WEBVTT

NOTE
This file was generated by Descript 

00:00:05.474 --> 00:00:09.794
Hey folks, and welcome to
the small tech podcast.

00:00:09.824 --> 00:00:13.154
I'm your host Raph from Ephemere Creative.

00:00:13.154 --> 00:00:18.164
And today we are going to
be talking about AI tools.

00:00:18.664 --> 00:00:24.604
So, again, this is going to be a bit
more of a loosey goosey kind of episode,

00:00:24.604 --> 00:00:29.374
because, you know, everyone has just
been talking about these AI tools and.

00:00:30.154 --> 00:00:31.654
They're pretty neat.

00:00:32.134 --> 00:00:33.964
And I think.

00:00:34.444 --> 00:00:39.424
In the context of building small
tech, there are a lot of little

00:00:39.424 --> 00:00:42.754
like productivity gains that
you can get out of some of them.

00:00:43.114 --> 00:00:45.754
I don't think they'll
make sense for everyone.

00:00:46.144 --> 00:00:47.644
And I don't think.

00:00:48.634 --> 00:00:51.604
I have lots of thoughts about
whether or not this is, this is

00:00:51.604 --> 00:00:54.034
a big game-changing thing or not.

00:00:54.534 --> 00:01:00.924
Fundamentally, I do think this is a new
way for people to interact with machines,

00:01:01.194 --> 00:01:04.794
which I think is pretty game-changing.

00:01:05.454 --> 00:01:11.514
Um, specifically thinking about tools
like chat GPT and, and those types

00:01:11.544 --> 00:01:14.874
of, uh, large language models and the.

00:01:15.444 --> 00:01:18.264
The chat driven interface.

00:01:18.654 --> 00:01:20.334
Uh, for interacting with them.

00:01:20.834 --> 00:01:25.424
In terms of how all of that stuff
fits into building small tech.

00:01:26.414 --> 00:01:33.014
I found that there's a couple
ways in which AI tools fit into

00:01:33.044 --> 00:01:35.984
my workflow, our workflow at EDC.

00:01:36.044 --> 00:01:36.584
And.

00:01:37.454 --> 00:01:40.424
Just generally things that, that, that.

00:01:40.924 --> 00:01:43.774
Are involved in building small tech.

00:01:43.834 --> 00:01:47.044
So the first one I'm
going to talk about is.

00:01:47.544 --> 00:01:48.054
Get hub.

00:01:48.084 --> 00:01:48.774
Co-pilot.

00:01:49.464 --> 00:01:52.794
The next is chat GPT itself.

00:01:53.294 --> 00:01:54.044
And.

00:01:54.544 --> 00:01:57.154
We'll touch on the AI.

00:01:57.154 --> 00:02:05.664
Um, Like the image generators that you can
find in, uh, in, in being, and, uh, and.

00:02:06.164 --> 00:02:07.334
What am I even talking about?

00:02:07.394 --> 00:02:13.994
Uh, like the, the stable diffusion, uh,
mid journey, Dolly, all of those things.

00:02:14.834 --> 00:02:16.454
I find those are less relevant.

00:02:16.784 --> 00:02:21.134
Um, But maybe in like
how they fit into your.

00:02:21.634 --> 00:02:23.074
Potentially into a workflow.

00:02:23.074 --> 00:02:26.224
And then, um, finally there's.

00:02:27.214 --> 00:02:29.014
Uh, what was the last one?

00:02:29.044 --> 00:02:31.204
Oh, well like being, being searched.

00:02:31.204 --> 00:02:36.754
So Bing has an image generator, uh, but,
uh, Bing search and another one that is,

00:02:36.814 --> 00:02:40.414
I assume, built off of, uh, open AI eyes.

00:02:40.504 --> 00:02:42.874
Uh, Chad G well, GBT four.

00:02:43.624 --> 00:02:44.224
Or GPT.

00:02:44.284 --> 00:02:46.234
3.5 or whatever.

00:02:46.474 --> 00:02:49.684
Um, it's a tool called perplexity,
which I find kind of interesting.

