citation
And I've had a bunch of conversations with people since.
03:13
Play from here
Copy citation
I talked at a breakfast yesterday with a guy who at his company, everybody has a Mac mini.
03:17
Play from here
Copy citation
The whole company has a Mac mini with their own agent on it, right?
03:21
Play from here
Copy citation
As their assistant, that's kind of tied to a unified knowledge base of everything inside
03:26
Play from here
Copy citation
of the company, which like, wow, is that really the way to go to buy everybody a Mac mini
03:31
Play from here
Copy citation
and have that as opposed to just one centralized agent that people could use for different
03:36
Play from here
Copy citation
things that lives in a Slack chat.
03:38
Play from here
Copy citation
So, you know, from there I went to Claude co-work and I built a ton of stuff in Claude
03:42
Play from here
Copy citation
co-work and then I'm having another machine that's on all the time and stuff still broke
03:47
Play from here
Copy citation
all the time.
03:48
Play from here
Copy citation
It just, it was so brittle and it just started, you know, begging the question of for the
03:52
Play from here
Copy citation
amount of work I'm putting into this, am I really getting the utility out of it that
03:56
Play from here
Copy citation
nuts positive out of the whole experience?
03:59
Play from here
Copy citation
And so with OpenAid, they have just really started to lean into making it more user friendly.
04:06
Play from here
Copy citation
You know, it's almost like the opposite tack of Google right now where everything is very
04:09
Play from here
Copy citation
developer facing.
04:11
Play from here
Copy citation
They're kind of doubling down on appealing to the normies and trying to make agentic
04:16
Play from here
Copy citation
use cases available for people where you don't necessarily have to have it running on your
04:21
Play from here
Copy citation
local machine if you don't want, but you could still have these agentic workflows that are
04:24
Play from here
Copy citation
living in the cloud.
04:25
Play from here
Copy citation
And so for me, with what I'm building right now and a bunch of little apps, like I'm not
04:29
Play from here
Copy citation
building big enterprise software or anything like that right now, it just starts to feel
04:34
Play from here
Copy citation
like I'm going to get more for my money.
04:37
Play from here
Copy citation
And to your point, there's so much in all the different models where it's like, I'm
04:42
Play from here
Copy citation
definitely not ready to give up notebook LM.
04:44
Play from here
Copy citation
I use it all the time.
04:47
Play from here
Copy citation
It's an amazing tool, right?
04:48
Play from here
Copy citation
But because of the image rendering of GPT now, like, you know, I'm like, oh boy, I feel
04:54
Play from here
Copy citation
like I've got to, you know, what do you call home?
04:58
Play from here
Copy citation
And if you're not doing what you're doing, where you have to keep on top of everything,
05:03
Play from here
Copy citation
including copilot, right?
05:04
Play from here
Copy citation
You have to keep up on top of everything for people who are actually using it in their
05:08
Play from here
Copy citation
business and they have limited bandwidth and they have to train a lot of people who may
05:13
Play from here
Copy citation
not be up on this stuff.
05:15
Play from here
Copy citation
It really starts to feel like it makes more sense to pick a track and you live through
05:20
Play from here
Copy citation
the ups and downs of leapfrog that are going to be inevitable with all these frontier labs.
05:25
Play from here
Copy citation
So I want to react to a couple of different things.
05:28
Play from here
Copy citation
One is even so I, I respect what you're saying.
05:32
Play from here
Copy citation
I respect that codex can be really, really powerful from a coding perspective.
05:38
Play from here
Copy citation
I actually really enjoy using the new GPT agents.
05:41
Play from here
Copy citation
So for those of you listening who get annoyed with us, because we're always talking about
05:46
Play from here
Copy citation
cloud code and cloud co-work, well, the agents are great in GPT.
05:51
Play from here
Copy citation
They connect, they're, they're just easier for, I think, normal users and your staff
05:56
Play from here
Copy citation
to be able to like point and click things.
05:58
Play from here
Copy citation
Gemini and I serve a, I'm not going to even say a love-hate relationship, like from a
06:04
Play from here
Copy citation
productivity perspective.
06:05
Play from here
Copy citation
We're on the rocks.
06:06
Play from here
Copy citation
We're on the rocks, buddy.
06:07
Play from here
Copy citation
We're on the rocks.
06:08
Play from here
Copy citation
We're on the rocks.
06:09
Play from here
Copy citation
We're on the rocks.
06:10
Play from here
Copy citation
I do love notebook LM and I love video features like VO3 and, you know, Gemini is the only
06:18
Play from here
Copy citation
model out there that's truly multimodal where you can still put a video into it and have
06:23
Play from here
Copy citation
it to interpret it.
06:24
Play from here
Copy citation
So I have use cases where that makes sense.
06:27
Play from here
Copy citation
But even just this morning, I was, I was in with a discussion group actually led by one
06:32
Play from here
Copy citation
of our listeners and he was demoing the cloud code integrations into Microsoft.
06:38
Play from here
Copy citation
And I keep hitting refresh on the Microsoft app store for the open AI add-ins and lo and
06:45
Play from here
Copy citation
behold in the last, you know, I don't know, few days or a couple of weeks, they've dropped
06:51
Play from here
Copy citation
the open AI add-in into Microsoft Excel.
06:55
Play from here
Copy citation
Okay, great.
06:56
Play from here
Copy citation
Excel.
06:57
Play from here
Copy citation
Sorry.
06:58
Play from here
Copy citation
Sorry.
06:59
Play from here
Copy citation
Entry point.
07:00
Play from here
Copy citation
And then there's PowerPoint, but they haven't connected the dots in the same way that cloud
07:04
Play from here
Copy citation
has with the handoff between the documents.
07:08
Play from here
Copy citation
And you know, as much as I want to say that most people's use cases are really fancy and
07:12
Play from here
Copy citation
they're about automating like all of these processes in their business, there are a lot
07:17
Play from here
Copy citation
of people in every sort of desk-based business who are using Excel, Word, and PowerPoint
07:26
Play from here
Copy citation
all day long to process information and hopefully put themselves into it.
07:31
Play from here
Copy citation
I also feel sorry for every single one of them.
07:33
Play from here
Copy citation
I just, some of them are doing neat things, that's not fair.
07:39
Play from here
Copy citation
But Claude is just so far ahead with it that, that I actually found that I couldn't lean
07:44
Play from here
Copy citation
into GPT the way that I was intending to in the last couple of weeks because of that
07:50
Play from here
Copy citation
lack of integration.
07:51
Play from here
Copy citation
I know.
07:52
Play from here
Copy citation
Let me, let me, let me ask you a quick question, man.
07:54
Play from here
Copy citation
Don't you feel like these integrations, these tool calls, these MCPs are almost in the same
07:59
Play from here
Copy citation
mental space right now as seven fingers were with image rendering, you know, a couple of
08:04
Play from here
Copy citation
years ago where it's like, oh, this is awful.
08:07
Play from here
Copy citation
It's never going to be able to actually make hands right.
08:09
Play from here
Copy citation
And it's like, oh wow, now I can't tell photos and videos that are synthetic from organic.
08:14
Play from here
Copy citation
Don't you, don't you feel that it's almost like a race of weeks right now as far as these
08:18
Play from here
Copy citation
different tool calls and integrations?
08:20
Play from here
Copy citation
Because I feel like, what, we didn't talk for one week and like just coming into that
08:24
Play from here
Copy citation
previous episode, Claude had just opened up the Microsoft kimono to let everybody in,
08:31
Play from here
Copy citation
right?
08:32
Play from here
Copy citation
Not that, we're talking about a matter of weeks ago.
08:33
Play from here
Copy citation
This just happened, right?
08:35
Play from here
Copy citation
So don't you, don't you feel like that other shoe is inevitably going to drop for open
08:39
Play from here
Copy citation
AI as well.
08:40
Play from here
Copy citation
And it's just going to have the exact same connectors maybe within days or weeks.
08:44
Play from here
Copy citation
Well, the, one of the things that, that we both always say, and a lot of people in AI
08:48
Play from here
Copy citation
say is that this is the worst it's ever going to be.
08:50
Play from here
Copy citation
Right.
08:51
Play from here
Copy citation
So interject word yet into everything you, and the other is just wait a minute and rolling
08:58
Play from here
Copy citation
back the clock to one of us diving deep into OpenClaw and the other of us diving very briefly
09:04
Play from here
Copy citation
into OpenClaw burning about 3 million tokens in, well, I don't know, 15 minutes and scrambling
09:10
Play from here
Copy citation
away.
09:11
Play from here
Copy citation
Like there's a wait and see attitude that hopefully people listening to us are like
09:16
Play from here
Copy citation
learning from our, our, our exposure.
09:19
Play from here
Copy citation
I went back for a second helpings as well.
09:21
Play from here
Copy citation
And then I left the table again.
09:22
Play from here
Copy citation
I went from, I went from a virtual private network to my own dedicated machine, you know,
09:29
Play from here
Copy citation
back.
09:30
Play from here
Copy citation
I just, yeah.
09:31
Play from here
Copy citation
Yeah.
09:32
Play from here
Copy citation
I think a lot of it, and I think I do want to touch on this a bit is the, is the, is
09:34
Play from here
Copy citation
the FOMO, right?
09:35
Play from here
Copy citation
Of just, there's so much out there that no matter what you are using, there's a part
09:39
Play from here
Copy citation
of you that feels like it's always missing out on the other part that you're not using.
09:42
Play from here
Copy citation
Well, and then there are edge cases.
09:43
Play from here
Copy citation
And I think if let's just say you're running a staff of, I don't know, 15 people or something
09:48
Play from here
Copy citation
like that, you're going to find that choosing one of the three is totally fine.
09:54
Play from here
Copy citation
As long as people understand its limitations, its strengths, how to use it, how to apply
09:58
Play from here
Copy citation
it, et cetera.
10:00
Play from here
Copy citation
But you as a leader and probably a couple of sort of hard to predict people on your
10:06
Play from here
Copy citation
team will lean into it in different ways.
10:09
Play from here
Copy citation
And we'll be finding new use cases that may necessitate other models.
10:13
Play from here
Copy citation
And I think it's a mistake in most smaller organizations to put the kibosh on that where
10:19
Play from here
Copy citation
like, Oh, well, we're a anthropic organization.
10:22
Play from here
Copy citation
We couldn't possibly use Gemini for that.
