Cashing in on AI: Streamlining Bank Operations

WEBVTT

1
00:00:02.879 --> 00:00:07.320
You are listening to Leaders in Lending
from Upstart, a podcast dedicated to helping

2
00:00:07.360 --> 00:00:13.000
consumer lenders grow their programs and improve
their product offerings. Each week, here

3
00:00:13.039 --> 00:00:17.559
decision makers in the finance industry offer
insights into the future of the lending industry,

4
00:00:18.039 --> 00:00:22.800
best practices around digital transformation, and
more. Let's get into the show.

5
00:00:26.440 --> 00:00:29.920
Welcome everyone to another episode of Leaders
in Lending. I'm met snow here

6
00:00:30.039 --> 00:00:33.679
with Will Robinson from and Capture.
Will Welcome to the show. Hey,

7
00:00:33.679 --> 00:00:37.719
Matt, thanks for having me on. Absolutely Maybe let's get started by a

8
00:00:37.759 --> 00:00:41.759
little introduction, talk about what you
do and Capture what they do, and

9
00:00:41.799 --> 00:00:45.920
how you got to the role of
CEO there today. Yeah, it's kind

10
00:00:45.920 --> 00:00:49.880
of a wild journey. But in
Capture we are a software company in the

11
00:00:49.920 --> 00:00:56.439
intelligent document processing space, so that
means we have artificial intelligence that can read

12
00:00:56.479 --> 00:01:00.560
documents and understand what type of document
is looking at and then can go find

13
00:01:00.600 --> 00:01:07.680
important data that in those documents and
pull it out automatically. We work in

14
00:01:07.840 --> 00:01:12.000
a bunch of different use cases across
a bank, mostly around the kind of

15
00:01:12.079 --> 00:01:21.920
loan origination and processing space for mortgage, commercial, consumer, indirect auto.

16
00:01:22.799 --> 00:01:26.760
We do stuff in the branch as
well, but essentially any workflow where there's

17
00:01:26.760 --> 00:01:30.640
a lot of incoming paperwork and there's
important data in those documents, we can

18
00:01:30.640 --> 00:01:36.239
come in and really automate the manual
data entry type tasks and the manual stare

19
00:01:36.280 --> 00:01:41.040
and compare. That is a prevalent
in the back office of many financial institutions.

20
00:01:42.159 --> 00:01:44.439
Gotcha, Yeah, it makes sense. And I know, as we've

21
00:01:44.480 --> 00:01:49.319
talked a lot of similarities that we've
had. You've been on both sides now

22
00:01:49.560 --> 00:01:53.959
in a bank and supporting banks on
the service side. What's your experience been

23
00:01:53.079 --> 00:01:57.640
like on both sides of that you
compare and contrast, like, what are

24
00:01:57.879 --> 00:02:00.920
the similarities and differences you've seen?
Yeah, uh, you know, it's

25
00:02:00.439 --> 00:02:04.879
there's a lot of paperwork in banks. And it's funny. We have a

26
00:02:04.920 --> 00:02:13.199
famous story around our business, uh
where we had a newly appointed chief digital

27
00:02:13.280 --> 00:02:15.000
officer of a bank, of a
big bank, and we had a meeting

28
00:02:15.039 --> 00:02:20.039
with him and he confidently said,
you know, my mandate is within the

29
00:02:20.080 --> 00:02:24.680
next twenty four months, we're going
to have no more paper around the bank.

30
00:02:24.759 --> 00:02:29.560
This thing is going to be completely
digital. That conversation was twelve years

31
00:02:29.560 --> 00:02:34.639
ago, and needless to stay,
there's still plenty of paper flying around and

32
00:02:35.080 --> 00:02:38.680
that chief Digital Officer no longer is
in his role. It did not last

33
00:02:38.759 --> 00:02:42.520
very long, so uh, I
think it's it's a it's a it's an

34
00:02:42.520 --> 00:02:47.719
interesting dynamic in our in our industry
where banks are trying to become more digital,

35
00:02:49.199 --> 00:02:53.120
more automated, more efficient, but
yet you've got legacy processes. You

36
00:02:53.159 --> 00:02:58.960
have a lot of regulations that drive
collecting, uh collecting documents from your borrowers

37
00:02:58.960 --> 00:03:02.759
and your customers. Uh. And
there's a lot of compliance uh that that

38
00:03:02.919 --> 00:03:07.759
still revolves around having a physical or
even a digital copy at someone's driver's license

39
00:03:07.879 --> 00:03:13.280
or a bank statement or a pay
staff And so how do banks kind of

40
00:03:13.879 --> 00:03:17.319
figure out ways to uh move into
the next next, the next century,

41
00:03:17.719 --> 00:03:22.560
uh, but also uh satisfy these
these demands on them. So it's a

42
00:03:22.560 --> 00:03:24.080
pretty unique. It's it's a hard
thing relative to other industries. I think

43
00:03:24.080 --> 00:03:28.199
it's hard for banks to try to
figure out how to how to navigate all

44
00:03:28.199 --> 00:03:31.319
this. Yeah, a lot lots
we could dig into from there, for

45
00:03:31.360 --> 00:03:36.560
sure, especially like I said,
the requirements around gathering uh documentation. I

46
00:03:36.560 --> 00:03:39.080
know we see that could be different
to the state down to the county level.

47
00:03:39.159 --> 00:03:45.199
Like how do you guys think about
again, you think banking and regulation

48
00:03:45.280 --> 00:03:51.000
and government like being two very legacy
driven, slow to change, Uh,

49
00:03:51.360 --> 00:03:58.639
areas like how do you manage across
all that different nuance and process. Yeah,

50
00:03:58.639 --> 00:04:01.759
that's a good thing about about AI
these days and as it's evolved.

51
00:04:01.960 --> 00:04:05.000
Uh, there is a lot of
variance, and there's these you know,

52
00:04:05.199 --> 00:04:11.879
legacy regulations. Uh, every every
local jurisdiction has a different way of doing

53
00:04:11.960 --> 00:04:13.879
things, Every bank has a different
way of doing things, and a lot

54
00:04:13.879 --> 00:04:15.800
of the way, a lot of
the policies at the bank are kind of

55
00:04:15.920 --> 00:04:23.480
entrenched, and uh, trying to
completely overhaul that is is probably kinney,

56
00:04:23.519 --> 00:04:27.959
It's probably too hard for most banks
and lenders. And so instead it's how

57
00:04:27.959 --> 00:04:31.839
do you use new technology? How
can you use AI to really augment your

58
00:04:31.879 --> 00:04:35.600
existing workflows? You're not going to
try to change the whole paradigm of the

59
00:04:35.639 --> 00:04:40.720
bank or the regulatory landscape under which
you you operate, but can you use

60
00:04:40.800 --> 00:04:44.920
AI to automate some of the most
kind of mundane, boring, routine tasks