00:02:50.224 --> 00:02:54.244
Um, and I guess we can talk a little
bit about Canva and what they've been

00:02:54.244 --> 00:03:01.714
doing, uh, with, uh, with these large
language models and the image generation.

00:03:02.344 --> 00:03:03.364
Uh, tools.

00:03:03.864 --> 00:03:07.194
So, yeah, little loosey goosey.

00:03:07.854 --> 00:03:09.624
Uh, but let's start with.

00:03:09.684 --> 00:03:10.674
Get hub, copilot.

00:03:11.424 --> 00:03:13.614
Uh, and there are other tools like this.

00:03:13.614 --> 00:03:18.954
I think, uh, Microsoft has their own,
there are a few others out there.

00:03:19.254 --> 00:03:21.204
But essentially tools
that help you write code.

00:03:22.104 --> 00:03:24.174
And there was some controversy.

00:03:24.714 --> 00:03:27.234
Actually with all of these tools
that are controversy, right?

00:03:27.294 --> 00:03:33.954
Uh, With the large language models,
there is controversy surrounding truth.

00:03:33.984 --> 00:03:34.944
What is it?

00:03:35.064 --> 00:03:38.784
And how is this going
to flood the zone with.

00:03:39.144 --> 00:03:40.764
Things that are hard to.

00:03:41.264 --> 00:03:43.094
Determine whether they're true or not.

00:03:43.094 --> 00:03:46.544
And whether people are going to
do the homework required of them.

00:03:46.964 --> 00:03:53.054
To look into things that, that are
generated by these, by these models, with

00:03:53.084 --> 00:03:59.204
the image generators, there's controversy
around IP assignment and who gets to.

00:03:59.564 --> 00:04:04.424
Own the images that are generated,
but also should these models be

00:04:04.424 --> 00:04:09.734
trained on, uh, on images from
artists who have not given permission?

00:04:10.574 --> 00:04:12.914
There's there's all kinds of stuff there.

00:04:13.484 --> 00:04:17.984
With get hub and copilot in particular,
there were people who were talking

00:04:18.014 --> 00:04:21.824
about how, you know, they were
seeing snippets of code that looked a

00:04:21.824 --> 00:04:25.784
whole lot like snippets of code that
were right out of their code bases.

00:04:26.414 --> 00:04:29.384
Um, and that is definitely not ideal.

00:04:29.534 --> 00:04:31.514
Um, that's, that's not great.

00:04:32.054 --> 00:04:37.304
Um, I think some of the examples I saw
were pretty short snippets of code.

00:04:37.814 --> 00:04:40.454
And I do think that if you're given.

00:04:41.354 --> 00:04:46.064
A certain context and a certain
language there with, with

00:04:46.094 --> 00:04:47.744
shorter pieces of code, right.

00:04:47.744 --> 00:04:50.204
If something can be done in five lines.

00:04:50.774 --> 00:04:52.304
Or even 10 lines.

00:04:53.264 --> 00:04:58.094
There are definitely situations
where there's not going to be many

00:04:58.094 --> 00:05:00.134
ways to, to put that together.

00:05:00.164 --> 00:05:02.624
So that's, that's a,
that's an awkward one.

00:05:03.194 --> 00:05:07.424
Um, I think anything longer
is, you know, not ideal.

00:05:07.964 --> 00:05:11.804
Uh, and there should be protections
built into these systems for that.

00:05:12.194 --> 00:05:16.634
That being said, I have
been finding that get hub.

00:05:16.664 --> 00:05:18.974
Copilot is very helpful for me.

00:05:19.474 --> 00:05:25.384
For the most part, I actually would
say it doesn't do a whole lot in

00:05:25.384 --> 00:05:28.924
terms of generating longer functions.

00:05:29.854 --> 00:05:34.354
I just find that the productivity
improvement I can get from it, just

00:05:34.354 --> 00:05:36.574
generating one or two lines at a time.

00:05:37.174 --> 00:05:45.124
So if I've got a variable and I've got, or
let's say I've got a list, Of something.