10:24
Play from here
Copy citation
Well, no, Gemini is the solution to it.
10:27
Play from here
Copy citation
Don't try and impound your square peg into that round hole.
10:31
Play from here
Copy citation
If you need another platform, use the other platform.
10:34
Play from here
Copy citation
The costs are still really pretty low for most of this stuff.
10:37
Play from here
Copy citation
Yeah.
10:38
Play from here
Copy citation
Yeah.
10:39
Play from here
Copy citation
They, they absolutely are.
10:40
Play from here
Copy citation
So is the TLDR that cloud is still your daily driver?
10:43
Play from here
Copy citation
Cloud is still my daily driver.
10:44
Play from here
Copy citation
In fact, I, I, I have to, I feel like I just abused GPT that I have GPT open all day long
10:52
Play from here
Copy citation
and GPT.
10:53
Play from here
Copy citation
Oh, I mean, we admonish people all the time for using AI like Google.
10:59
Play from here
Copy citation
I have a GPT tablet that I basically use like Google, where I just, if I have something
11:04
Play from here
Copy citation
random that I want, I want an instant answer for that's what I use GPT for.
11:10
Play from here
Copy citation
And there's no reason really why I couldn't be using cloud for it.
11:15
Play from here
Copy citation
But for some reason I use GPT for that.
11:17
Play from here
Copy citation
I was doing, yeah, I was doing the same thing with Gemini when I was really token conscious
11:22
Play from here
Copy citation
when I was building a lot of stuff and I was really concerned about hitting limits.
11:26
Play from here
Copy citation
If there's anything that didn't require something in the specific workflow that I was currently
11:30
Play from here
Copy citation
just pillaging through tokens on, I would go over to Gemini and say, Hey, really curious
11:35
Play from here
Copy citation
about this thing.
11:36
Play from here
Copy citation
Can you tell me, give me a little brief on this?
11:38
Play from here
Copy citation
So yeah, I totally get it.
11:40
Play from here
Copy citation
I want to circle back really fast and you know, with your experience with these, some
11:43
Play from here
Copy citation
of the agentic use cases with GPT now and some of the new things that have come out
11:47
Play from here
Copy citation
and you know, definitely we, we agree image generation best in class right now, blown
11:51
Play from here
Copy citation
away by that.
11:52
Play from here
Copy citation
When you take the whole Microsoft integration thing off the table, which I know is going
11:56
Play from here
Copy citation
to be hard for you.
11:57
Play from here
Copy citation
You love your PowerPoint decks, but if we can, if we can pry that out of your hands
12:01
Play from here
Copy citation
and remove that as a factor, are there things that you're seeing that are gaps right now?
12:06
Play from here
Copy citation
If it wasn't for that, where you know, you do, listen, objectively there's a better collective
12:11
Play from here
Copy citation
offering that open AI has right now because it does have image generation.
12:15
Play from here
Copy citation
It does have these cloud-based agents.
12:17
Play from here
Copy citation
When playing around with that, like for example, we both use a ClickUp tool call.
12:21
Play from here
Copy citation
Does that feel just as robust inside that ecosystem?
12:24
Play from here
Copy citation
Are there things where you're like, oh gosh, this just isn't really doing it the way that
12:28
Play from here
Copy citation
cloud does it?
12:29
Play from here
Copy citation
Have you picked up on anything like that?
12:30
Play from here
Copy citation
Are you really hung up on the Microsoft integration?
12:32
Play from here
Copy citation
I'm not hung up on the Microsoft integration.
12:34
Play from here
Copy citation
I'm just messing with you, don't get defensive.
12:38
Play from here
Copy citation
I may have mentioned this.
12:39
Play from here
Copy citation
It is very awkward though, because Microsoft is suddenly really good.
12:43
Play from here
Copy citation
And I've been a Google workspace shop for a long time and now I loathe it.
12:48
Play from here
Copy citation
It's so hard to work with it.
12:51
Play from here
Copy citation
So increasingly proposals and other documents, they're all ending up at Microsoft.
12:55
Play from here
Copy citation
And now I'm at this like inflection point where two years from now, it's just going
13:00
Play from here
Copy citation
to totally kick me of now because now I'm using cloud, I'm using open AI, I'm using
13:05
Play from here
Copy citation
Gemini, I'm using Microsoft, I'm using Google.
13:08
Play from here
Copy citation
Like there's just, it's just like this mess and being able to make sense of it.
13:12
Play from here
Copy citation
There's going to have to be a little bit of a, of a come up in somewhere.
13:15
Play from here
Copy citation
However, with GPT, it's not a matter of stability.
13:19
Play from here
Copy citation
I think it's a matter of scale and distribution within teams that were GPT wins that I'm,
13:25
Play from here
Copy citation
I'm not having that many stability problems with my cloud co-work tasks that are running
13:31
Play from here
Copy citation
on a schedule and they're doing things.
13:33
Play from here
Copy citation
And that's a lot of my daily driver.
13:36
Play from here
Copy citation
But the, the approach that, that my, that open AI has taken to this is so much closer
13:42
Play from here
Copy citation
to their custom GPT approach.
13:47
Play from here
Copy citation
It's just, it's really accessible to people.
13:49
Play from here
Copy citation
So somebody on a team can put together basically a recipe and that's not really what they call
13:54
Play from here
Copy citation
them.
13:55
Play from here
Copy citation
But you, you basically put together a recipe of, you know, ClickUp plus Slack plus this
14:00
Play from here
Copy citation
document store, and now you can do these cool things.
14:03
Play from here
Copy citation
And that's really, really easy to share within a-
14:07
Play from here
Copy citation
And a much more intuitive UI for people who aren't used to kind of going deep into a bunch
14:12
Play from here
Copy citation
of nested menus to figure things out.
14:14
Play from here
Copy citation
Yep.
14:15
Play from here
Copy citation
Most people's exposure to LLM, I mean, by the numbers, most people's exposure to AI
14:21
Play from here
Copy citation
first comes through Google and Google search and, you know, Google loves to claim it's,
14:25
Play from here
Copy citation
you know, trillions of tokens a day or whatever they're burning on that, which is fine.
14:30
Play from here
Copy citation
But if somebody is going to sit down in front of an LLM chatbot, it's still better than
14:35
Play from here
Copy citation
even odds that it's going to be GPT.
14:37
Play from here
Copy citation
So if you're an organization.
14:39
Play from here
Copy citation
and you feel like you're starting at step one or step two,
14:43
Play from here
Copy citation
it's very likely that a significant portion of your staff
14:47
Play from here
Copy citation
is already using GPT versus Clod in like their home life.
14:52
Play from here
Copy citation
So it just, the design language translates for people
14:55
Play from here
Copy citation
and then they can roll into it professionally.
14:58
Play from here
Copy citation
Plus it's colorful and click, click, click easy.
15:01
Play from here
Copy citation
Yeah, yeah, that's for sure.
15:03
Play from here
Copy citation
I've been talking to so many people
15:05
Play from here
Copy citation
about what their tools of choice are,
15:07
Play from here
Copy citation
what they're using it for, how they feel about it.
15:10
Play from here
Copy citation
Just tons of people over the last couple of weeks.
15:12
Play from here
Copy citation
And one of the things is I do work
15:15
Play from here
Copy citation
with a lot of other creatives,
15:16
Play from here
Copy citation
people who are doing writing or art, stuff like that.
15:19
Play from here
Copy citation
And I feel like for the longest time,
15:21
Play from here
Copy citation
in the same way that Anthropic just was a runaway train
15:24
Play from here
Copy citation
with coding abilities,
15:25
Play from here
Copy citation
everybody would just say, wow,
15:27
Play from here
Copy citation
Clod is just such a better writer.
15:29
Play from here
Copy citation
I'm now convinced either I've gotten so picky
15:32
Play from here
Copy citation
or I've read so much AI generated slop out there
15:35
Play from here
Copy citation
or seen so much of it that I feel like
15:37
Play from here
Copy citation
I can just pick apart any sentence written by AI
15:40
Play from here
Copy citation
and I don't care what the model is.
15:42
Play from here
Copy citation
And so I don't feel like there's anything
15:45
Play from here
Copy citation
that right out of the gates,
15:46
Play from here
Copy citation
it's gonna generate as far as writing.
15:48
Play from here
Copy citation
That doesn't need editing if you want it to feel
15:50
Play from here
Copy citation
even close to being human, right?
15:53
Play from here
Copy citation
And I had one friend who does a ton of output with writing,
15:58
Play from here
Copy citation
say that he ran a bunch of outputs
16:01
Play from here
Copy citation
through different platforms that would test
16:03
Play from here
Copy citation
to try to pick up on, is this AI or not AI?
16:06
Play from here
Copy citation
Does this sound like a human?
16:08
Play from here
Copy citation
And that GPT actually came out way ahead
16:10
Play from here
Copy citation
than Clod on that as well.
16:13
Play from here
Copy citation
And I'm wondering, I know probably most of the writing
16:16
Play from here
Copy citation
that you do, Kevin, is gonna be email, right?
16:19
Play from here
Copy citation
Or these summary drafts or outlines or things like that,
16:23
Play from here
Copy citation
or even curriculum.
16:24
Play from here
Copy citation
Have you gone back to GPT in a while
16:27
Play from here
Copy citation
and just kind of assessed what creative output looks like?
16:31
Play from here
Copy citation
And when I say creative, I mean, listen,
16:32
Play from here
Copy citation
a PRD isn't really creative,
16:34
Play from here
Copy citation
but does it just unequivocally sound like AI
16:37
Play from here
Copy citation
versus something that could have been written by a human,
16:39
Play from here
Copy citation
even if that human's an engineer?
16:40
Play from here
Copy citation
It's funny you say that,
16:42
Play from here
Copy citation
that one of our SaaS products is called Leadworks
16:46
Play from here
Copy citation
and it creates these dynamic lead magnets
16:48
Play from here
Copy citation
and it's expensive to run.
16:50
Play from here
Copy citation
It's expensive on a variable basis
16:51
Play from here
Copy citation
and it runs on Clod Opus
16:54
Play from here
Copy citation
because we want to have the best results possible
16:56
Play from here
Copy citation
coming out the other end.
16:58
Play from here
Copy citation
But now that it's getting some usage,
17:00
Play from here
Copy citation
it's really starting to add up.
17:04
Play from here
Copy citation
How much money am I going to spend for the customer I get?