61
00:04:45.000 --> 00:04:46.959
that that you've had to do in
the past, And so, uh,

62
00:04:47.360 --> 00:04:50.639
it's it's interesting I hear you know, there's obviously a lot of talk around

63
00:04:50.800 --> 00:04:56.160
right now about AI This is something
that we've been doing for years and years,

64
00:04:56.199 --> 00:04:59.600
so we're not we're not new to
it, but we chuckle because,

65
00:05:00.079 --> 00:05:04.639
uh, finding like really practical use
cases is still a challenge for a lot

66
00:05:04.680 --> 00:05:08.959
of a lot of banks as well
as a lot of technology partners that want

67
00:05:09.000 --> 00:05:13.399
to sell AI stuff into into a
bank. You have to show up with

68
00:05:13.439 --> 00:05:17.879
a really proven, uh kind of
uh really ROI effective use case, and

69
00:05:18.160 --> 00:05:21.160
that's that's hard for a lot of
folks to do. Yeah, I want

70
00:05:21.199 --> 00:05:25.000
to dig into some of those use
cases because I don't know if you've experienced

71
00:05:25.040 --> 00:05:28.839
this as well, and kind of
getting the sense that maybe you have that

72
00:05:29.399 --> 00:05:32.000
when when banks talk about, you
know, becoming digital or making their process

73
00:05:32.040 --> 00:05:39.120
digital, they're just taking that same
like tired, cumbersome offline process and putting

74
00:05:39.160 --> 00:05:42.639
it on the internet. So they're
not you know, thinking about how can

75
00:05:42.680 --> 00:05:46.160
we use these new technologies tools with
AI to make it easier for the consumer

76
00:05:46.360 --> 00:05:50.759
or even internal staff, and it's
just digitizing, uh you know, the

77
00:05:50.800 --> 00:05:55.759
same process. How do you think
about are those things different or similar?

78
00:05:55.959 --> 00:05:59.839
What's what's yours? It's really important
at that, you know, and we

79
00:06:00.160 --> 00:06:02.120
this a lot when we start out
with the bank and we start talking about

80
00:06:02.160 --> 00:06:08.199
implementing our own product and capture to
help streamline, you know, a specific

81
00:06:08.199 --> 00:06:11.439
workflow. We have to get really
specific. What is it that we're actually

82
00:06:11.480 --> 00:06:14.600
doing at the bank, What are
we actually replacing? What steps are you

83
00:06:14.639 --> 00:06:17.040
guys doing today that either you're not
going to have to do anymore or you

84
00:06:17.079 --> 00:06:21.560
may have to do them different.
Uh. That change management process is probably

85
00:06:21.560 --> 00:06:28.759
one of the biggest risk points for
a bank that's trying to adopt AI,

86
00:06:29.720 --> 00:06:31.000
UH, because if it does,
you know, for it to work,

87
00:06:31.079 --> 00:06:35.480
it requires you to maybe think about
changing your process and you can't just necessarily

88
00:06:35.519 --> 00:06:39.120
always layer it on the existing way
you do things. A lot of times

89
00:06:39.160 --> 00:06:42.639
you should try to start with that, but uh, it will require maybe

90
00:06:42.680 --> 00:06:46.279
a different way of thinking about a
specific process or people's roles and responsibilities.

91
00:06:46.879 --> 00:06:50.560
And uh, you know, we've
seen this unfortunately on our side, where

92
00:06:51.079 --> 00:06:55.399
you know, we'll come in,
we'll build a really good solution that that

93
00:06:55.480 --> 00:06:58.879
can save that bank a lot of
time, a lot of money. Uh,

94
00:06:58.959 --> 00:07:02.879
But then there's so much resistance to
changing a process to accommodate this this

95
00:07:02.879 --> 00:07:05.879
this AI that like they don't get
much value out of it. And that's

96
00:07:05.920 --> 00:07:11.120
where we have to have these hard
conversations on it's not just the technology that's

97
00:07:11.120 --> 00:07:14.360
going to show up and solve all
your problems. You have to be willing

98
00:07:14.399 --> 00:07:16.680
to change. You have to be
willing to adapt, and you know,

99
00:07:16.720 --> 00:07:19.720
we're going to try to make that
as easy as possible. But it's it's

100
00:07:19.800 --> 00:07:23.399
very much a mindset as well as
you go into a you know, kind

101
00:07:23.439 --> 00:07:29.319
of an AI specific project. Yeah, that's really true about the change management

102
00:07:29.319 --> 00:07:30.079
of all of this, because,
like you said, you want to get

103
00:07:30.079 --> 00:07:34.040
really specific on the use cases and
but not make it a complete overall of

104
00:07:34.040 --> 00:07:39.439
a system, because then you kind
of get paralysis analysis. So you don't

105
00:07:39.480 --> 00:07:42.439
know where to start if you think
you have to rip and replace an entire

106
00:07:42.600 --> 00:07:46.680
process or system. So I'm sure
you know, understanding the entire landscape of

107
00:07:46.720 --> 00:07:49.759
a process and then digging in on
those use cases is probably pretty critical to

108
00:07:49.800 --> 00:07:55.600
getting folks excited about starting on that
journey and then keeping going. Well,

109
00:07:55.639 --> 00:07:58.439
and it's it's you know, it's
funny too. I guess this is human

110
00:07:58.519 --> 00:08:03.000
nature, but people, you know, people get very comfortable with the way

111
00:08:03.000 --> 00:08:07.920
they do things. And we had
a situation recently where we had built a

112
00:08:07.959 --> 00:08:15.759
solution to help a bank spread all
the tax returns that they gather as part

113
00:08:15.800 --> 00:08:20.120
of their small business lending process.
So you know, a small business lender,

114
00:08:20.160 --> 00:08:22.959
a banker at the bank is going
to work with a barrower and they're

115
00:08:22.959 --> 00:08:26.399
going to be sending in tax return
documents instead of the bank. Now,

116
00:08:26.439 --> 00:08:28.800
instead of these credit analysts or these
processors having to manually go through the tax

117
00:08:28.839 --> 00:08:33.639
returns and enter all the income information, we do that automatically. We can

118
00:08:33.679 --> 00:08:37.679
read through the documents, extract the
data, we put it in the system.

119
00:08:37.279 --> 00:08:41.679
And we were talking to one lady
and about her team, and I

120
00:08:41.679 --> 00:08:43.519
said, hey, how's it working. She goes, Oh, your software

121
00:08:43.519 --> 00:08:46.399
work's great, but nobody's using it. And I was like, well why.