00:05:45.604 --> 00:05:49.924
And I let's say, okay, I'll take an
example from our project that you restack.

00:05:50.764 --> 00:05:54.004
And there's a bunch of components
as part of a chewy stock project.

00:05:54.034 --> 00:05:54.484
And.

00:05:54.994 --> 00:05:58.714
When we're building the dependency
graph for these different

00:05:58.714 --> 00:06:00.094
components and how they interact.

00:06:00.394 --> 00:06:03.934
There's a point in the process
where I want to just get the

00:06:03.934 --> 00:06:05.524
infrastructure components.

00:06:06.244 --> 00:06:10.384
And so if I already have a function where
I've said, here are the components, I've

00:06:10.384 --> 00:06:12.124
loaded them using this other function.

00:06:12.664 --> 00:06:14.194
And I start typing.

00:06:14.914 --> 00:06:18.454
Uh, the name of a variable
called infrastructure components.

00:06:18.934 --> 00:06:20.104
I find that.

00:06:20.494 --> 00:06:24.544
Um, co-pilot will write
the right code to you.

00:06:24.544 --> 00:06:28.234
Usually just a single line or maybe
three lines to say, all right, let's loop

00:06:28.264 --> 00:06:32.584
through all of these components and let's
filter by type equals infrastructure.

00:06:33.084 --> 00:06:34.914
And it's, it's the sort of thing.

00:06:34.914 --> 00:06:37.314
That's just a really, really small.

00:06:37.974 --> 00:06:38.574
Thing.

00:06:39.564 --> 00:06:44.964
But when you add that up to like three
seconds or five seconds that it saves me.

00:06:45.504 --> 00:06:46.854
When I'm typing that out.

00:06:47.634 --> 00:06:51.864
If I do that, I don't know, a
hundred times and a project.

00:06:52.254 --> 00:06:53.424
Uh, in a session.

00:06:54.324 --> 00:06:57.984
Um, That actually adds up
to quite a lot of time.

00:06:57.984 --> 00:06:59.844
500 seconds is a lot.

00:06:59.904 --> 00:07:03.744
And I do find that it's every, you
know, every few seconds it'll, it'll

00:07:03.744 --> 00:07:05.124
autocomplete, something like that.

00:07:05.784 --> 00:07:07.914
Um, so I think there's
a lot of value there.

00:07:08.094 --> 00:07:08.694
Um,

00:07:09.194 --> 00:07:12.224
The next one that I want to talk about.

00:07:12.254 --> 00:07:15.644
So that's code and how you can just
build products a little bit faster

00:07:15.644 --> 00:07:17.954
using AI tools, like get hub copilot.

00:07:18.254 --> 00:07:20.564
I think one of the other
ones out there is a.

00:07:21.064 --> 00:07:23.164
Microsoft in Telecode.

00:07:23.524 --> 00:07:24.634
I believe.

00:07:25.054 --> 00:07:30.064
Um, they have had a product named
IntelliSense for a long time, which

00:07:30.064 --> 00:07:32.344
is just their normal code completion.

00:07:32.854 --> 00:07:40.084
Uh, but in tele code, I think is this,
uh, uh, predictive, predictive model,

00:07:40.174 --> 00:07:45.214
um, So, yeah, I haven't tried it, but
it might be neat to give it a shot.

00:07:45.784 --> 00:07:49.654
The next thing I want to talk
about is Chad GBT itself.

00:07:50.314 --> 00:07:55.384
Um, I have found actually what I
will likely do with this, uh, podcast

00:07:55.384 --> 00:07:59.314
episode when I'm done is I'm going to
take the transcript of the episode.

00:07:59.644 --> 00:08:01.594
I'm going to go to Chad GPT.

00:08:01.804 --> 00:08:05.284
I now pay for GPT for, um, and.

00:08:05.944 --> 00:08:10.444
What I will do is I will tell
Chad GBT, here's the transcript.

00:08:11.044 --> 00:08:15.814
Can you clean this up into a blog,
post maintaining a bit of the tone.