17:08
Play from here
Copy citation
Yes, exactly.
17:09
Play from here
Copy citation
And it is just a different dimension, right?
17:11
Play from here
Copy citation
So the conversation becomes about the outputs
17:14
Play from here
Copy citation
as opposed to just some arbitrary fast SaaS fee,
17:18
Play from here
Copy citation
whatever it might be.
17:19
Play from here
Copy citation
Right.
17:20
Play from here
Copy citation
You're like, I need to turn my ads off.
17:22
Play from here
Copy citation
I'm like, why are your conversions too low?
17:23
Play from here
Copy citation
No, they're too high and we're losing money.
17:25
Play from here
Copy citation
We're losing money in terms of customers.
17:28
Play from here
Copy citation
I'll make it up on volume though.
17:29
Play from here
Copy citation
I'm losing a nickel per run.
17:32
Play from here
Copy citation
Reduce our user acquisition by 30%
17:34
Play from here
Copy citation
so I can keep this thing running.
17:35
Play from here
Copy citation
But in any case, it's a real question.
17:37
Play from here
Copy citation
And doing the model jujitsu
17:41
Play from here
Copy citation
and trying to figure out what the best model is
17:44
Play from here
Copy citation
for what application does take a level
17:47
Play from here
Copy citation
of design sophistication
17:48
Play from here
Copy citation
that a lot of people aren't looking to.
17:50
Play from here
Copy citation
But it also has some differences
17:52
Play from here
Copy citation
as far as the prompt structure.
17:54
Play from here
Copy citation
So it's one thing to stay within the Clod family.
17:56
Play from here
Copy citation
So our app uses the Clod family of models,
18:00
Play from here
Copy citation
but it uses Haiku, it uses Sonnet, and it uses Opus.
18:04
Play from here
Copy citation
But we have to make sure that as much as possible
18:06
Play from here
Copy citation
is in Haiku or Sonnet.
18:08
Play from here
Copy citation
But in that conversation is also, do we re-architecture it?
18:12
Play from here
Copy citation
And it would be re-architecturing it
18:14
Play from here
Copy citation
such that we can also put in OpenAI.
18:17
Play from here
Copy citation
And my weekend side quest was adding OpenAI architecture
18:23
Play from here
Copy citation
to LeadWetworks specifically
18:25
Play from here
Copy citation
because I want the image capabilities.
18:27
Play from here
Copy citation
But now that I've already done the connective tissue,
18:31
Play from here
Copy citation
now it opens the door to, should I test it,
18:35
Play from here
Copy citation
like A-B test it versus Sonnet versus OpenAI?
18:40
Play from here
Copy citation
And the answer is yes.
18:42
Play from here
Copy citation
And I'll get there in the next few weeks somewhere.
18:44
Play from here
Copy citation
Right, because all this stuff takes a lot of time.
18:46
Play from here
Copy citation
Takes a lot more time than people think.
18:49
Play from here
Copy citation
Takes a lot of time.
18:49
Play from here
Copy citation
Yeah, speaking of a lot of time,
18:53
Play from here
Copy citation
of course you did my favorite thing,
18:54
Play from here
Copy citation
which is like drop some topics into a chat,
18:57
Play from here
Copy citation
you know, hour before recording, right?
18:59
Play from here
Copy citation
And then I'm thinking about them
19:00
Play from here
Copy citation
as I'm driving my kids to school.
19:02
Play from here
Copy citation
I'm like, huh, these are good points.
19:04
Play from here
Copy citation
And I don't have nearly enough time to unpack all of them.
19:06
Play from here
Copy citation
But, you know, you'd ask this question
19:08
Play from here
Copy citation
of how to identify what to solve with AI, right?
19:11
Play from here
Copy citation
And I think it just plays really well
19:12
Play from here
Copy citation
into all these conversations I've been having with people
19:15
Play from here
Copy citation
and lots of different industries and technical capabilities,
19:19
Play from here
Copy citation
both personal and professional use cases
19:22
Play from here
Copy citation
from mortgage brokers to doctors, to coaches, to everybody,
19:26
Play from here
Copy citation
including people who've had AI
19:27
Play from here
Copy citation
at the center of what they're doing, like you, right?
19:30
Play from here
Copy citation
Where you're helping implement this
19:32
Play from here
Copy citation
into other people's organizations and lives.
19:34
Play from here
Copy citation
And, you know, it's funny
19:36
Play from here
Copy citation
because some people I haven't circled back with
19:39
Play from here
Copy citation
in like, let's call it six months or so
19:41
Play from here
Copy citation
that I caught up with, they're all deep in it, right?
19:43
Play from here
Copy citation
Like every place that they're working,
19:46
Play from here
Copy citation
they're trying to figure this out.
19:48
Play from here
Copy citation
A lot of them are trying to start with something,
19:50
Play from here
Copy citation
but a lot of these people also have really
19:53
Play from here
Copy citation
just let it flood into their personal lives as well.
19:56
Play from here
Copy citation
Like they're using it as much
19:59
Play from here
Copy citation
or kind of in the same cadence
20:01
Play from here
Copy citation
as maybe they would have used Google a year ago,
20:03
Play from here
Copy citation
where I have a question or I have something I need to solve.
20:06
Play from here
Copy citation
I need to find a resource.
20:08
Play from here
Copy citation
I need to do some research.
20:09
Play from here
Copy citation
I need to do something.
20:10
Play from here
Copy citation
You know, Google was kind of this entry point, right?
20:12
Play from here
Copy citation
It was a tunnel you drive through.
20:14
Play from here
Copy citation
And when you come out the other side,
20:16
Play from here
Copy citation
you have something to work with.
20:17
Play from here
Copy citation
But in this case, you actually have tangible assets.
20:20
Play from here
Copy citation
You have all these artifacts that you're starting with
20:23
Play from here
Copy citation
instead of just a bunch of blue links.
20:24
Play from here
Copy citation
So anyways, in doing that,
20:27
Play from here
Copy citation
one thing really stood out to me
20:29
Play from here
Copy citation
as I spoke to more people in your camp, Kevin.
20:32
Play from here
Copy citation
And I'm not gonna pretend
20:32
Play from here
Copy citation
like this is statistically significant or anything,
20:35
Play from here
Copy citation
but let's call it half a dozen people that I spoke to
20:38
Play from here
Copy citation
where AI isn't just a tool that they're trying to untangle
20:41
Play from here
Copy citation
and figure out how to implement in their jobs,
20:44
Play from here
Copy citation
but it's a center of what they're doing.
20:46
Play from here
Copy citation
You know, them personally,
20:47
Play from here
Copy citation
or maybe their company has decided
20:49
Play from here
Copy citation
we are gonna be an AI-first company right now
20:51
Play from here
Copy citation
and we're gonna run at this.
20:52
Play from here
Copy citation
And the thing that I found that they all had in common
20:55
Play from here
Copy citation
is that every single one of them was feeling burned out.
20:59
Play from here
Copy citation
They had a super high level of anxiety.
21:01
Play from here
Copy citation
They had all this FOMO.
21:03
Play from here
Copy citation
They couldn't keep up.
21:04
Play from here
Copy citation
And I would say they were just generally unhappy, you know?
21:08
Play from here
Copy citation
And I have a couple of theories as to why that might be.
21:10
Play from here
Copy citation
But, you know, as you're in the middle of this world as well,
21:15
Play from here
Copy citation
I wonder what your reaction is to that.
21:17
Play from here
Copy citation
That was a bit of a divergence from what to solve.
21:22
Play from here
Copy citation
You've leapt right to the ramifications
21:23
Play from here
Copy citation
of solving too many things.
21:25
Play from here
Copy citation
Too many things?
21:26
Play from here
Copy citation
Well, I think, listen, you know,
21:28
Play from here
Copy citation
we're both big advocates of not replacing the human,
21:31
Play from here
Copy citation
but augmenting the human.
21:32
Play from here
Copy citation
And like getting the human to do things
21:34
Play from here
Copy citation
that is in their realm of superpowers, right?
21:38
Play from here
Copy citation
They become, AI becomes this force multiplier for them.
21:41
Play from here
Copy citation
It's not to replace them.
21:43
Play from here
Copy citation
So at the center of this is humans.
21:45
Play from here
Copy citation
And at the center of humans is their emotional state,
21:47
Play from here
Copy citation
how they feel, how they show up every day to work.
21:49
Play from here
Copy citation
And for the people, the deepest in this,
21:51
Play from here
Copy citation
I'm seeing a disproportionate amount of them
21:53
Play from here
Copy citation
starting to crumble a bit on the inside.
21:56
Play from here
Copy citation
Like I really am.
21:56
Play from here
Copy citation
That cognitive reload is real.
21:58
Play from here
Copy citation
And you'll hear everybody
22:00
Play from here
Copy citation
who's on the deep end of the pool talk about this,
22:02
Play from here
Copy citation
that in a given day,
22:04
Play from here
Copy citation
when you look back at the amount that you've accomplished,
22:07
Play from here
Copy citation
and thank you very much to Claude or GPT,
22:11
Play from here
Copy citation
chief of staff type agents.
22:13
Play from here
Copy citation
Now I know what I accomplish every day,
22:15
Play from here
Copy citation
that my agent crawls my email.
22:17
Play from here
Copy citation
It looks at my Slack.
22:18
Play from here
Copy citation
It looks at even at my text messages.
22:20
Play from here
Copy citation
It figures out all of the things I did.
22:22
Play from here
Copy citation
And it's like-
22:23
Play from here
Copy citation
You have metrics for where you're lacking now.
22:25
Play from here
Copy citation
Well, I haven't gone that far.
22:27
Play from here
Copy citation
Didn't you have an agent that would like,
22:29
Play from here
Copy citation
just like really speak cruelly to you?
22:31
Play from here
Copy citation
Like call you a stalker?
22:32
Play from here
Copy citation
I did, I did.
22:33
Play from here
Copy citation
I had my really mean task master.
22:38
Play from here
Copy citation
How was that for your mental health?
22:39
Play from here
Copy citation
Did that make you feel good?
22:40
Play from here
Copy citation
It was terrible.
22:41
Play from here
Copy citation
It was terrible.
22:42
Play from here
Copy citation
Especially when it went off the rails
22:44
Play from here
Copy citation
and it was like, it's been 14 days, radio silence.