122
00:08:46.519 --> 00:08:50.559
She said, well, they don't
like skipping this step. They really

123
00:08:50.679 --> 00:08:54.159
like manually entering all the data because
they say it quote unquote helps them understand

124
00:08:54.200 --> 00:08:56.720
the loan better and get to know
the loan and get to know the borrower

125
00:08:58.000 --> 00:09:00.960
and they can kind of think as
they go along. And I kind of

126
00:09:00.960 --> 00:09:03.320
threw my hands up and laughed.
I said, I don't know if I

127
00:09:03.320 --> 00:09:05.080
can do anything about that, and
she laughed too. She said, no,

128
00:09:05.159 --> 00:09:09.360
it's it's hard to get people to
think about changing their behavior, especially

129
00:09:09.399 --> 00:09:13.039
if it's becoming grained in the way
that they work, in the way that

130
00:09:13.080 --> 00:09:16.960
they make decisions. And so we
talked about a couple of kind of practical

131
00:09:16.000 --> 00:09:22.320
ways to maybe let these people trust
the system to do it correctly and they

132
00:09:22.320 --> 00:09:24.639
can go back and review later.
But that's that's a hard thing, you

133
00:09:24.639 --> 00:09:28.120
know. And I would you know, if you're if you are kind of

134
00:09:28.159 --> 00:09:33.879
an exec for a lender and you're
thinking about, you know, implementing new

135
00:09:33.919 --> 00:09:37.279
technology, whether it's AI or anything
else, I think it's good to honestly

136
00:09:37.279 --> 00:09:41.639
ask yourself, do I have a
team that is willing to change, embrace

137
00:09:41.759 --> 00:09:45.720
change, try things differently, or
can I go find maybe a few folks

138
00:09:45.759 --> 00:09:48.000
that can be leaders in this area
and can kind of be the advocates for

139
00:09:48.039 --> 00:09:50.600
the change, Because if you if
you're not willing to change or really push

140
00:09:50.639 --> 00:09:56.080
that and push your team to change, whatever you implement, regardless of technology,

141
00:09:56.120 --> 00:10:00.240
it probably is not going to work
very well. Yeah, that's true,

142
00:10:00.240 --> 00:10:03.759
And I think that's a really good
story, really specific example, but

143
00:10:03.879 --> 00:10:07.480
maybe highlights some things that we could
dig in more broadly around AI. As

144
00:10:07.519 --> 00:10:11.919
you said, like being able to
trust us any system really is probably key

145
00:10:11.960 --> 00:10:18.039
to adoption and AI for better or
worse. I think it brings with it

146
00:10:18.080 --> 00:10:22.480
a reputation of being a quote unquote
black box or something harder to understand.

147
00:10:22.879 --> 00:10:26.759
So maybe we could start a little
bit just like getting more specific around the

148
00:10:26.879 --> 00:10:28.960
termin as you've used it a couple
of times, like when you think about,

149
00:10:30.519 --> 00:10:33.240
you know, what a capture does
and how you guys use AI,

150
00:10:33.440 --> 00:10:35.360
like what does that specifically mean,
or what are some of the approaches that

151
00:10:35.399 --> 00:10:41.120
you're taking to make make your bags
easier. It's funny, Matt, we've

152
00:10:41.120 --> 00:10:46.519
been using AI and specifically what we
use as supervised machine learning. I can

153
00:10:46.679 --> 00:10:50.879
explain what that is in a second, but we've been using this for like

154
00:10:50.919 --> 00:10:54.440
the last four to five years,
and you know, up until about a

155
00:10:54.480 --> 00:10:56.799
year and a half ago, we
didn't even say the words AI or say

156
00:10:56.840 --> 00:11:03.240
machine learning because it would just scare
everybody. UH people have these preconceived ideas

157
00:11:03.279 --> 00:11:05.399
of what AI means, what it
is when it isn't what it does at

158
00:11:05.399 --> 00:11:09.679
the bank, and we just felt
there was so much barrier, there was

159
00:11:09.679 --> 00:11:15.080
so much kind of pre existing UH
biases against it, that we really referred

160
00:11:15.120 --> 00:11:20.519
ourselves more to kind of an automation
technology or automation platform, and it wasn't

161
00:11:20.600 --> 00:11:24.200
until chat GPT about eighteen months ago
kind of entered the public lexicon and everybody

162
00:11:24.600 --> 00:11:28.639
kind of got comfortable talking about AI
and realized, hey, for better or

163
00:11:28.679 --> 00:11:31.639
for worse, I'm gonna have to
figure out a way to use some flavor

164
00:11:31.639 --> 00:11:37.120
of AI going forward. That's really
allowed us to speak I think a lot

165
00:11:37.159 --> 00:11:39.879
more candidly about it. And so, uh, you know, I won't

166
00:11:39.879 --> 00:11:43.519
get in all the weeds and all
the types of the AI that are out

167
00:11:43.559 --> 00:11:50.159
there, but you know, there's
there are several different uh strategies and techniques

168
00:11:50.159 --> 00:11:52.759
of using artificial intelligence. That's a
really really broad term. Uh. You

169
00:11:52.799 --> 00:11:58.399
know what we do around supervised machine
learning is we are it's almost like we

170
00:11:58.559 --> 00:12:01.399
have you know this this for that
almost behave someone like like a fresh college

171
00:12:01.440 --> 00:12:05.120
grad that's maybe in their very first
job, and they're smart, they're willing

172
00:12:05.159 --> 00:12:07.960
to learn, but you really have
to teach them what the job is and

173
00:12:09.000 --> 00:12:11.200
what they're expected to do. And
so in our use case, you know,

174
00:12:11.360 --> 00:12:18.879
we have trained our machine learning models
to you know, recognize certain types

175
00:12:18.879 --> 00:12:20.279
of documents. So we'll feed it
samples of documents, and we'll feed it

176
00:12:20.320 --> 00:12:24.120
a bunch of different driver's licenses and
say hey, these are driver's licenses.

177
00:12:24.360 --> 00:12:26.519
And then we'll feed a bunch of
samples of bank statements and say hey,

178
00:12:26.519 --> 00:12:30.399
these are bank statements, and all
the while the system is looking at these

179
00:12:30.399 --> 00:12:33.240
documents and looking at maybe some of
the similarities and some of the differences between

180
00:12:33.240 --> 00:12:37.519
the document types, so that you
know, one day I can show up

181
00:12:37.519 --> 00:12:39.879
with kind of a mystery document,
give it to the system, and they

182
00:12:39.879 --> 00:12:41.480
can look at it and say,
Okay, is this a bank statement,

183
00:12:41.960 --> 00:12:46.200
is this a driver's license, or
is this something else? And you know,

184
00:12:46.279 --> 00:12:50.039
the more that you could train them
and teach them and give them hints,

185
00:12:50.039 --> 00:12:54.559
give them rules, give them reminders, or reinforce kind of positive decision

186
00:12:54.559 --> 00:12:58.799
making the better it is at kind
of I would say, guessing the right

187
00:13:00.240 --> 00:13:03.639
the right outcome in terms of what
documents looking at or the data it's looking

188
00:13:03.639 --> 00:13:07.120
for. So that's a really kind
of simplified version of how our our machine