00:08:16.324 --> 00:08:21.814
And structure, but turn it into something
a bit more clean, few paragraphs.

00:08:22.414 --> 00:08:26.434
And I will use that as a base to
sort of expand on, on what I'm

00:08:26.434 --> 00:08:30.094
talking about now and turn it
into a, into a, into a blog post.

00:08:30.904 --> 00:08:32.914
And I think there's a lot of value there.

00:08:33.244 --> 00:08:39.424
In terms of using these large language
models, not to generate anything novel.

00:08:39.454 --> 00:08:41.644
I don't think that is their role.

00:08:42.394 --> 00:08:47.284
But to take language that you
have, or that exists out in the

00:08:47.284 --> 00:08:49.444
world and doing something with it.

00:08:50.014 --> 00:08:58.384
So if I feed in a longer chunk of text and
I want to get something out of it, I think

00:08:58.384 --> 00:09:01.114
that works out generally pretty decently.

00:09:01.114 --> 00:09:02.584
You still have to double-check.

00:09:03.094 --> 00:09:07.804
Uh, so in this case, the transcription
and I want to get a blog post.

00:09:08.374 --> 00:09:13.324
Um, I find it works well,
not for summarizing, but for

00:09:13.324 --> 00:09:16.204
reorganizing and cleaning things up.

00:09:16.834 --> 00:09:20.944
Um, so with, with the
transcription it'll, it'll.

00:09:21.444 --> 00:09:22.314
I can tell it.

00:09:22.314 --> 00:09:25.074
This is a transcription
with transcription errors.

00:09:25.764 --> 00:09:27.534
And basically it'll clean up that text.

00:09:28.104 --> 00:09:32.724
Um, and it will format it to some
extent, uh, the way that I might

00:09:32.724 --> 00:09:35.424
direct it to, uh, which is practical.

00:09:36.114 --> 00:09:39.264
It usually takes a few tries.

00:09:40.104 --> 00:09:44.484
But I think it's still a lot faster than
me just writing the whole thing from

00:09:44.484 --> 00:09:48.984
scratch, especially since I've already
spoken it out in this podcast episode.

00:09:49.484 --> 00:09:50.864
So that was number two.

00:09:50.984 --> 00:09:52.034
Uh, Chad GBT.

00:09:52.244 --> 00:09:55.664
Uh, I find that the 20 bucks a
month is worth it for me, uh,

00:09:55.724 --> 00:09:58.904
especially for this kind of thing,
turning a podcast into a blog post.

00:09:59.414 --> 00:10:03.824
Or I've had a few other
situations, um, that I find useful.

00:10:04.244 --> 00:10:08.474
Oh, actually as a developer myself,
one of the things that I find useful as

00:10:08.474 --> 00:10:18.404
well is using it to, uh, remember the
format of a, uh, command in the terminal.

00:10:18.854 --> 00:10:26.144
So if I want to use a tool like our
sync, which will, uh, sync files

00:10:26.144 --> 00:10:29.624
from one directory to another,
or from one server to another.

00:10:29.924 --> 00:10:34.394
And I can't quite remember the flags, for
example, to make it do certain things.

00:10:34.574 --> 00:10:39.854
For those of you who aren't, uh,
super technical, um, It will, uh,

00:10:39.884 --> 00:10:41.774
basically spit out a command for me.

00:10:42.494 --> 00:10:47.654
That will do the things I needed to, and
I can give it sort of a natural language.

00:10:47.954 --> 00:10:54.484
So I will say, Hey, what's the arson
command to, uh, Sync this directory

00:10:54.484 --> 00:10:59.794
with, uh, this directory on our
remote server with this IP address.

00:11:00.364 --> 00:11:06.874
And I want to, uh, make sure that
everything is archived and small.

00:11:07.384 --> 00:11:09.454
And I don't know, something like that.

00:11:09.454 --> 00:11:11.914
And it'll spit out the right,
the right, uh, command.

00:11:12.484 --> 00:11:12.904
So.