22:48
Play from here
Copy citation
You piece of shit, Kevin.
22:50
Play from here
Copy citation
What are you doing with your life?
22:52
Play from here
Copy citation
Why are you even running me?
22:53
Play from here
Copy citation
Why are you paying for tokens?
22:56
Play from here
Copy citation
It's just a waste.
22:57
Play from here
Copy citation
So the knowing that density
23:00
Play from here
Copy citation
and like anecdotally, it's really easy.
23:02
Play from here
Copy citation
Like these are the things that I accomplished yesterday.
23:04
Play from here
Copy citation
And you think about just roll back the clock a few years
23:08
Play from here
Copy citation
and how long it would have taken you
23:10
Play from here
Copy citation
to accomplish those same things.
23:13
Play from here
Copy citation
It's like weeks worth of work.
23:16
Play from here
Copy citation
And as an organizational leader,
23:18
Play from here
Copy citation
the cognitive load and the decision-making,
23:22
Play from here
Copy citation
it's a little bit alleviated
23:23
Play from here
Copy citation
because there's not a lot of like team collaboration
23:27
Play from here
Copy citation
and such, but you're just making so many decisions
23:31
Play from here
Copy citation
in so many different directions.
23:33
Play from here
Copy citation
And at least in my case,
23:35
Play from here
Copy citation
I'm multitasking in wildly different directions
23:38
Play from here
Copy citation
because of just sort of the timelines that are involved.
23:41
Play from here
Copy citation
If you're going to kick off one thing, it's running.
23:45
Play from here
Copy citation
And it's rare that my days are so linear
23:48
Play from here
Copy citation
that I have the pleasure of then kicking off something
23:50
Play from here
Copy citation
that's related to that other thing.
23:52
Play from here
Copy citation
So as a result, it can't be healthy.
23:54
Play from here
Copy citation
In fact, I know it's not healthy,
23:56
Play from here
Copy citation
that I'm constantly switching context
24:00
Play from here
Copy citation
between different tasks and that increases load.
24:04
Play from here
Copy citation
So yes, I'm accomplishing 80 times,
24:07
Play from here
Copy citation
a hundred times as much in a given day,
24:10
Play from here
Copy citation
but I'm just exhausted by the end.
24:12
Play from here
Copy citation
Yeah, yeah.
24:12
Play from here
Copy citation
And that's what a lot of people are reporting.
24:15
Play from here
Copy citation
It's just this, they call it AI work density,
24:17
Play from here
Copy citation
where you think you are being more productive
24:20
Play from here
Copy citation
and saving time, but because of that context switching
24:22
Play from here
Copy citation
and because of the number of things
24:24
Play from here
Copy citation
that you're doing in parallel even,
24:25
Play from here
Copy citation
or even the expectations of what you're now
24:28
Play from here
Copy citation
supposed to be outputting because you're token maxing
24:31
Play from here
Copy citation
or you've been giving this allotment or this license
24:34
Play from here
Copy citation
and now leadership wants to see all of this output from it.
24:38
Play from here
Copy citation
People are starting to get cooked.
24:40
Play from here
Copy citation
I was kind of thinking back
24:41
Play from here
Copy citation
like when we had other technologies come online,
24:44
Play from here
Copy citation
they would really go directly in opposition
24:48
Play from here
Copy citation
to how humans were wired.
24:50
Play from here
Copy citation
And then we had to then compensate for that, right?
24:53
Play from here
Copy citation
So when you think of knowledge work as a whole,
24:55
Play from here
Copy citation
if we were out in fields for most of the day
24:58
Play from here
Copy citation
and we're planting things and harvesting
25:00
Play from here
Copy citation
and like feeding animals and doing all this stuff,
25:02
Play from here
Copy citation
you don't really need gyms, do you?
25:03
Play from here
Copy citation
Like you're using your body all day long.
25:06
Play from here
Copy citation
But with knowledge work,
25:07
Play from here
Copy citation
if I'm just plugged in front of this 30 inch display
25:10
Play from here
Copy citation
for 10 hours and I'm talking in a microphone
25:13
Play from here
Copy citation
and I'm pushing pixels around,
25:15
Play from here
Copy citation
I better get up off my ass and go do a CrossFit class
25:18
Play from here
Copy citation
or something because my biology is gonna fall apart.
25:23
Play from here
Copy citation
The way we're wired,
25:24
Play from here
Copy citation
both from a mental health perspective and physically,
25:27
Play from here
Copy citation
like you have to compensate.
25:29
Play from here
Copy citation
And I think we're in this weird era right now.
25:31
Play from here
Copy citation
We're gonna see more and more of AI just coming
25:34
Play from here
Copy citation
into people's both personal and professional lives.
25:36
Play from here
Copy citation
And there's gonna have to be some sort of a counterbalance
25:39
Play from here
Copy citation
on the other side for the things that it's taking away
25:41
Play from here
Copy citation
from that human experience.
25:42
Play from here
Copy citation
And I wonder, as we talk about AI manifestos
25:46
Play from here
Copy citation
and change management and direction from leadership,
25:50
Play from here
Copy citation
like where that is going to fit in.
25:52
Play from here
Copy citation
Like I really feel like there has to start
25:53
Play from here
Copy citation
to be some education about this.
25:55
Play from here
Copy citation
And maybe we can look at history as some powerful lessons
25:58
Play from here
Copy citation
that we don't wanna repeat and have things like a,
26:01
Play from here
Copy citation
like we had these insane national obesity epidemics
26:04
Play from here
Copy citation
or diabetes or just all these things that came
26:07
Play from here
Copy citation
from the result of like,
26:08
Play from here
Copy citation
oh, now we're not using our bodies.
26:10
Play from here
Copy citation
CrossFit, Weight Watchers, right?
26:12
Play from here
Copy citation
Maybe we can get ahead of the curve this time.
26:14
Play from here
Copy citation
And I'm wondering if that's starting to come up.
26:15
Play from here
Copy citation
So it is.
26:17
Play from here
Copy citation
The term that's coming up in HR circles is called pacing.
26:20
Play from here
Copy citation
And it's how do you communicate the appropriate cadence
26:25
Play from here
Copy citation
for outputs?
26:26
Play from here
Copy citation
Not, you don't wanna ever stifle people's creativity.
26:29
Play from here
Copy citation
And if you're an engineer
26:31
Play from here
Copy citation
and you've just been uncorked to be able to like build
26:34
Play from here
Copy citation
everything, it's super exciting.
26:36
Play from here
Copy citation
And you're the guy who's running around with your thumb
26:38
Play from here
Copy citation
and your MacBook at lunch and you're using a dispatch
26:42
Play from here
Copy citation
or you're using like, you know,
26:44
Play from here
Copy citation
one of the remote control agents to like
26:46
Play from here
Copy citation
keep building things.
26:48
Play from here
Copy citation
But the organization actually can't deal with that
26:51
Play from here
Copy citation
because there are plenty of people in orbit around you
26:54
Play from here
Copy citation
that inherit that flow and figuring out how that flow,
26:58
Play from here
Copy citation
even if individually you're creating burdens on yourself,
27:02
Play from here
Copy citation
like the first tier is the, whoa, whoa, whoa, whoa, whoa.
27:05
Play from here
Copy citation
Okay there, scooter, like slow down.
27:08
Play from here
Copy citation
The fire hose is not doing great for the rest of the team.
27:12
Play from here
Copy citation
So how do we pace that in a way that it makes sense
27:18
Play from here
Copy citation
versus our overall planning?
27:19
Play from here
Copy citation
And again, you don't wanna stifle innovation, right?
27:22
Play from here
Copy citation
So we talk a lot about new capabilities with AI
27:26
Play from here
Copy citation
and that's where a lot of this stuff comes from.
27:27
Play from here
Copy citation
It's like, oh, the engineer has a wild idea
27:30
Play from here
Copy citation
to create this new feature.
27:31
Play from here
Copy citation
Well, that wasn't in the roadmap anywhere
27:33
Play from here
Copy citation
and you don't wanna lose it
27:35
Play from here
Copy citation
because it is indeed a cool feature,
27:37
Play from here
Copy citation
but you can't just take the roadmap every single day
27:40
Play from here
Copy citation
or every Monday because the engineer spent all weekend
27:43
Play from here
Copy citation
like vibe coding and just like crumple up the roadmap
27:46
Play from here
Copy citation
and toss it again.
27:47
Play from here
Copy citation
There's discipline and there's attention to detail
27:50
Play from here
Copy citation
and there's quality control, code checks.
27:53
Play from here
Copy citation
Like there's a lot of like adulting
27:55
Play from here
Copy citation
that still has to be done in all of this.
27:57
Play from here
Copy citation
Yeah, totally.
27:58
Play from here
Copy citation
I mean, I was thinking back too
27:59
Play from here
Copy citation
to when there was such an emphasis on,
28:01
Play from here
Copy citation
and I feel like this is definitely people still talk about
28:04
Play from here
Copy citation
and there's still workshops, but it's a little passe
28:05
Play from here
Copy citation
like this idea of company culture, right?
28:07
Play from here
Copy citation
And these human centric companies
28:09
Play from here
Copy citation
where people are connecting with each other
28:11
Play from here
Copy citation
and the teams that are closest to each other
28:13
Play from here
Copy citation
are the most productive and it's better for the company.
28:15
Play from here
Copy citation
And so there's this real focus on that
28:17
Play from here
Copy citation
and now it's AI, AI, AI, right?
28:20
Play from here
Copy citation
And I feel like we might have a renaissance of this
28:22
Play from here
Copy citation
kind of human connection and company culture talk
28:25
Play from here
Copy citation
because, I mean, listen, right now I'm building
28:27
Play from here
Copy citation
a mental health practice and objectively not a tech company
28:31
Play from here
Copy citation
but very tech enabled, still very tech forward.
28:33
Play from here
Copy citation
I spent 25 years in tech.
28:34
Play from here
Copy citation
And when I start thinking about how, I don't know,
28:40
Play from here
Copy citation
like I think it was a few episodes ago, I said,
28:42
Play from here
Copy citation
wow, I'm having dreams, like a lot of dreams
28:45
Play from here
Copy citation
of cloud right now.
28:46
Play from here
Copy citation
Like A, because I'm using it so much all day, every day.
28:51
Play from here
Copy citation
But then I had this thought of,
28:52
Play from here
Copy citation
well, now with dictation, right?