189
00:13:07.159 --> 00:13:11.240
learning works, our AI. That's
actually very different than what chat GPT is

190
00:13:11.399 --> 00:13:16.159
as you as you guys have heard
around the generative AI. That's a that's

191
00:13:16.159 --> 00:13:20.440
a different type of AI than instead
of kind of guessing an answer uh to

192
00:13:20.519 --> 00:13:24.639
a specific question based on previous training, this AI is more about creating new

193
00:13:24.679 --> 00:13:28.039
content and trying to figure out what's
the next word that I should say in

194
00:13:28.080 --> 00:13:31.519
the Senate so that the sentence is
complete and makes sense. Relative the question

195
00:13:31.559 --> 00:13:37.080
that was asked, so use cases
there for generative AI are are are a

196
00:13:37.080 --> 00:13:41.399
bit different than kind of the AI
we use for reading documents. And I

197
00:13:41.440 --> 00:13:45.279
think it's important for you know,
someone who's looking to bring on AI to

198
00:13:45.360 --> 00:13:48.600
understand, Hey, tell me how
your AI works. Tell me how much

199
00:13:48.639 --> 00:13:52.240
of a black box it is,
because you know, in our case,

200
00:13:52.279 --> 00:13:54.919
we have kind of intentionally designed ra
AI to where it's less of a black

201
00:13:54.960 --> 00:13:58.279
box. In some ways. You
would say, well, maybe it's not

202
00:13:58.320 --> 00:14:01.120
as effective. I would argue,
well, we at least know how it's

203
00:14:01.159 --> 00:14:05.000
making the decisions it's making, and
so every time it makes it a decision,

204
00:14:05.279 --> 00:14:09.480
we can point to kind of the
way it went about that versus it

205
00:14:09.559 --> 00:14:13.039
being some just magical thing that comes
up with an answer and we don't know

206
00:14:13.039 --> 00:14:16.080
how it gets there. So that's
a big you know, that's a big

207
00:14:16.120 --> 00:14:18.559
part of it. And then the
other thing is, especially for banks and

208
00:14:18.639 --> 00:14:22.159
lenders credit unions, how are you
know, where are you using AI in

209
00:14:22.200 --> 00:14:26.440
your organization? It's one thing if
you're using AI like ours, which is

210
00:14:26.519 --> 00:14:31.159
just really kind of automating these Monday
tasks, you know, doing replacing manual

211
00:14:31.240 --> 00:14:35.159
data entry. It's different if you're
trying to use AI to make a credit

212
00:14:35.159 --> 00:14:41.399
decision, or you're using AI to
have a conversation with a customer or provide

213
00:14:41.480 --> 00:14:46.399
kind of a more native kind of
conversational onboarding experience for a customer. Those

214
00:14:46.440 --> 00:14:50.240
all have kind of their own unique
risks and challenges. And again for lenders

215
00:14:50.240 --> 00:14:52.919
specifically, there's a lot of regulation
that would say, hey, if you're

216
00:14:52.919 --> 00:14:54.679
going to put AI in to make
a decision on the loan, you better

217
00:14:54.679 --> 00:14:58.799
have a lot of confidence that you
understand how that system is making that decision

218
00:15:00.279 --> 00:15:05.919
so that you can prove it's not
discriminating against certain types of borrowers. Hey

219
00:15:05.960 --> 00:15:09.799
there, former host Jeff Kelviner here
to let you know about an exciting opportunity

220
00:15:09.799 --> 00:15:15.000
to strengthen your understanding of artificial intelligence
and financial services. If you want to

221
00:15:15.000 --> 00:15:18.799
better understand the fundamentals of how AI
models work and how they will impact the

222
00:15:18.879 --> 00:15:22.879
financial services industry, your in Luck
Upstart is launched an online course and certification

223
00:15:22.960 --> 00:15:26.679
designed to do just that. This
course will give you the knowledge you need

224
00:15:26.679 --> 00:15:33.000
to lead the conversation about how your
organization can best leverage and manage this revolutionary

225
00:15:33.000 --> 00:15:37.279
technology. If that sounds interesting,
check out the course as it's currently free

226
00:15:37.360 --> 00:15:41.120
for a limited time. You can
learn more at upstart dot com slash AI

227
00:15:41.200 --> 00:15:46.960
Certification. Again, that's upstart dot
Com slash AI Certification. Thanks and now

228
00:15:46.159 --> 00:15:50.159
back to the show. Yeah,
totally. I think every application, we

229
00:15:50.240 --> 00:15:54.120
keep coming back to the idea of
use cases, but defining those use cases,

230
00:15:54.240 --> 00:15:58.000
understanding the risk and trade offs are
definitely important. I think when you're

231
00:15:58.080 --> 00:16:03.759
embarking on this kind of a journey
or deciding if an AI application is right

232
00:16:03.879 --> 00:16:07.960
for a certain use case and really
understanding uh, the benefits you get there

233
00:16:07.320 --> 00:16:11.679
And you didn't you mentioned compliance too. I think sometimes you know, folks

234
00:16:11.720 --> 00:16:18.720
maybe view that as a more adversarial
relationship, or different levels of government will

235
00:16:18.720 --> 00:16:23.879
come out with their perspective or recommendations
on use of AI. But we talked

236
00:16:23.919 --> 00:16:27.480
before about you know, how how
can you use AI more of a proactive

237
00:16:27.519 --> 00:16:32.759
way or being more compliant or helping
uh, you know, stay within regulation,

238
00:16:32.919 --> 00:16:36.840
regulation and guidelines. How have you
guys thought about that and in ways

239
00:16:36.879 --> 00:16:41.080
that you can actually use this technology
to help you maybe show up better in

240
00:16:41.080 --> 00:16:47.000
those cases. Yeah, it's it's
funny. There has been this growth in

241
00:16:48.279 --> 00:16:52.919
UH regulations for banks and lenders and
how you know, UH, there's more

242
00:16:52.960 --> 00:16:55.919
and more data that has to be
reported to the government to make sure that

243
00:16:55.960 --> 00:17:02.840
you're not discriminating in your lending practices. There's recently been a revamp of you

244
00:17:02.879 --> 00:17:04.799
know, the CRA rules, which
have been around for fifty years, but

245
00:17:04.880 --> 00:17:10.279
those have expanded. There's this upcoming
you know Dodd frankedin seventy one rule.