00:11:13.804 --> 00:11:19.444
That's another neat use of Chad GPT and
there's a bunch of small ways to do that.

00:11:19.444 --> 00:11:23.194
I find it's good for, for
code and terminal commands.

00:11:23.344 --> 00:11:27.604
Um, if you just need to spit out
little, little chunks that are useful.

00:11:28.444 --> 00:11:31.414
The next one, the image generation stuff.

00:11:31.924 --> 00:11:33.784
I have had almost zero use for this.

00:11:33.784 --> 00:11:35.824
I think it's more of like a curiosity.

00:11:36.154 --> 00:11:41.074
Um, I've heard of people who use
those tools like a stapled, a fusion,

00:11:41.494 --> 00:11:46.954
uh, and, uh, Dolly and mid journey
to, uh, generate content, to put

00:11:46.954 --> 00:11:48.454
into their marketing materials.

00:11:48.664 --> 00:11:52.834
I feel like our brand doesn't really
match with that and I still feel.

00:11:53.644 --> 00:11:57.834
A bit more weird about the, uh, The art.

00:11:58.614 --> 00:12:02.934
I don't know, there's, there's a lot
of stuff going on with artists, uh,

00:12:02.964 --> 00:12:05.394
who are very unhappy about it and like,

00:12:05.894 --> 00:12:08.234
I kind of see where they're coming from.

00:12:08.264 --> 00:12:08.804
It's weird.

00:12:08.804 --> 00:12:14.504
Cause I don't like I put a lot of
code up on GitHub, open source and.

00:12:15.434 --> 00:12:20.624
From my context, doing that, I'm
perfectly happy for these models to

00:12:20.624 --> 00:12:22.934
be trained on my code and my writing.

00:12:22.964 --> 00:12:26.594
I have a fair bit of writing out there on
the internet, and that doesn't bother me.

00:12:27.374 --> 00:12:31.094
Um, I have some art out there online.

00:12:31.594 --> 00:12:35.884
And frankly, it doesn't bother
me that it's trained on my art.

00:12:36.064 --> 00:12:39.964
But for someone who's living
is made off of their art.

00:12:40.464 --> 00:12:45.234
Uh, and who is perhaps invested a lot
more time into refining their craft.

00:12:45.774 --> 00:12:47.514
I can understand where
they're coming from.

00:12:47.694 --> 00:12:48.744
Uh, so.

00:12:49.644 --> 00:12:51.114
Yeah, anyway, that's that?

00:12:51.354 --> 00:12:56.214
I do know some people who do get
a fair bit of value out of, uh,

00:12:56.214 --> 00:13:00.204
the image generation tools for
marketing materials, as you know,

00:13:00.264 --> 00:13:02.964
the header of their newsletter or
something like that, they want.

00:13:03.564 --> 00:13:08.274
Uh, they're talking about something that
will speed up your website and they want.

00:13:08.784 --> 00:13:10.404
Uh, computer running.

00:13:11.244 --> 00:13:15.654
With shoes on fire or something like
that, something silly and the AI

00:13:15.654 --> 00:13:17.124
spits it out and it's kind of fun.

00:13:17.364 --> 00:13:20.214
Um, and yeah, it doesn't
need to be perfect.

00:13:20.784 --> 00:13:24.684
It's not something that they
would've paid someone for anyways.

00:13:25.104 --> 00:13:29.664
So it's just kind of a neat, neat,
uh, additional value for them.

00:13:30.294 --> 00:13:32.724
Uh, so that's.

00:13:33.414 --> 00:13:34.704
Yeah.

00:13:34.704 --> 00:13:35.424
That's that?

00:13:36.084 --> 00:13:40.974
And the, oh, actually on the subject
of the image generation tools.

00:13:41.844 --> 00:13:47.364
If you haven't used Canva before,
I think Canva is an amazing tool.

00:13:47.394 --> 00:13:50.004
I feel like I may have talked
about it in a previous episode.

00:13:50.514 --> 00:13:56.334
Um, if you want to do fairly simple
designs for social media posts.