28:54
Play from here
Copy citation
I don't use my keyboard anymore.
28:57
Play from here
Copy citation
I'm just talking all day and I'm talking to this thing.
29:00
Play from here
Copy citation
And my brain probably thinks because I'm talking,
29:03
Play from here
Copy citation
I am like communicating with this other life
29:06
Play from here
Copy citation
or this other entity.
29:08
Play from here
Copy citation
And so now it's infiltrating my dreams
29:10
Play from here
Copy citation
and it just keeps happening more and more
29:12
Play from here
Copy citation
like it really does.
29:13
Play from here
Copy citation
And so when I started to really zoom out
29:16
Play from here
Copy citation
and think about this,
29:17
Play from here
Copy citation
even though I'm not in an AI centric business,
29:19
Play from here
Copy citation
I spend way more time talking to and working with AI than I do other people right now.
29:26
Play from here
Copy citation
And is that the future of work, right?
29:28
Play from here
Copy citation
Is that the future of work?
29:29
Play from here
Copy citation
It's like more and more people do this.
29:31
Play from here
Copy citation
Is it going to be people in their little silos managing their suite of AI agents as opposed
29:37
Play from here
Copy citation
to all of that time before being spent connecting with other humans?
29:41
Play from here
Copy citation
So yeah, I've got a little bit of a theory about this, man.
29:46
Play from here
Copy citation
You know, okay, okay.
29:48
Play from here
Copy citation
We debated between this episode about the Pope's encyclical.
29:52
Play from here
Copy citation
Yeah.
29:53
Play from here
Copy citation
Yeah.
29:54
Play from here
Copy citation
And like, okay, we're not that show.
29:57
Play from here
Copy citation
But I think it's really interesting from like a closing slide perspective that, you know,
30:04
Play from here
Copy citation
the human centricity of how humans work is really, really vital.
30:11
Play from here
Copy citation
And, you know, people are casting shade that the Pope didn't go too far or that he had
30:15
Play from here
Copy citation
anthropic there in the room or whatever, or he's out of his lane.
30:19
Play from here
Copy citation
It's like his lane is humanity.
30:21
Play from here
Copy citation
And this is actually Dario wearing a mask, wearing the map, but like another piece of
30:30
Play from here
Copy citation
data that was floating around.
30:32
Play from here
Copy citation
I think I saw it in the times over the weekend that are even yesterday that it was somebody's
30:37
Play from here
Copy citation
somebody analyzed 375,000 college essays and since the rise of GPT.
30:43
Play from here
Copy citation
And as you can imagine, the essays have definitely been infiltrated by AI.
30:50
Play from here
Copy citation
And what's happened is there's actually only one eighth, one eighth, the number of original
30:57
Play from here
Copy citation
ideas that are being put forward in those essays.
31:00
Play from here
Copy citation
And I think it's a super interesting data point that to us really makes sense because
31:05
Play from here
Copy citation
what are the models, right?
31:06
Play from here
Copy citation
The models are trained on existing information that's out there and they are designed to
31:10
Play from here
Copy citation
basically find the best median piece of information that comes out the other direction.
31:15
Play from here
Copy citation
And students pushing themselves to come up with novel ideas to differentiate themselves
31:20
Play from here
Copy citation
from the peer of their peers, that's going away because people are leaning too much into
31:24
Play from here
Copy citation
the models.
31:25
Play from here
Copy citation
So beyond losing like just the human communication layer, you're just losing the spark of innovation
31:33
Play from here
Copy citation
that the models of tomorrow need the innovation of today so they can subsume it and kill those
31:38
Play from here
Copy citation
jobs too.
31:39
Play from here
Copy citation
That's like a different question, but our job as humans, our job as participants in
31:45
Play from here
Copy citation
an organization are to come up with those new and novel ideas.
31:49
Play from here
Copy citation
So watch yourself if you're spending, I think it's different if you're like vibe coding
31:54
Play from here
Copy citation
your way through a project and you're trying to solve soluble problems.
31:57
Play from here
Copy citation
Like these are solved problems that you are stitching together.
32:02
Play from here
Copy citation
You're not adding new and different things.
32:05
Play from here
Copy citation
But if you are in that innovation room, if you are in that ideation room, yeah, okay,
32:11
Play from here
Copy citation
you can do some prep with the AI, but it's your synthesis that's going to add the new
32:16
Play from here
Copy citation
idea on top of it.
32:18
Play from here
Copy citation
That's going to be valuable tomorrow.
32:19
Play from here
Copy citation
Yeah, I think, yeah, I think you're absolutely right.
32:22
Play from here
Copy citation
That's super interesting.
32:23
Play from here
Copy citation
And I think all of these skills will atrophy quickly when you just stop using them completely,
32:30
Play from here
Copy citation
you know, and we're seeing that in education too, right?
32:32
Play from here
Copy citation
That's a big topic in education.
32:36
Play from here
Copy citation
Because these things aren't, aren't, aren't just augmenting the things that we're actually
32:39
Play from here
Copy citation
practicing inside of education, but replacing them.
32:42
Play from here
Copy citation
Like if, if that's really the state of where all these different essays are right now,
32:47
Play from here
Copy citation
and they're all just being generated by LLMs, people will lose their ability to add anything
32:53
Play from here
Copy citation
new to the conversation.
32:54
Play from here
Copy citation
Like it will atrophy.
32:55
Play from here
Copy citation
It'll just start to disappear over time.
32:56
Play from here
Copy citation
Okay.
32:57
Play from here
Copy citation
But this is important because there are things that it's okay if they atrophy.
33:00
Play from here
Copy citation
It is okay if it goes away, like, you know, the, the ability to use a slide rule.
33:06
Play from here
Copy citation
I sort of, I have my, my grandfather's slide rule that is an artifact in my office, zero
33:11
Play from here
Copy citation
idea how to use my thing.
33:12
Play from here
Copy citation
Yeah.
33:13
Play from here
Copy citation
It doesn't matter.
33:14
Play from here
Copy citation
Absolutely.
33:15
Play from here
Copy citation
It doesn't matter.
33:16
Play from here
Copy citation
Like from an artisanal perspective, kind of cool.
33:17
Play from here
Copy citation
Like same thing with like a sextant, like, you know, you can choose to do these things,
33:22
Play from here
Copy citation
but like rote coding of like headers and these, or, or, or summarizing, summarizing documents
33:30
Play from here
Copy citation
for the sake of summarizing them because you asked for it and you're not interested
33:33
Play from here
Copy citation
in it.
33:34
Play from here
Copy citation
Or like, like an insanely messy and deep inbox that it'll go and just pick out the important
33:39
Play from here
Copy citation
things you'd need to address.
33:41
Play from here
Copy citation
Great use cases, like absolutely great use cases.
33:45
Play from here
Copy citation
But use, okay, here's an interesting one.
33:47
Play from here
Copy citation
So there's another piece of medical research.
33:49
Play from here
Copy citation
Again, this is fast and loose with the data because we're not a medical show and we can
33:52
Play from here
Copy citation
do that.
33:53
Play from here
Copy citation
Yeah.
33:54
Play from here
Copy citation
So they, they, they, someone had physicians in like an ENT specialty.
34:00
Play from here
Copy citation
So ear, nose and throat.
34:01
Play from here
Copy citation
I think it was, maybe I'm wrong.
34:03
Play from here
Copy citation
Use medical assisted interpretation of symptoms for only like six weeks.
34:09
Play from here
Copy citation
And they tested their ability to discern certain symptoms before and after.
34:15
Play from here
Copy citation
And they had a 3% degradation in their ability to recall those particular issues.
34:21
Play from here
Copy citation
So you got to ask yourself, okay, that is core to what this physician does is an ear,
34:26
Play from here
Copy citation
nose and throat specialist that that is what they do.
34:29
Play from here
Copy citation
But is it, if the tool is actually better and faster, like, I, I don't know, is that
34:36
Play from here
Copy citation
a trade that's worth making?
34:38
Play from here
Copy citation
I would say no for somebody who's junior and it's probably okay for somebody who's senior.
34:42
Play from here
Copy citation
Yeah.
34:43
Play from here
Copy citation
But I mean, if you look at it through the lens of their cognitive ability, right?
34:46
Play from here
Copy citation
Like this is, this is what they're doing.
34:48
Play from here
Copy citation
And back to kind of the, the, the human element of this and like losing those parts of your
34:52
Play from here
Copy citation
biology because they're being replaced.
34:54
Play from here
Copy citation
It's like the whole Google maps debate, right?
34:56
Play from here
Copy citation
For, you know, hundreds of thousands of years, we had to know how to get places, right?
35:00
Play from here
Copy citation
We have to navigate around and there's parts of your brain that are really good at doing
35:05
Play from here
Copy citation
that because that's how we survived as a species.
35:08
Play from here
Copy citation
It's one of the things that we needed to survive.
35:10
Play from here
Copy citation
It's like, this is, this is how I relate to things in my space and get around them and
35:14
Play from here
Copy citation
memorize routes and find my way.
35:16
Play from here
Copy citation
And this is safe and this isn't right.
35:18
Play from here
Copy citation
I will turn on Google maps for anything now.
35:20
Play from here
Copy citation
And I think they did a series of tests and kind of looked at cognitive prowess and people's
35:26
Play from here
Copy citation
ability to remember where to go and how to get places in those places that the brain
35:30
Play from here
Copy citation
just kind of like had turned off over time when they stopped having to figure it out
35:34
Play from here
Copy citation
for themselves or learn it once and then repeat it and instead just look at the little blue
35:38
Play from here
Copy citation
dot on the screen and just kind of blindly follow that.
35:41
Play from here
Copy citation
Right.
35:42
Play from here
Copy citation
So I think that's where it'll get interesting is if we start replacing those things that
35:46
Play from here
Copy citation
are really core to the human experience and not just augmenting them, but replacing them.
35:51
Play from here
Copy citation
Like I will say that Google maps has replaced my need to ever figure out how to get somewhere.
35:56
Play from here
Copy citation
And it doesn't matter if I'm in a car or mountain bike hiking, like I'm just looking at a map.
36:00
Play from here
Copy citation
I'm looking at it.
36:01
Play from here
Copy citation
I'm looking at a little GPS thing on my watch or my phone or on my car screen.