246
00:17:10.319 --> 00:17:11.880
And I realize I'm saying a lot
of acronyms here, Matt, but I'm

247
00:17:11.880 --> 00:17:15.440
sure your listeners will know what these
are. There's just this growing body of

248
00:17:15.480 --> 00:17:19.200
reporting data to the government. And
you know, what we've found for our

249
00:17:19.319 --> 00:17:23.599
use case is there's actually a really
there's a really good solution here. A

250
00:17:23.640 --> 00:17:30.079
lot of times these compliance teams are
having to go review you know, loan

251
00:17:30.160 --> 00:17:33.960
packages to make sure that the data
they've pulled out of their system that they're

252
00:17:33.960 --> 00:17:37.960
about to report to the regulator is
accurate and complete, and so they spend

253
00:17:37.000 --> 00:17:42.599
a lot of time hunting and pecking
through big loan package PDFs, you know,

254
00:17:42.640 --> 00:17:48.039
double checking various data points against the
system. It's very time intensive and

255
00:17:48.200 --> 00:17:52.880
kind of cumbersome process. Most teams
can't even check every loan. They don't

256
00:17:52.920 --> 00:17:55.279
have the staff or the time to
do that. So we'll just sample a

257
00:17:55.279 --> 00:17:59.599
few loans. And that's a great
use case for AI. And again we're

258
00:17:59.599 --> 00:18:03.880
not you know, it's it's it's
not something where we're manipulating data and reporting

259
00:18:03.880 --> 00:18:06.839
without your you know, it's not
this black box scenario. It's simply saying,

260
00:18:06.880 --> 00:18:11.000
okay, AI, I want you
to go through all of my loan

261
00:18:11.039 --> 00:18:12.720
packages and I want you to pull
out the data that I need to report.

262
00:18:12.720 --> 00:18:15.160
I want you to compare it to
my system, and I want you

263
00:18:15.200 --> 00:18:19.359
to flag any discrepancies. And you
know what we found is ninety to ninety

264
00:18:19.359 --> 00:18:23.079
five percent of the data points that
are being reported, they're accurate, they're

265
00:18:23.119 --> 00:18:26.359
consistent, and you don't need to
do anything with them. But maybe it's

266
00:18:26.359 --> 00:18:30.720
five or ten percent of your data
points are they're missing data fields or it's

267
00:18:30.720 --> 00:18:33.039
a different name. You know,
it should say Will Robinson, but instead

268
00:18:33.039 --> 00:18:37.200
it says Matt Snow and you need
to go figure out why that is.

269
00:18:37.279 --> 00:18:42.200
And so it cuts down a lot
of this kind of mundane data scrubbing activity

270
00:18:42.200 --> 00:18:45.279
that these compliance teams do. So
I think that's kind of an interesting thing

271
00:18:45.319 --> 00:18:49.960
about you know, you hear about
compliance and banking and regulations going to restrict

272
00:18:51.039 --> 00:18:57.279
AI. Ironically, banks are using
AI to facilitate their obligations under the regulations.

273
00:18:57.319 --> 00:19:00.200
So, you know, I think
that's kind it's kind of been an

274
00:19:00.240 --> 00:19:06.480
ironic twist of AI. AI helping
banks even though it's going to get cracked

275
00:19:06.480 --> 00:19:08.119
down on I think more and more. Yeah, I think you're right.

276
00:19:08.160 --> 00:19:12.000
I think that's a good example too
of where, especially with human resources that

277
00:19:12.000 --> 00:19:17.279
are very expensive and limited, like, can you use a combination of these

278
00:19:17.720 --> 00:19:22.079
two tools, right the AI that
can help you focus and prioritize or find

279
00:19:22.519 --> 00:19:25.599
higher risk signals. I think we
see that a lot in terms of you

280
00:19:25.640 --> 00:19:30.359
know, verification in underwriting as well, Like you're seeing tens of thousands of

281
00:19:30.440 --> 00:19:33.960
applications in any day, which ones
do you focus on, where do you

282
00:19:33.960 --> 00:19:38.039
put the human capital against that?
And where should you focus on? Types

283
00:19:38.079 --> 00:19:42.400
of documents or types of fraud?
And being able to partner those two together,

284
00:19:42.559 --> 00:19:47.519
each with their unique skills and either
handling volume or you know, novelty

285
00:19:47.720 --> 00:19:51.759
in the data I think can be
a great win. Yeah, you know,

286
00:19:52.119 --> 00:19:52.960
We get asked this question a lot, and you know, it's like

287
00:19:52.960 --> 00:19:56.359
every conference you go to, that's
a big question, how do I use

288
00:19:56.400 --> 00:20:02.359
AI my bank? And I kind
of chuckle because I had not heard many

289
00:20:02.359 --> 00:20:07.880
people show up on these panels and
give like three or five or seven or

290
00:20:07.880 --> 00:20:10.839
even you know, ten good answers, Like usually there's kind of one use

291
00:20:10.920 --> 00:20:15.880
case or two. And I think
it's because sometimes you hear about all the

292
00:20:15.880 --> 00:20:22.640
potential of AI and it's hard to
figure out where to start, and there's

293
00:20:22.720 --> 00:20:29.559
so much hype around what it could
do that I think people miss like what

294
00:20:29.599 --> 00:20:32.480
are some really really practical things that
can do today? So we always just

295
00:20:32.559 --> 00:20:33.920
encourage people, and again this is
unique to our business a little bit,

296
00:20:33.960 --> 00:20:37.920
and the fact that we you know, we process, we process documents.

297
00:20:37.680 --> 00:20:41.319
But go look through your existing workflows
and look at where people are spending a

298
00:20:41.359 --> 00:20:48.119
lot of time doing repetitive tasks that
are that are highly rules based. Right,

299
00:20:48.160 --> 00:20:51.160
you probably have a rule or a
process because in those situations you can

300
00:20:51.279 --> 00:20:55.240
use technology to automate that. It
doesn't even have to be AI. So

301
00:20:55.319 --> 00:20:57.799
much of what a load origination system
is, which doesn't been around forever,

302
00:20:57.960 --> 00:21:03.880
is really just making sure you're applying
your bank policies and bank rules around a

303
00:21:03.960 --> 00:21:07.240
process and around an underwriting set of
decision making. So that and there's no

304
00:21:07.319 --> 00:21:12.200
AI in that. So I do
think it's an It's important you know that

305
00:21:12.279 --> 00:21:18.200
you first take stock of do we
have tight, repetitive processes that can be

306
00:21:18.240 --> 00:21:22.240
automated. The flip side of that
is, don't try to go use technology

307
00:21:22.680 --> 00:21:26.039
or AI until you've got that part
down. We struggled with this a lot

308
00:21:26.079 --> 00:21:30.000
with some of our customers. We
will go they will be so excited about

309
00:21:30.079 --> 00:21:33.079
using AI. Oh, you know, you're going to automate our loan origination

310
00:21:33.240 --> 00:21:37.640
process and you're going to help,
you know, make sure that we're following

311
00:21:37.640 --> 00:21:40.039
all these rules and we're collecting all
the right documents. And we're like,

312
00:21:40.119 --> 00:21:41.880
yeah, that's great. Tell us
your process, tell us your rules so

313
00:21:41.920 --> 00:21:45.039
we can make sure the AI is
following your rules. And then quickly you