00:13:56.784 --> 00:14:03.084
Uh, presentations, documents, uh,
short videos, that sort of thing.

00:14:03.114 --> 00:14:06.414
And you want access to a bunch
of royalty free content built

00:14:06.414 --> 00:14:07.944
directly into your editor.

00:14:08.364 --> 00:14:13.764
And you want to have your branding
materials available, like fonts, colors.

00:14:13.854 --> 00:14:18.084
Uh, graphics, other things
that, that use to make your.

00:14:18.954 --> 00:14:21.774
Uh, collateral recognizable as your own.

00:14:22.274 --> 00:14:23.774
Uh, Canva is great for that.

00:14:24.164 --> 00:14:28.244
And one of the things that they added
recently is a bunch of tools built on

00:14:28.484 --> 00:14:34.274
both large language models and these,
uh, image generation, uh, tools.

00:14:34.904 --> 00:14:38.174
And so now within Canva, you can.

00:14:38.774 --> 00:14:40.004
Generate an image.

00:14:40.334 --> 00:14:43.844
Uh, that you might want to use
as like the background for,

00:14:43.904 --> 00:14:45.974
uh, for a social media post.

00:14:46.394 --> 00:14:51.374
Uh, and so you could tell it, Hey,
I want a wavy green fields with

00:14:51.734 --> 00:14:54.404
a Hawk flying through the sky.

00:14:54.644 --> 00:14:56.054
Or something like that.

00:14:56.414 --> 00:14:58.394
And, uh, yeah.

00:14:58.424 --> 00:15:02.204
Have it generate that for your background
and throw in some texts and whatever.

00:15:02.714 --> 00:15:04.394
Um, so that's kind of neat.

00:15:04.454 --> 00:15:05.654
I've tried it once.

00:15:05.954 --> 00:15:09.704
It's a little slow as all of these
things are and the quality is.

00:15:09.884 --> 00:15:10.484
It's okay.

00:15:11.054 --> 00:15:15.694
Um, They also added in some,
uh, tools to generate text.

00:15:16.234 --> 00:15:18.574
Uh, directly in there.

00:15:19.024 --> 00:15:22.984
I find that a little awkward again,
for me, the primary value that

00:15:23.014 --> 00:15:26.254
comes out of these large language
models is not in generating text.

00:15:26.284 --> 00:15:28.744
It's in restructuring.

00:15:29.164 --> 00:15:31.414
Texts that I have generated.

00:15:31.744 --> 00:15:36.004
Um, again, I don't think the
large language models are.

00:15:36.454 --> 00:15:39.784
They don't create anything truly novel.

00:15:40.144 --> 00:15:43.654
Um, Well, that's not exactly true.

00:15:43.834 --> 00:15:46.114
They don't come up with novel ideas.

00:15:46.354 --> 00:15:50.344
So they'll come up with novel
texts, but you still have to give.

00:15:50.974 --> 00:15:52.624
Them something to work with.

00:15:52.684 --> 00:15:56.884
And so if you're trying to
generate text for a, uh, for

00:15:56.884 --> 00:15:59.014
marketing material, I don't know.

00:15:59.014 --> 00:16:03.484
I feel like you have to have something
pretty good in there to start with.

00:16:04.054 --> 00:16:07.654
And it just feels like a
weird, weird place for me.

00:16:08.164 --> 00:16:10.834
Um, to, to be doing that in Canva.

00:16:11.334 --> 00:16:13.584
But yeah, it is, it is neat.

00:16:13.644 --> 00:16:17.544
They have this, uh, this new,
uh, tool that I haven't tried

00:16:17.544 --> 00:16:21.414
yet where you're supposed to be
able to describe a presentation.

00:16:21.954 --> 00:16:24.234
And it will spit out the presentation.

00:16:24.654 --> 00:16:28.914
Um, with your brand or,
uh, and, and your content.

00:16:28.944 --> 00:16:33.054
So I am curious to give that a shot,
especially if you can pump in enough

00:16:33.084 --> 00:16:36.564
content to say, Hey, I want to do a
presentation on all of these things.