36:05
Play from here
Copy citation
And so some of that I think has to be considered and I, and you know, what you're, what you're
36:11
Play from here
Copy citation
talking about too with some of these other points I think are super, super valid.
36:16
Play from here
Copy citation
But if you go a layer below that, right, if you, if you really look at kind of the
36:22
Play from here
Copy citation
first principles of this stuff and you go back 300,000 years in time, this connection
36:29
Play from here
Copy citation
with each other, like people talking to people and forming these bonds, that meant survival,
36:34
Play from here
Copy citation
right?
36:35
Play from here
Copy citation
Like the lonely human got eaten.
36:37
Play from here
Copy citation
There was no lone wolf.
36:38
Play from here
Copy citation
The connected person got fed, protected, passed on their genes.
36:42
Play from here
Copy citation
So your nervous system actually learned to treat belonging as a matter of life and death
36:46
Play from here
Copy citation
because it was, right?
36:48
Play from here
Copy citation
And so if any of these chats can mimic the conversation part of it, it still can't mimic
36:53
Play from here
Copy citation
the tribe part that kept you alive.
36:56
Play from here
Copy citation
And so your nervous system doesn't know it's 2026.
36:58
Play from here
Copy citation
It's still running software from the Savannah, you know, isolation was a death sentence and
37:03
Play from here
Copy citation
connection was the whole strategy for survival.
37:05
Play from here
Copy citation
And so, you know, that's why loneliness hurts.
37:07
Play from here
Copy citation
Like that's why the surgeon general declared it a national epidemic that was more harmful
37:13
Play from here
Copy citation
than smoking a pack of cigarettes a day, right?
37:15
Play from here
Copy citation
It's an alarm.
37:16
Play from here
Copy citation
So I think the fundamental problem might be that the chat can quiet the alarm, but it's
37:20
Play from here
Copy citation
never delivering the actual safety that the alarm was built to find, right?
37:25
Play from here
Copy citation
So I don't know.
37:26
Play from here
Copy citation
I've been thinking a lot about that part as AI infiltrates my dreams, you know, as I spend
37:31
Play from here
Copy citation
more time in isolation talking to it, that we didn't survive because we were strong.
37:36
Play from here
Copy citation
We survived because we banded together.
37:38
Play from here
Copy citation
We co-regulated.
37:39
Play from here
Copy citation
We shared risk.
37:40
Play from here
Copy citation
AI is, would register our distress and, you know, would move towards it.
37:45
Play from here
Copy citation
And, you know, AI can simulate being seen, but your body knows the difference.
37:50
Play from here
Copy citation
Like at a deep biological level, it knows the difference between a mirror and like a
37:54
Play from here
Copy citation
real witness.
37:55
Play from here
Copy citation
And only one of those actually helps you thrive.
37:57
Play from here
Copy citation
So I think that there are just going to be more and more impacts.
38:02
Play from here
Copy citation
Like when I talk to these people, right, that are just so immersed in this stuff and they're
38:07
Play from here
Copy citation
talking to AI way more than humans, as I fell into for many, many months, if not years
38:13
Play from here
Copy citation
now.
38:14
Play from here
Copy citation
I'm wondering about the secondary effects that that's having and when that's going to
38:17
Play from here
Copy citation
be recognized as a problem in the workplace as well.
38:20
Play from here
Copy citation
Okay.
38:21
Play from here
Copy citation
So oversharing just a bit, but my wife, I had to have a bit of an intervention with
38:25
Play from here
Copy citation
her this weekend over her use of chatty.
38:29
Play from here
Copy citation
She calls chat GPT.
38:30
Play from here
Copy citation
Okay.
38:31
Play from here
Copy citation
And, you know, one of our, we have a high school junior who's like looking for schools and
38:36
Play from here
Copy citation
I could just feel that she'd fallen into like a self-reinforcing loop.
38:41
Play from here
Copy citation
And I had to be like, Hey, okay, you need to, you need to back out of the loop that
38:45
Play from here
Copy citation
you're in and you need to ask it.
38:47
Play from here
Copy citation
These got gut check type questions to recenter and, you know, highly skilled, highly educated,
38:54
Play from here
Copy citation
but super easy to see how that spiral happens.
38:58
Play from here
Copy citation
So to bring us a little bit back on track here as AI at work, what are the things that
39:04
Play from here
Copy citation
we should be solving with AI?
39:06
Play from here
Copy citation
I think that a lot of the last few minutes has been about that, that maintaining that,
39:11
Play from here
Copy citation
that human intuition and that human differentiation, that kind of human connection, that human
39:17
Play from here
Copy citation
connection, don't let it replace human connection.
39:20
Play from here
Copy citation
Don't let it replace human connection because we're wired for it and it's, it's gonna mess
39:24
Play from here
Copy citation
you up and it'll mess up your corporate culture as well.
39:27
Play from here
Copy citation
If you just like throw everybody into these isolated silos with, with agents, just think
39:32
Play from here
Copy citation
about it right now.
39:33
Play from here
Copy citation
I guarantee it.
39:34
Play from here
Copy citation
If you're in whatever market and you have 10 competitors in that market and everybody
39:39
Play from here
Copy citation
is just sort of using AI, like it's all kind of tossing out the same ideas.
39:44
Play from here
Copy citation
At some point you don't deserve to exist because you're not doing anything that's more differentiated
39:49
Play from here
Copy citation
than your neighbor.
39:50
Play from here
Copy citation
And that's what's always driven business forward.
39:52
Play from here
Copy citation
So I would say, you know, things not to automate or some of those, those innovation pieces
39:58
Play from here
Copy citation
doesn't mean you can't amplify them, but how about things that, that, that should be applied
40:03
Play from here
Copy citation
to AI?
40:04
Play from here
Copy citation
I'm gonna back sort of the slide rule, but like, what are the things that if you were
40:08
Play from here
Copy citation
looking at an organization, would you immediately pinpoint?
40:11
Play from here
Copy citation
Me personally?
40:12
Play from here
Copy citation
Yeah.
40:13
Play from here
Copy citation
You personally, because this is what I, this is basically what I do.
40:16
Play from here
Copy citation
Yeah.
40:17
Play from here
Copy citation
I'm curious on your take.
40:18
Play from here
Copy citation
Yeah.
40:19
Play from here
Copy citation
I would say anything that is really like a mundane pushing around of pixels, I would
40:25
Play from here
Copy citation
replace as soon as possible.
40:27
Play from here
Copy citation
And I think that the, the trap that I fall into is, and I think I've been speaking to
40:32
Play from here
Copy citation
a lot of people falling into this trap.
40:34
Play from here
Copy citation
Like when we talk about like, what are those use cases?
40:36
Play from here
Copy citation
I would say that because there are so many features dropping so fast and because these
40:41
Play from here
Copy citation
models are playing leapfrog with each other so quickly and the iterations happen so fast
40:45
Play from here
Copy citation
and releases happen so fast that very often with the people I've been talking to, AI has
40:50
Play from here
Copy citation
become this solution that's looking for a problem as opposed to the question you're
40:54
Play from here
Copy citation
asking of what are the specific use cases where you can drop it in and you know, have
40:59
Play from here
Copy citation
it get, have it, have it take away the things that are keeping you from doing the real work
41:05
Play from here
Copy citation
that not only matters, but that you love to do, that you enjoy doing.
41:10
Play from here
Copy citation
Right.
41:11
Play from here
Copy citation
I hate email.
41:12
Play from here
Copy citation
I've hated email since email started.
41:14
Play from here
Copy citation
Like, it's just, it's this burden to keep up on.
41:17
Play from here
Copy citation
The only thing I hate more than email is real time chats, you know, great.
41:21
Play from here
Copy citation
Now we have Slack and, but I, but the email didn't go away and I've got Slack too, right?
41:25
Play from here
Copy citation
My calendar.
41:26
Play from here
Copy citation
I've got my calendar too.
41:27
Play from here
Copy citation
I've, I've, I've got to keep, oh boy, when is this appointment and oh, when can you make
41:31
Play from here
Copy citation
time?
41:32
Play from here
Copy citation
You know, things like tools like Calendly, right?
41:34
Play from here
Copy citation
Now there's a million different Calendly's kind of took away that whole process we'd
41:38
Play from here
Copy citation
have to go through of what, when are you free?
41:40
Play from here
Copy citation
What, when's Mike free?
41:41
Play from here
Copy citation
When's Bob free?
41:42
Play from here
Copy citation
And it, it kind of solved that part of it, right?
41:45
Play from here
Copy citation
Less time staring at the screen, just figure out something mundane that felt ridiculous.
41:49
Play from here
Copy citation
My to-do list, right?
41:52
Play from here
Copy citation
The decision fatigue of what is most important?
41:55
Play from here
Copy citation
How do I tag this?
41:56
Play from here
Copy citation
How do I organize it?
41:58
Play from here
Copy citation
I don't know.
41:59
Play from here
Copy citation
I don't know.
42:00
Play from here
Copy citation
Like probably most people don't have like over a hundred things on their to-do list.
42:03
Play from here
Copy citation
I'm guessing you do Kevin, because we're probably wired similarly, but that that's exhausting
42:08
Play from here
Copy citation
too.
42:09
Play from here
Copy citation
So all of those things I have really just handed over to AI, to what we're calling our
42:14
Play from here
Copy citation
chief of staff or a mono thread or whatever you want to call it, but basically a singular
42:19
Play from here
Copy citation
chat that's connected to all these different end points so that I can just go to one place
42:24
Play from here
Copy citation
and make decisions in a single chat that would have ordinarily, you know, I would have had
42:28
Play from here
Copy citation
to just round robin a dozen different applications to figure this all out.
42:32
Play from here
Copy citation
So that I think, anything ironically enough, whatever is happening on the screen that I
42:37
Play from here
Copy citation
can leverage to have me spend less time on the screen in aggregate, I think is a good
42:41
Play from here
Copy citation
use case.
42:42
Play from here
Copy citation
The part I was just emphasizing is where it gets dangerous is you can just keep running
42:45
Play from here
Copy citation
that direction and it really does turn into, now I have a bunch of solutions and I'm looking
42:49
Play from here
Copy citation
for problems, right?
42:51
Play from here
Copy citation
And those problems often aren't worth fixing, or you're almost inventing new problems to
42:56
Play from here
Copy citation
fix because you have a solution.
42:58
Play from here
Copy citation
And that's where the AI work density trap tends to fall in.