314
00:21:45.079 --> 00:21:48.240
realize they don't really have clean rules. They don't have consistency on what the

315
00:21:48.319 --> 00:21:52.599
rules are. We did. This
was a couple of years ago, but

316
00:21:52.640 --> 00:21:55.960
we rolled out a solution based on
what the project team at the bank had

317
00:21:55.960 --> 00:21:59.119
said. These are our rules and
this is how we process loans, and

318
00:21:59.160 --> 00:22:03.839
so we we started testing it with
actual users and they revolted. They were

319
00:22:03.839 --> 00:22:06.559
like, this thing is terrible,
and we said, well, why is

320
00:22:06.599 --> 00:22:08.839
it terrible? And then we uncovered
it's because they could not agree internally on

321
00:22:08.880 --> 00:22:12.799
what even the rules were or the
standard processes were, and so they thought

322
00:22:14.359 --> 00:22:15.119
that, you know, it was
AI's fault. And we were like,

323
00:22:15.160 --> 00:22:18.559
hey, we we can make this
do anything you want, but you have

324
00:22:18.599 --> 00:22:22.559
to have consistent standards at the bank
that you can follow. Otherwise this is

325
00:22:22.559 --> 00:22:25.799
not going to help at all.
So I think starting with kind of,

326
00:22:25.880 --> 00:22:29.559
you know, making sure you have
clear process, making sure that if you're

327
00:22:29.559 --> 00:22:33.319
going to be using data, your
own data to help with an AI enabled

328
00:22:33.319 --> 00:22:37.319
project, making sure that data is
clean and well structured is really important.

329
00:22:37.720 --> 00:22:40.680
Again, this happens a lot with
us. We look at we look at

330
00:22:40.680 --> 00:22:42.400
documents, and so a lot of
times we'll go into a bank and say,

331
00:22:42.400 --> 00:22:47.680
hey, pull us samples of these
documents so we can make sure our

332
00:22:47.680 --> 00:22:49.720
machine learning model can read it.
Well. They don't have their documents safe

333
00:22:49.759 --> 00:22:52.799
down very well, and they don't
know where they are, or they have

334
00:22:52.839 --> 00:22:57.279
a miss misclassified or misindexed, and
so we start with a really dirty sample

335
00:22:57.319 --> 00:23:00.000
set and it just doesn't go well, So, uh, there's a there's

336
00:23:00.000 --> 00:23:03.440
a lot to do here on the
AI front. But I think just taking

337
00:23:03.440 --> 00:23:06.599
a step back and looking at hey, what are the what are the most

338
00:23:06.680 --> 00:23:08.960
mundane, routine, repetitive things that
my team's doing every day that's taking a

339
00:23:10.000 --> 00:23:12.319
lot of time and maybe it's not
very much fun and how can I go

340
00:23:12.359 --> 00:23:17.119
automate that? And they'll typically be
a good use case there. Yeah,

341
00:23:17.279 --> 00:23:19.759
that makes sense. So as if
you were advising somewhere, you know,

342
00:23:19.799 --> 00:23:25.039
at a bank or a credit union, thinking about automation use of AI,

343
00:23:26.799 --> 00:23:29.839
how do you think about the you
know, bill versus buy or what are

344
00:23:29.839 --> 00:23:32.400
the different dimensions? Is it like, like you said, the type of

345
00:23:32.480 --> 00:23:34.759
data that I have, or maybe
for a third party would have access to

346
00:23:36.240 --> 00:23:40.839
the skills I have natively, you
know, and evaluating each of these different

347
00:23:40.880 --> 00:23:44.279
use cases that might exist. How
would you advise someone to think about,

348
00:23:44.400 --> 00:23:45.400
you know, is this something we
should try to tackle on our own or

349
00:23:45.440 --> 00:23:51.640
start looking for help outside. I
think it's really a function of how big

350
00:23:51.720 --> 00:23:53.559
is your bank, how big is
your budget, and what are you trying

351
00:23:53.599 --> 00:23:57.680
to do. I do think the
biggest banks, you know, the JP

352
00:23:57.799 --> 00:24:02.759
Morgan's of the world, they have
the budget, in the in the appetite

353
00:24:02.759 --> 00:24:06.200
to kind of build as much as
they can in house. At the same

354
00:24:06.240 --> 00:24:10.480
time, we work with some of
those those biggest names, and what we

355
00:24:10.599 --> 00:24:15.279
found is the problem that we're solving
for them is really important to the business,

356
00:24:15.039 --> 00:24:19.599
but from like a from an IT
or product perspective inside the bank,

357
00:24:19.839 --> 00:24:22.480
it's not very it's not very important. So the bank's not going to get

358
00:24:22.519 --> 00:24:26.039
to it even though they could build
it there, it's going to take them

359
00:24:26.039 --> 00:24:27.000
a year or two to get there, and so the business is like,

360
00:24:27.039 --> 00:24:32.960
hey, I need this solve now. So you know, I would say,

361
00:24:32.960 --> 00:24:34.759
if you're like a line of business
leader at a lender, I think

362
00:24:36.599 --> 00:24:38.720
just being honest with yourself and with
your team on maybe my bank says they

363
00:24:38.759 --> 00:24:41.039
want to build it, or they
say they can, or they say they're

364
00:24:41.039 --> 00:24:45.000
working on it, but when's it
really going to be ready? Versus if

365
00:24:45.000 --> 00:24:48.359
I can find a vendor that can
solve this, and what's the cost of

366
00:24:48.400 --> 00:24:51.759
that and what's the timeline up getting
that in, I think you'll you'll realize

367
00:24:51.799 --> 00:24:55.640
even if the long term plan is
for the bank to build it themselves,

368
00:24:56.680 --> 00:24:59.400
I think you can probably get pretty
far with the vendor up early on.

369
00:24:59.559 --> 00:25:02.920
So in terms of by most banks, I mean just cutting to the chase,

370
00:25:02.960 --> 00:25:08.000
most banks are ill equipped to build
technology internally. Just most banks are.