00:16:36.564 --> 00:16:37.464
Here's like my.

00:16:37.944 --> 00:16:41.094
Intro the stuff that I want to
go over and like a conclusion.

00:16:41.514 --> 00:16:44.574
Can you put that together in
a nice cohesive presentation?

00:16:44.574 --> 00:16:45.864
For me, that would be kind of neat.

00:16:46.554 --> 00:16:49.674
Um, So, yeah, we'll see if
that's actually any good.

00:16:50.174 --> 00:16:56.594
Um, and the last thing I wanted to talk
about is the search, uh, and sort of

00:16:56.744 --> 00:17:00.884
finding information in sort of a coherent.

00:17:01.184 --> 00:17:02.984
Natural language format.

00:17:03.764 --> 00:17:04.364
And.

00:17:04.864 --> 00:17:08.464
It's weird because I don't
like my instinct was that this

00:17:08.464 --> 00:17:11.554
is not something that large
language models would be good at.

00:17:12.004 --> 00:17:14.944
And that I wouldn't find much value in it.

00:17:15.814 --> 00:17:20.164
But, uh, I have been sort of proven wrong.

00:17:20.554 --> 00:17:22.834
Uh, to some extent, uh, especially.

00:17:23.554 --> 00:17:27.484
Framing this again, in the context of
building a small, small tech product.

00:17:27.934 --> 00:17:28.384
Um,

00:17:28.884 --> 00:17:29.784
I think.

00:17:30.624 --> 00:17:34.944
If you're operating business, there is
some value in being able to quickly parse.

00:17:35.424 --> 00:17:38.694
Uh, websites information, whether
you're looking for grants.

00:17:38.784 --> 00:17:44.394
Uh, you, uh, are trying to find people
to connect with, uh, potential clients.

00:17:44.424 --> 00:17:44.934
I don't know.

00:17:45.264 --> 00:17:49.434
Uh, users for your app, and you're
trying to figure out how, how

00:17:49.464 --> 00:17:51.474
all of your, your things connect.

00:17:52.434 --> 00:17:54.504
Perplexity, I think has been kind of neat.

00:17:55.164 --> 00:18:00.684
Uh, it's a tool that is structured
a bit like a search engine, but

00:18:00.684 --> 00:18:03.984
it provides a bit more context.

00:18:04.674 --> 00:18:11.664
Um, so it's very much the same value
prop as, uh, Bing's a new chat interface.

00:18:11.964 --> 00:18:16.524
Which is, you can ask it a
question in natural language and

00:18:16.524 --> 00:18:19.314
it will do some searching for you.

00:18:19.434 --> 00:18:23.874
Uh, some more traditional search
pull out pages that contain the

00:18:23.874 --> 00:18:27.174
information that you want and sort
of summarize and spit that out.

00:18:27.684 --> 00:18:28.824
In a.

00:18:29.334 --> 00:18:31.524
Uh, natural language format.

00:18:32.024 --> 00:18:35.414
Now, one of the things I find
really annoying about being is

00:18:35.474 --> 00:18:39.704
you have to use Microsoft edge
browser, which I do not want to use.

00:18:39.854 --> 00:18:42.134
Uh, so I will not be using being.

00:18:42.634 --> 00:18:46.864
Which is I think a weird move
on Microsoft's part because

00:18:46.864 --> 00:18:50.284
I could absolutely see myself
using, being more often.

00:18:50.854 --> 00:18:55.684
If it weren't tied to the edge
browser, but I use arc, which is a

00:18:55.684 --> 00:18:57.814
wonderful chromium based browser.

00:18:58.384 --> 00:19:01.894
And I don't want to move away from it.

00:19:02.044 --> 00:19:02.704
It's lovely.

00:19:02.824 --> 00:19:04.024
So no being for me.

00:19:04.414 --> 00:19:07.054
But perplexity gives me
something very, very similar.

00:19:07.084 --> 00:19:13.324
It will pull out, uh, websites
that it uses as, uh, sources.