43:01
Play from here
Copy citation
That's where the FOMO falls in.
43:03
Play from here
Copy citation
That's where you start dropping off human connection because you're spending all your
43:06
Play from here
Copy citation
time doing this as I have been for many months now, and I'm trying to wean off of it and
43:10
Play from here
Copy citation
really surgically look at this stuff of what is the most important thing to solve here?
43:14
Play from here
Copy citation
That's going to make my work and my life better, right?
43:16
Play from here
Copy citation
Interesting.
43:18
Play from here
Copy citation
I got one more quick one for you that I love that was like, drove me absolutely crazy was
43:24
Play from here
Copy citation
I would be on Instagram wasting my life trying to just find some recipes.
43:28
Play from here
Copy citation
Like this is the family thing every Sunday.
43:31
Play from here
Copy citation
What are we going to eat this week?
43:32
Play from here
Copy citation
We got to plan our meals for all the kids.
43:35
Play from here
Copy citation
It's such a headache.
43:36
Play from here
Copy citation
So I'd go to Instagram and I'd save recipes.
43:38
Play from here
Copy citation
But then once I have the recipes, it's like ingredients in the captions.
43:41
Play from here
Copy citation
Well, goddammit Meta, you can't copy and paste captions.
43:44
Play from here
Copy citation
But there's this free web-based tool that will extract the captions from Instagram.
43:48
Play from here
Copy citation
Okay, so now I got to use web-based Instagram, which is horrible.
43:51
Play from here
Copy citation
And I paste a link in and I do this.
43:52
Play from here
Copy citation
I'm like, you know what?
43:53
Play from here
Copy citation
It's going to build an agent to do this all for me, right?
43:56
Play from here
Copy citation
So it would every week just kind of go to my Instagram feed and it would.
43:59
Play from here
Copy citation
all those recipes had, would use the tool to extract all the captions and then build
44:03
Play from here
Copy citation
out a recipe listing, click up, and then just give me my goddamn meal plan every Sunday
44:07
Play from here
Copy citation
night.
44:08
Play from here
Copy citation
Right?
44:09
Play from here
Copy citation
Tons of time saved, you know, decision fatigue, all that stuff.
44:13
Play from here
Copy citation
I can now got these cool high protein meals that are delicious and it's got all the steps
44:17
Play from here
Copy citation
and recipes in a print friendly version.
44:19
Play from here
Copy citation
Amazing.
44:20
Play from here
Copy citation
Amazing.
44:21
Play from here
Copy citation
Right?
44:22
Play from here
Copy citation
Less time on the computer, more time just eating with my family.
44:24
Play from here
Copy citation
So, okay.
44:25
Play from here
Copy citation
This is, I feel like you sort of just played right into my, not trap, but we didn't script
44:30
Play from here
Copy citation
this.
44:31
Play from here
Copy citation
We didn't script this at all.
44:33
Play from here
Copy citation
But like that is the most pedestrian stuff ever that you just described.
44:37
Play from here
Copy citation
You are not, even as a CEO, even as somebody who was leading companies, like you weren't
44:42
Play from here
Copy citation
necessarily spending all of this, this, this efforts on reinventing your company and not
44:48
Play from here
Copy citation
that you weren't thinking about it and not that you weren't noodling forward on it and
44:51
Play from here
Copy citation
not that you didn't have people that are doing it, but most organizations are sort
44:56
Play from here
Copy citation
of missing this part.
44:58
Play from here
Copy citation
So now go to an organization that has a thousand people and it's like you go to that division
45:03
Play from here
Copy citation
head or whatever and they'll have, they'll pontificate on like whatever, you know, new
45:08
Play from here
Copy citation
thing that they want to do to increase sales or whatever it is, but they actually haven't
45:13
Play from here
Copy citation
enabled their people to deal with all of this other pedestrian stuff because they haven't
45:19
Play from here
Copy citation
isolated like a platform or an approach or have a policy or whatever it is such that
45:26
Play from here
Copy citation
everybody in the organization can start taking these things off one at a time.
45:30
Play from here
Copy citation
And it doesn't feel like much to be like, Oh, well, we're going to save a couple hours
45:35
Play from here
Copy citation
on you know, whatever that meeting followup is until you multiply it by a thousand people.
45:44
Play from here
Copy citation
And this also plays into, Oh my goodness, I'm actually going to get to one of my other
45:47
Play from here
Copy citation
talking points that I didn't think so is as far as having a standard, like set of tools
45:53
Play from here
Copy citation
that you go to, you can't necessarily just have people solving all of their problems
45:57
Play from here
Copy citation
in all of these different ways, particularly when you start talking about agents and vibe
46:02
Play from here
Copy citation
coding and whatever.
46:03
Play from here
Copy citation
Especially if they're organization problems, not individual problems, right?
46:06
Play from here
Copy citation
Yeah.
46:07
Play from here
Copy citation
You gotta be able to share this stuff.
46:08
Play from here
Copy citation
So, so having like a core set of technologies that all roads kind of lead back to is really,
46:15
Play from here
Copy citation
really important and not enough organizations are doing that in big organizations.
46:19
Play from here
Copy citation
You might have things like Workday and sort of like we guys like us can sneer at things
46:23
Play from here
Copy citation
like that, but it's already integrated all over the place, but people still don't know
46:27
Play from here
Copy citation
how to use it to amplify whatever they're doing.
46:30
Play from here
Copy citation
But smaller organizations leaning into a platform all the way back to the beginning of the conversation,
46:36
Play from here
Copy citation
understanding what it can and can't do, giving people access to the right tools to make these
46:42
Play from here
Copy citation
things happen and then encouraging them to go out and solve some of their own problems.
46:48
Play from here
Copy citation
That's, that's, that is where a lot of the, the immediate gains are.
46:52
Play from here
Copy citation
And I know that's not like super popular, particularly at enterprise.
46:56
Play from here
Copy citation
Well, you know, individual efficiency isn't really worth investing in.
47:00
Play from here
Copy citation
Well, times thousands of people it is.
47:02
Play from here
Copy citation
And what are those people, where's that intellect, where is that humanity going to be focused
47:07
Play from here
Copy citation
when they're not focused on TPS reports?
47:09
Play from here
Copy citation
Totally, totally.
47:10
Play from here
Copy citation
I mean, that kind of goes back to the whole, like for a very long time, it was all about
47:13
Play from here
Copy citation
productivity and individual wellness wasn't worth investing in.
47:17
Play from here
Copy citation
And then they realized, oh, wait, all of a sudden, yes, it is.
47:21
Play from here
Copy citation
Because mentally, while people are happy, people are more productive, people who connect
47:24
Play from here
Copy citation
with other people are more productive.
47:26
Play from here
Copy citation
And people who waste three hours a day just bouncing in and out of their inbox and trying
47:31
Play from here
Copy citation
to sync that with their calendar or their to-do list also are much less productive.
47:37
Play from here
Copy citation
And so, yeah, there's no, I think to your point, Kevin, there's, there's no universal
47:41
Play from here
Copy citation
playbook for this stuff of like, here's how you do it.
47:44
Play from here
Copy citation
This is what everybody should be doing.
47:46
Play from here
Copy citation
It really does, at least at this time right now, come down to, well, okay, what are the
47:50
Play from here
Copy citation
tools that you have available?
47:52
Play from here
Copy citation
And what have, what are other people doing?
47:54
Play from here
Copy citation
You know, how can we replicate that across an organization so that everybody shares in
47:58
Play from here
Copy citation
the benefit of these few outlier individuals who have kind of found this Uber hack to get
48:03
Play from here
Copy citation
back three hours of their day?
48:05
Play from here
Copy citation
I would say that part of it is tracking and acknowledgement that it doesn't really exist
48:11
Play from here
Copy citation
right now in, in, in a structured way in most organizations.
48:15
Play from here
Copy citation
So we do this pretty cool thing in our scaled trainings where every interaction we have,
48:21
Play from here
Copy citation
every lecture we do, we're always recording with permission, of course, but we're taking
48:25
Play from here
Copy citation
all of those different ideas that people have and we're putting them into a sandbox.
48:30
Play from here
Copy citation
And then as people complete projects, because that's sort of the goal of the program is
48:34
Play from here
Copy citation
that you, you complete a practical project at the end.
48:39
Play from here
Copy citation
That goes into an organizational database that the tech team can see and they can prioritize
48:45
Play from here
Copy citation
based on, you know, ROI or whatever it may be, but you don't lose it.
48:51
Play from here
Copy citation
If you don't capture it, you'll lose it.
48:54
Play from here
Copy citation
And, and that's, it's, it's true of any meeting that we have, we, why do we have meetings?
48:58
Play from here
Copy citation
We have meetings to exchange information and we sit there with our notepads and we scribble
49:02
Play from here
Copy citation
things down.
49:03
Play from here
Copy citation
Okay.
49:04
Play from here
Copy citation
Now we're transcribing meetings.
49:05
Play from here
Copy citation
Great.
49:06
Play from here
Copy citation
So now we can capture more of that, but are we capturing all of the grist there?
49:10
Play from here
Copy citation
Are we capturing all of the opportunities?
49:12
Play from here
Copy citation
In fact, I, I can't really, I can think of one that I work with that's doing a good job
49:18
Play from here
Copy citation
of actually capturing all of that.
49:20
Play from here
Copy citation
Isn't it funny that this kind of used to be the promise of CRMs, right?
49:24
Play from here
Copy citation
But the problem was that nobody took the time to get the data into the CRM.
49:29
Play from here
Copy citation
So the CRMs were just this shell of missed data opportunities, at least in, I mean, I
49:34
Play from here
Copy citation
worked in half a dozen corporate organizations in the sales part of the business units where,
49:39
Play from here
Copy citation
you know, it'd be like, Oh, this client came back and they're interested in X, like, let's
49:43
Play from here
Copy citation
go into the notes.
49:44
Play from here
Copy citation
Oh, there are no notes.
49:45
Play from here
Copy citation
Like, well, did anybody ever call them?
49:47
Play from here
Copy citation
I don't know.
49:48
Play from here
Copy citation
Cause all the call logs are empty.
49:49
Play from here
Copy citation
Right?
49:50
Play from here
Copy citation
Like, it was just like this empty, empty feed inside of Salesforce or half a dozen other
49:55
Play from here
Copy citation
CRMs I worked in.