371
00:25:08.039 --> 00:25:10.880
They're just not set up for that. It's not in their DNA,

372
00:25:10.920 --> 00:25:14.839
it's not their culture. They don't
they don't have success with it. So

373
00:25:15.039 --> 00:25:18.200
again, most banks, it's better
to partner with vendors who've done this for

374
00:25:18.240 --> 00:25:22.519
a long time, and again,
do do really good diligence on your on

375
00:25:22.559 --> 00:25:26.599
your vendors. Ask for referrals,
Uh, make sure you know that that

376
00:25:26.759 --> 00:25:32.559
they've got a long body of evidence
of this being successful. Ask for a

377
00:25:32.559 --> 00:25:37.319
lot of details around their process and
how their technology works. A good vendor,

378
00:25:37.319 --> 00:25:40.440
who's who's been doing this a while, we'll have no problem sharing that,

379
00:25:40.480 --> 00:25:44.319
and we'll have good documentation around that. If you feel like you're working

380
00:25:44.319 --> 00:25:48.759
with someone who's maybe trying to figure
this out alongside you, that's okay,

381
00:25:48.799 --> 00:25:51.359
But just be aware of that that
they're trying to figure it out alongside you,

382
00:25:51.400 --> 00:25:52.519
and it may not. It may
not be go as well as you

383
00:25:52.960 --> 00:25:56.680
as you're hearing in the sales process. So you know, I think there's

384
00:25:56.680 --> 00:26:00.000
gonna be more regulation around this as
well. We were talking about that earlier,

385
00:26:00.079 --> 00:26:04.039
Matt, kind of you know where
where's regulation going around this? The

386
00:26:04.079 --> 00:26:11.680
state of Colorado just passed a law
requiring organizations that use AI and these high

387
00:26:11.759 --> 00:26:15.759
they call it, these kind of
these high risk or high impact areas to

388
00:26:17.119 --> 00:26:21.920
have a lot better governance around what
is the AI doing, what data is

389
00:26:21.960 --> 00:26:23.039
it touching, what is it doing
with that data? How is it making

390
00:26:23.079 --> 00:26:27.079
decisions? How is it impacting the
outcome of the bank, is it discriminating?

391
00:26:27.680 --> 00:26:32.039
Is it being biased towards certain things? So I think more and more

392
00:26:32.039 --> 00:26:34.400
regulation is going to come out on
that. And again, vendors like ourselves,

393
00:26:34.400 --> 00:26:37.240
we you know, we we have
answers for all that, and so

394
00:26:37.400 --> 00:26:41.000
making sure that your vendor understands what's
going on, I think we'll give a

395
00:26:41.000 --> 00:26:45.319
lot of confidence. Yeah, the
transparency and trust I think is key.

396
00:26:45.799 --> 00:26:48.119
And bringing back to what you said
earlier about you know, maybe the LMS

397
00:26:48.160 --> 00:26:52.640
and chat GPT bringing more awareness to
the space in general. Have you seen

398
00:26:53.039 --> 00:26:57.400
your partners, you know, asking
harder questions, getting into deeper diligence or

399
00:26:57.400 --> 00:27:00.599
are you still having to do a
lot of education on you know what what

400
00:27:00.880 --> 00:27:06.079
some of the terminology is, the
mechanisms you're using in this space to help

401
00:27:06.559 --> 00:27:11.000
demystify some of that. You know, it's a it's a wide spectrum.

402
00:27:11.200 --> 00:27:14.160
Most I will say this, most
banks are doing a good job that when

403
00:27:14.200 --> 00:27:18.480
they hear AI kind of their the
red flag goes up and they ask more

404
00:27:18.559 --> 00:27:21.559
questions, like, this is not
just a routine. I'm not just buying

405
00:27:21.640 --> 00:27:25.920
you know more more you know,
telephones for my office desks at my you

406
00:27:25.960 --> 00:27:29.880
know, at my company. This
is this is something I need to have

407
00:27:29.920 --> 00:27:33.079
a better understanding up. And so
we do a lot of education around that.

408
00:27:33.319 --> 00:27:36.440
And again, AI is such a
broad term. There's so many different

409
00:27:36.440 --> 00:27:40.319
techniques that, like, what I
tell you about how our stuff works does

410
00:27:40.319 --> 00:27:41.920
not mean every other AI vendor out
there is going to work the same way.

411
00:27:41.960 --> 00:27:45.079
In fact, most of them won. And so getting a good understanding,

412
00:27:45.119 --> 00:27:51.240
making sure that your vendor can explain
it crisply and simply, because you're

413
00:27:51.240 --> 00:27:52.839
gonna have to go turn around and
explain it to your bosses. What is

414
00:27:52.839 --> 00:27:55.920
it this we're putting in, How
does it work? How do we make

415
00:27:55.960 --> 00:27:59.960
sure it won't you know, run
a foul of any of our fair lind

416
00:28:00.240 --> 00:28:03.559
regulations? How am I gonna get
people to buy into it? And so

417
00:28:03.599 --> 00:28:07.960
I think if you can understand how
it works simply, then you'll have more

418
00:28:07.960 --> 00:28:11.240
confidence that will work well. Yeah, I agree. And one of the

419
00:28:11.240 --> 00:28:15.240
things I've found is while it's certainly
important, you know, pre contract and

420
00:28:15.279 --> 00:28:19.720
as a one time event, it's
also an ongoing exercise because you know,

421
00:28:19.799 --> 00:28:25.359
companies like ours, I think,
are constantly trying to find the next new

422
00:28:25.400 --> 00:28:30.599
innovative approach or like you said,
AI is so broad that the way we're

423
00:28:30.680 --> 00:28:33.680
leveraging those types of technology is changing
day by day. So it's really a

424
00:28:33.680 --> 00:28:37.640
part of the ongoing relationship. I
think if you're using a third party to

425
00:28:37.920 --> 00:28:44.000
stay in touch with them and how
they're evolving their thinking and kind of it's

426
00:28:44.000 --> 00:28:48.480
like ongoing diligence of the relationship.
Absolutely, I mean it's a it's a

427
00:28:48.519 --> 00:28:53.799
never ending If you find the right
vendor, there's there's typically a good chance

428
00:28:53.799 --> 00:28:56.000
they can do more than just one
thing for you. And if they can

429
00:28:56.039 --> 00:28:59.119
do one more than one thing,
that means they're gonna be sticking around for

430
00:28:59.119 --> 00:29:00.599
a while and they want to do
more than one thing. So I think

431
00:29:00.599 --> 00:29:04.559
it's important when you're not just looking
at the technology or the specific use case,

432
00:29:04.559 --> 00:29:07.519
but talking about vendor and saying,
hey is this a type of partner.

433
00:29:07.559 --> 00:29:10.559
Have they been around a while,
are they going to be around?

434
00:29:10.640 --> 00:29:14.839
Are they really committed to my industry, MySpace, my use cases? Do

435
00:29:14.920 --> 00:29:15.960
they want to do more with me? Or am I going to be kind

436
00:29:15.960 --> 00:29:19.279
of a one trick pony where they
just show up, do their thing and

437
00:29:19.279 --> 00:29:23.279
then leave and they never hear from
me and collect a connector collector renewal check.

438
00:29:25.720 --> 00:29:27.519
I think getting a good sense of
that's important because you can probably go

439
00:29:27.640 --> 00:29:33.039
find, you know, just a
handful or less of good vendors that can

440
00:29:33.079 --> 00:29:36.480
do a lot for you across the
bank, and you'll make a lot more

441
00:29:36.559 --> 00:29:40.960
progress sticking with those lenders and are
sticking with those vendors and going deep than

442
00:29:41.000 --> 00:29:44.440
trying to find thirty different vendors who
kind of each do one thing for you.