00:19:13.654 --> 00:19:18.034
And so it'll tell you basically where the
information came from, that it summarizes.

00:19:18.694 --> 00:19:21.544
But the other thing that I really
like about it is it's got a Chrome

00:19:21.574 --> 00:19:26.884
extension that you can use to sort of
summarize or ask questions about a page.

00:19:27.094 --> 00:19:29.164
So if you're browsing.

00:19:29.614 --> 00:19:36.184
And you come across, uh, for example,
a very long grant application document.

00:19:36.754 --> 00:19:39.064
You can ask it things about that document.

00:19:39.064 --> 00:19:43.174
You can ask it to summarize and,
uh, figure out what's what's

00:19:43.174 --> 00:19:44.464
in there that might be notable.

00:19:44.914 --> 00:19:46.534
Um, and that's something that.

00:19:46.894 --> 00:19:49.114
I find valuable, even if it's.

00:19:49.774 --> 00:19:50.704
Infrequently.

00:19:51.634 --> 00:19:55.054
The value that I get out of it is, is.

00:19:55.504 --> 00:19:56.674
I'd say significant.

00:19:57.304 --> 00:19:58.114
Um, yeah.

00:19:58.114 --> 00:20:02.014
So those are the AI tools that I've
used that I've poked around that.

00:20:02.584 --> 00:20:06.094
I would love to hear more about
what, what you all are using.

00:20:06.544 --> 00:20:12.394
Uh, it's, uh, it's it feels like quite
the explosion of AI tools at the moment.

00:20:12.634 --> 00:20:16.774
I guess it's been going for a little
while, but especially since chat GBT.

00:20:17.284 --> 00:20:19.324
Uh, came, came on the scene.

00:20:19.594 --> 00:20:21.394
Uh, things have.

00:20:21.894 --> 00:20:23.514
I'd say very much.

00:20:24.354 --> 00:20:28.254
Exploded and I'm so, so
curious to see where they go.

00:20:28.464 --> 00:20:29.574
I think there's a.

00:20:30.534 --> 00:20:31.674
There's a hype cycle.

00:20:32.064 --> 00:20:38.904
That happens with any new sort of
technology as it, it captures the

00:20:38.934 --> 00:20:42.744
public interest, uh, particularly
the consumer public interest.

00:20:43.244 --> 00:20:49.994
But there's something about these
AI tools that strikes me as a little

00:20:49.994 --> 00:20:54.404
different than some of the hype that
we've seen over the past five years or so.

00:20:55.034 --> 00:20:58.274
A lot of the technologies that I've
seen hyped over the past five years.

00:20:58.604 --> 00:21:03.074
I couldn't find any
directive value in them.

00:21:03.854 --> 00:21:09.434
Uh, so whether that's a crypto
V R a R, I know some people got

00:21:09.434 --> 00:21:10.814
a lot of value out of those.

00:21:11.314 --> 00:21:17.314
To me, I'm seeing a lot more people
get value out of the new AI tools.

00:21:17.974 --> 00:21:18.904
Very quickly.

00:21:19.294 --> 00:21:20.224
Including myself.

00:21:20.254 --> 00:21:25.354
And maybe that's just my own
context, but there's something

00:21:25.354 --> 00:21:27.964
there that I find very exciting.

00:21:28.684 --> 00:21:34.054
Um, And so I'm curious to see where
it, where it lands when the hype

00:21:34.054 --> 00:21:38.524
cycle dies down and we're left
with the actual evaluable stuff.

00:21:38.824 --> 00:21:40.264
So, yeah.

00:21:40.764 --> 00:21:44.784
Anyway, if you want to keep up with
this stuff about building small

00:21:44.784 --> 00:21:47.814
tech, then make sure to subscribe.

00:21:48.264 --> 00:21:52.434
If you're checking this out on YouTube,
make sure to hit that like button.

00:21:52.934 --> 00:21:56.534
I have been your host wrath, and we
all want to do some good in the world.

00:21:56.564 --> 00:21:59.414
So go out there and build something.

00:21:59.444 --> 00:22:00.674
Good friends.