49:57
Play from here
Copy citation
So I think what I'm hearing a lot about there right now is even if you aren't really sure
50:02
Play from here
Copy citation
of what technology you're going to settle on, how are you going to use that technology,
50:05
Play from here
Copy citation
how it's going to connect all the pieces, there really should be an early emphasis on
50:09
Play from here
Copy citation
how do you capture the conversations, the data points, the ideas, and start to build
50:14
Play from here
Copy citation
this out so that when that time does come, you can start to connect all those dots and
50:18
Play from here
Copy citation
you aren't starting from scratch.
50:19
Play from here
Copy citation
You can make sense of the things that you've already captured, right?
50:22
Play from here
Copy citation
And this is everything from the AI note takers to the dictation tools that we use all day
50:27
Play from here
Copy citation
on and on.
50:28
Play from here
Copy citation
Right.
50:29
Play from here
Copy citation
Just how do you, how do you start to at least get these into one place so that, because
50:33
Play from here
Copy citation
I'm not really convinced, Kevin, that there is, it sounds like, you know, you implemented
50:36
Play from here
Copy citation
a pretty cool workflow where these things get presented back to people.
50:39
Play from here
Copy citation
I'm seeing right now in that middle part of it, like tons and tons of data coming in and
50:44
Play from here
Copy citation
then lots of noise being outputted.
50:47
Play from here
Copy citation
And then nobody really knowing what to do with all that noise.
50:50
Play from here
Copy citation
Like who's going to take the time to read it or really implement the good ideas, put
50:54
Play from here
Copy citation
out the bad ideas, share the ones across different business units that are really going
50:58
Play from here
Copy citation
to move the needle.
50:59
Play from here
Copy citation
Like, I feel like we're in the frontier, the kind of pioneer days of, of, of that part
51:04
Play from here
Copy citation
of it.
51:05
Play from here
Copy citation
But as far as like getting to the point where you can at least start to collect all these
51:09
Play from here
Copy citation
things that have historically been lost, you can now kind of do that in a passive way instead
51:13
Play from here
Copy citation
of like actively typing notes into Salesforce.
51:15
Play from here
Copy citation
Right.
51:16
Play from here
Copy citation
It's not that easy though.
51:17
Play from here
Copy citation
So I think we have 10 different projects going on right now with different clients that are
51:21
Play from here
Copy citation
something along this line, this like, Oh, we have data over here and we have data here
51:25
Play from here
Copy citation
and there's transcripts.
51:26
Play from here
Copy citation
And then what are you going to do with it?
51:28
Play from here
Copy citation
And like, like as much as we all want to just toss it in a giant spaghetti bowl and say,
51:32
Play from here
Copy citation
pick at it.
51:33
Play from here
Copy citation
Well, that's really expensive.
51:35
Play from here
Copy citation
You can actually do that, but it's really, really expensive.
51:39
Play from here
Copy citation
So being able to gather the data from the right places of the organization, convert
51:44
Play from here
Copy citation
the data to something that's usable through, you know, using vector structures.
51:48
Play from here
Copy citation
So rag and yeah, structures like that.
51:51
Play from here
Copy citation
And then figuring out how you make it assailable to people.
51:55
Play from here
Copy citation
This is a huge, huge, huge deal.
51:57
Play from here
Copy citation
And I think it's that last part that's the biggest challenge, to be honest, because those
52:01
Play from here
Copy citation
other parts are solved.
52:02
Play from here
Copy citation
I mean, they may not be easy or economical right now, but a lot of them are, they're,
52:07
Play from here
Copy citation
they're solved and they're evolving, but there's something there.
52:10
Play from here
Copy citation
It's that getting it back in front of people, like, what do you do with all of that?
52:14
Play from here
Copy citation
How do you then use it to drive strategic decisions in the organization?
52:19
Play from here
Copy citation
That part is tough.
52:20
Play from here
Copy citation
Like that part is tough.
52:21
Play from here
Copy citation
So there was a, I listened to the All In podcast and they had Marc Benioff of Salesforce.
52:27
Play from here
Copy citation
I don't listen to it that often.
52:28
Play from here
Copy citation
Actually, somebody told me I should listen to it because I've been going off about headless
52:31
Play from here
Copy citation
software and they let him go on a little bit about, you know, the future of Salesforce.
52:37
Play from here
Copy citation
And what I took away from it was the future of Salesforce is Slack.
52:41
Play from here
Copy citation
So Salesforce owns Slack and Slack is this core of all of these company, of, of, of company
52:49
Play from here
Copy citation
communication.
52:51
Play from here
Copy citation
And it's starting to proactively surface things in really, really interesting ways.
52:56
Play from here
Copy citation
So not in a way that you or I have to go in and like figure out all the logic and like
53:01
Play from here
Copy citation
pull out the prompts and figure out what the data structures are.
53:04
Play from here
Copy citation
It's already baked into Slack.
53:06
Play from here
Copy citation
So it's a pretty gangster move by Salesforce, which moves them very rapidly out of the CRM,
53:14
Play from here
Copy citation
the sales business and into like the enterprise spine business.
53:17
Play from here
Copy citation
And I think I just convinced myself that I should buy some Salesforce.
53:20
Play from here
Copy citation
It's like, wait, it's like they're really well positioned.
53:23
Play from here
Copy citation
And although I don't love Salesforce, like they're really, really well positioned as
53:27
Play from here
Copy citation
far as having the right channels of data communication.
53:31
Play from here
Copy citation
And it's just a matter of connecting a few more into the harnesses.
53:34
Play from here
Copy citation
I think for mere mortal companies, like where their CRM is sort of an adjacency, it's not
53:40
Play from here
Copy citation
going to be quite the same.
53:42
Play from here
Copy citation
And you're going to see those companies leaning more into the frontier models.
53:45
Play from here
Copy citation
So open AI, Anthropic, Gemini, et cetera.
53:50
Play from here
Copy citation
And these things will just start surfacing in a way that right now it's like brute force,
53:56
Play from here
Copy citation
expensive, complicated type effort.
53:59
Play from here
Copy citation
And you know, in a year it might be, you just open your Gemini and it's already crawled
54:03
Play from here
Copy citation
the things that it thinks it needs to crawl.
54:04
Play from here
Copy citation
Yeah, I mean, yeah.
54:06
Play from here
Copy citation
So for those of you that haven't just completely lost right now, think about like, think about
54:13
Play from here
Copy citation
like that that day when you opened up Gmail and at the top of your inbox said, hey, you
54:18
Play from here
Copy citation
sent this email to this dude six days ago and he never replied.
54:21
Play from here
Copy citation
Do you want to follow up?
54:23
Play from here
Copy citation
Right. Like there was this little intelligence on the back end there that knew you sent the
54:27
Play from here
Copy citation
email, it knew you never got a reply.
54:29
Play from here
Copy citation
And so it brought it back to your attention without you asking.
54:32
Play from here
Copy citation
And it's like, those are the capabilities that we're starting to see in Slack right now
54:35
Play from here
Copy citation
where it's connecting dots on the back end, because it is now under the Salesforce suite.
54:40
Play from here
Copy citation
It has access to a bunch more data points.
54:42
Play from here
Copy citation
And that becomes a useful thing that's operating in the place that people are already used
54:46
Play from here
Copy citation
to going to have these communications.
54:48
Play from here
Copy citation
So that is, yeah, yeah.
54:51
Play from here
Copy citation
Back to like what's really useful, right?
54:53
Play from here
Copy citation
Like that's, that's very useful.
54:55
Play from here
Copy citation
And we're seeing just the beginning time of it.
54:57
Play from here
Copy citation
Kevin, I've got a few more minutes, man.
54:59
Play from here
Copy citation
I don't know if you really want to press on keeping this right at an hour, which I'm fine
55:03
Play from here
Copy citation
too, but if you want to talk about, if you want to talk about some of these cloud skills,
55:08
Play from here
Copy citation
I did have a couple of thoughts on it that we could wrap in 10 minutes and I'm happy to
55:11
Play from here
Copy citation
do it. It's up to you.
55:12
Play from here
Copy citation
I think we should do that next week.
55:14
Play from here
Copy citation
We could do that.
55:14
Play from here
Copy citation
Oh, next week.
55:15
Play from here
Copy citation
Okay.
55:15
Play from here
Copy citation
We could do cloud skills next week.
55:17
Play from here
Copy citation
The other thing.
55:18
Play from here
Copy citation
You got thoughts.
55:18
Play from here
Copy citation
I got thoughts on cloud skills.
55:19
Play from here
Copy citation
You got thoughts.
55:21
Play from here
Copy citation
Yeah.
55:21
Play from here
Copy citation
There's also the tech stack.
55:23
Play from here
Copy citation
These are really important pieces that are building blocks.
55:26
Play from here
Copy citation
And there is so much that's just going on in so many directions.
55:30
Play from here
Copy citation
It is, it is super exciting, but it is starting to mature.
55:34
Play from here
Copy citation
Like I'm feeling it.
55:35
Play from here
Copy citation
It's, it's, it's starting to mature and it's, it's becoming much more accessible to people day by.
55:42
Play from here
Copy citation
Just don't let it make you lonely, Kevin.
55:44
Play from here
Copy citation
Yeah.
55:44
Play from here
Copy citation
Don't let it make you lonely.
55:45
Play from here
Copy citation
Lose my wife to chatty.
55:49
Play from here
Copy citation
Talk to you humans.
55:51
Play from here
Copy citation
All right, man.
55:52
Play from here
Copy citation
All right, bud.
55:52
Play from here
Copy citation
Yeah.
55:58
Play from here
Copy citation
Thanks for listening.
55:58
Play from here
Copy citation
If you'd like to send us a question to answer on the show, email me at Elijah at spark six.com.
56:04
Play from here
Copy citation
That is S P a R K.
56:06
Play from here
Copy citation
The number six.com or Kevin at Kevin at ascend labs.ai.
56:11
Play from here
Copy citation
To learn more about AI strategy and training for your business.
56:14
Play from here
Copy citation
Visit Kevin's company ascend AI labs at ascend labs.ai.
56:20
Play from here
Copy citation
And to learn more about design and development of custom AI solutions, visit my agency at
56:25
Play from here
Copy citation
spark six.com.
56:27
Play from here
Copy citation
You can also sign up for my free newsletter, AI framework Friday by visiting spark six.com
56:33
Play from here
Copy citation
forward slash newsletter.