443
00:29:45.200 --> 00:29:48.279
Gotcha? Well, maybe it's a
good time to transition to a closing

444
00:29:48.440 --> 00:29:52.480
question for you. As I was
thinking about our conversation today, I was

445
00:29:52.519 --> 00:29:57.079
thinking back to a book I read
around this, Stuart Russell's Human Compatible.

446
00:29:57.119 --> 00:30:00.000
I'm not sure if you've read that, but always trying to find whether it's

447
00:30:00.039 --> 00:30:04.079
a podcast a thought leader book.
I thought that one was really interesting just

448
00:30:04.079 --> 00:30:11.480
around the idea setting up controls and
transparency around human and AI interactions. And

449
00:30:11.079 --> 00:30:15.000
while it was enlightening around AI and
some of the topics we've talked about,

450
00:30:15.000 --> 00:30:19.079
it helped me think about even you
know, human decision making and maybe processes

451
00:30:19.079 --> 00:30:23.400
we take for granted or that are
more automatic. So I highly recommend that

452
00:30:23.440 --> 00:30:26.599
if you haven't read it, but
curious if there's anyone you follow in this

453
00:30:26.680 --> 00:30:32.319
space to again kind of stay relevant
and keep your thoughts sharp in this this

454
00:30:32.400 --> 00:30:36.359
area, whether it's a podcast or
you know, maybe a writer that you

455
00:30:36.480 --> 00:30:40.759
follow. That's a good question,
Matt. I would if I mean very

456
00:30:40.759 --> 00:30:42.960
transparent. I feel like I'm following
twenty or thirty people right now and I'm

457
00:30:42.960 --> 00:30:48.279
still trying to figure out like who
is who is real and legit in this

458
00:30:48.440 --> 00:30:52.519
because it's and I also think kind
of the diversity of thought in this phase

459
00:30:52.519 --> 00:30:57.960
of AI. It's so early.
You have well, I think I'll give

460
00:30:57.960 --> 00:31:02.640
you an example. I think more
Andresen he's a prominent venture capitalist. He

461
00:31:02.680 --> 00:31:07.440
wrote a really good kind of long
form essay on AI and how we should

462
00:31:07.440 --> 00:31:10.319
be thinking about it, wrote it
several months ago, so I do encourage

463
00:31:10.359 --> 00:31:12.680
people to read that. But he
does talk about the concept of the Baptists

464
00:31:12.680 --> 00:31:17.319
and the bootleggers. Relating this back
to the prohibition, that there are people

465
00:31:17.319 --> 00:31:23.440
who are going to really call for
the the regulation of AI because they feel

466
00:31:23.440 --> 00:31:26.000
like in their hearts that's the right
thing for the world. But then there

467
00:31:26.000 --> 00:31:29.519
are people who are going to call
for the regulation of AI because they're the

468
00:31:29.559 --> 00:31:32.799
ones who are going to benefit the
most from it. And so I think

469
00:31:32.799 --> 00:31:36.440
in this early phase of AI and
how we use it, I think diversity

470
00:31:36.480 --> 00:31:41.200
of thought is really important so you
can hear kind of the different sides and

471
00:31:41.240 --> 00:31:45.759
then feel confident, you know,
making your own conclusions on what does AI

472
00:31:45.839 --> 00:31:48.319
mean for us as a as a
world, What does it mean for me

473
00:31:48.359 --> 00:31:51.880
in my business, what does it
mean for you know, my own professional

474
00:31:51.920 --> 00:31:57.960
development going forward. I did read
an interesting quote recently, and I can't

475
00:31:57.960 --> 00:32:01.680
remember who said it, but it
is essentially, you shouldn't worry about AI

476
00:32:01.759 --> 00:32:07.039
replacing your job. You should worry
about a person who knows how to use

477
00:32:07.079 --> 00:32:10.319
AI replacing your job. And so
I think, I think there's sometimes this

478
00:32:10.440 --> 00:32:15.400
fear that AI is going to replace
everybody, and I think there will be

479
00:32:15.440 --> 00:32:20.680
pockets where AI comes in and can
automate some of the things that people do,

480
00:32:20.720 --> 00:32:22.640
but I don't think those are the
most interesting jobs. And I think

481
00:32:22.680 --> 00:32:25.680
annially is going to create a lot
more opportunity for a lot more people and

482
00:32:25.960 --> 00:32:30.799
be a net benefit, just as
we've seen with every major technological advancement over

483
00:32:30.839 --> 00:32:34.880
the history of the history of me
and kind. So maybe a conversation for

484
00:32:34.920 --> 00:32:37.599
another day we can have, Matt, but I encourage you read, read

485
00:32:37.640 --> 00:32:42.359
diverse thoughts and continue to kind of
keep an up in mind on where this

486
00:32:42.400 --> 00:32:45.039
all goes. Yeah. I couldn't
agree more on that, will. I

487
00:32:45.079 --> 00:32:49.920
think that's really it's a really exciting
time to be in industry as these tools

488
00:32:49.920 --> 00:32:52.759
are coming on board, and I
think, like you said, the transparency,

489
00:32:52.799 --> 00:32:57.039
that openness in the dialogue. Let's
find ways to test and see what

490
00:32:57.240 --> 00:33:01.480
works and be open about the pros
and cons and see where this goes.

491
00:33:02.000 --> 00:33:06.480
So yeah, I'd love to follow
up with a conversation on that and appreciate

492
00:33:06.519 --> 00:33:09.920
your time with me today and thanks
again, Well, yeah, thanks so

493
00:33:10.000 --> 00:33:13.960
much for having me. Matt really
enjoyed it. Upstart partners with banks and

494
00:33:14.000 --> 00:33:19.039
credit unions to grow households and expand
consumer lending through its leading AI lending platform.

495
00:33:19.720 --> 00:33:23.359
Upstart powered banks and credit unions leverage
AI to offer higher approval rates and

496
00:33:23.480 --> 00:33:30.599
experience lower loss rates while simultaneously delivering
the exceptional digital first lending experience that consumers

497
00:33:30.680 --> 00:33:34.960
demand. Whether you're looking to grow
and enhance your existing personal and auto lending

498
00:33:35.000 --> 00:33:39.240
programs, or you're just getting started, upstart can help. Upstart offers an

499
00:33:39.319 --> 00:33:44.319
end to end solution that can help
you find more credit worthy borrowers within your

500
00:33:44.400 --> 00:33:49.799
risk profile with all digital underwriting,
verification, loan closing, and servicing,

501
00:33:50.240 --> 00:33:53.519
It's all possible with Upstart in your
corner. Learn more about finding new borrowers,

502
00:33:53.799 --> 00:33:59.440
enhancing your credit decisioning process, and
growing your business by visiting upstart dot

503
00:33:59.480 --> 00:34:02.960
com slash Lenders. That's upstart dot
com slash lenders