Name: Generative AI: Why is it important for financial services to pay attention to?
Uploaded: 2023-06-09T13:58:56.989Z
Duration: 387.52
Description: Sarah Hinkfuss, Partner, Bain Capital Ventures, breaks down the outlook on generative AI and shares her vision for its place within financial services. Stay up to date with the community on https://finovate.com/ Find out more about FinovateSpring on http://www.finovatespring.com/

FSPR.INT.005_500mb.mp4

Back

Or filter by your interest

Browse all our content

Finovate

Unlock all the content with only {{price}}

Sarah Hinkfuss.vtt

English

{{00:00:05}}
William Mills: Hi, I am William Mills.

{{00:00:06}}
I'm here at FinovateSpring
in San Francisco.

{{00:00:08}}
I'm pleased to be joined by Sarah Hinkfuss
who's a partner at Bain Capital Ventures.

{{00:00:12}}
How you doing, Sarah?

{{00:00:13}}
Great.

{{00:00:14}}
Thanks for having me.

{{00:00:15}}
Listen, I saw your fireside chat
this morning on generative AI,

{{00:00:19}}
and I thought that was very cool.

{{00:00:21}}
For those folks that weren't
at that panel, what was the key

{{00:00:24}}
takeaways and key points from that?

{{00:00:27}}
Sarah Hinkfuss: Yeah,
it was for 20 minutes.

{{00:00:28}}
We packed a lot in, so it was
a great discussion and I was

{{00:00:31}}
thrilled to have two CEOs join me.

{{00:00:33}}
So Alex, who's from vic.ai and
Lawrence from Slope Pay, who were

{{00:00:38}}
really on the forefront of thinking
about how to integrate generative

{{00:00:41}}
AI within financial services.

{{00:00:43}}
I would say the main takeaway, like
the most important thing that people

{{00:00:46}}
should know, is that generative AI
is a concept that is an ingredient to

{{00:00:51}}
add into financial services, right?

{{00:00:54}}
But in and of itself, it
doesn't solve all the problem.

{{00:00:57}}
William Mills: That's
just part of the solution.

{{00:00:59}}
One of the many tools.

{{00:01:00}}
Sarah Hinkfuss: Exactly
one of the many tools.

{{00:01:01}}
And I think that matters because
there's a lot of hope for it can be

{{00:01:06}}
a platform shift, which I think is
right, but that's only in combination

{{00:01:09}}
with great software, with other
types of more traditional AI models.

{{00:01:13}}
And so really seeing the entire
workflow and how you get your

{{00:01:15}}
entire company to embrace it is key
to the success of leveraging it.

{{00:01:20}}
William Mills: That's what I
was going to ask you about.

{{00:01:22}}
Why is generative of AI.

{{00:01:24}}
It's so important for us to
be paying attention to now.

{{00:01:27}}
There's a lot of buzz on AI in general.

{{00:01:29}}
Why this particular tool?

{{00:01:31}}
Sarah Hinkfuss: Yeah, so it
hasn't been possible until now.

{{00:01:34}}
So just in terms of why now there's
all of this technological change

{{00:01:38}}
that's enabled it to become available.

{{00:01:40}}
And the power of it is
absolutely incredible.

{{00:01:43}}
And so I would almost turn the question
back to you and say, for anyone

{{00:01:46}}
who doesn't think that they need
to care about it, their competitors

{{00:01:49}}
are going to, and disruptors in
their industry are going to as well.

{{00:01:52}}
And so for every company to actually have
a strategy and a way to integrate it,

{{00:01:57}}
because if they don't, other people will.

{{00:01:59}}
William Mills: So in what
areas do you see generative AI

{{00:02:02}}
revolutionizing the financial industry?

{{00:02:05}}
What, are the first steps?

{{00:02:06}}
Sarah Hinkfuss: Yeah.

{{00:02:06}}
In areas.

{{00:02:07}}
Yeah, totally.

{{00:02:08}}
So I, I wrote a blog post about this
actually, around the 10 applications

{{00:02:12}}
of generative AI within FinTech.

{{00:02:13}}
But maybe taking a step back from that,
if you think about what are the use cases

{{00:02:18}}
of generative ai, what does it do best?

{{00:02:21}}
It creates new content, so
generates content, right?

{{00:02:23}}
It enables much more efficient search,
so you can look through a lot of

{{00:02:27}}
documents and find exactly what you want.

{{00:02:30}}
And then it also enables summarization
of many documents as well.

{{00:02:33}}
So those are the three.

{{00:02:35}}
Reasons to be or, types of
work that it does, right?

{{00:02:39}}
You can think about those three
applications within the context

{{00:02:43}}
of financial services, right?

{{00:02:44}}
So financial services is about
distributing financial products,

{{00:02:48}}
manufacturing financial products,
and servicing financial products, and

{{00:02:52}}
every step of that journey requires.

{{00:02:54}}
Content generation
searching and summarization.

{{00:02:57}}
And so it's pretty transformative when it
comes to both the ability to generate new

{{00:03:02}}
products as well as the cost underlying
the provision of those financial services.

{{00:03:08}}
William Mills: That makes sense.

{{00:03:09}}
Being in the financial industry
and understand the importance of

{{00:03:12}}
compliance with the various regulators.

{{00:03:15}}
How is this gonna, how is a AI going
to influence or change the relationship

{{00:03:20}}
with the regulations that more
financial institutions have to follow?

{{00:03:24}}
Sarah Hinkfuss: Totally, and this,
is one of the reasons why I think

{{00:03:27}}
it's been a more complex topic
within financial services as well.

{{00:03:30}}
So if you think about marketing
applications or gaming or

{{00:03:34}}
code generation, right?

{{00:03:35}}
Those areas are, those domains are really
ahead of the curve on generative ai.

{{00:03:39}}
But within financial services,
there's a lot of concern around

{{00:03:42}}
what will regulators think?

{{00:03:43}}
How will it change my
relationship with customers?

{{00:03:46}}
There's a really big cost to
being wrong and sometimes it's

{{00:03:48}}
you're licensed to actually serve.

{{00:03:51}}
And so for that reason,
there's more reluctance.

{{00:03:53}}
So I, think a few things
I would say to that.

{{00:03:55}}
One is in speaking in, in my role as
an investor, I get to speak to a lot of

{{00:03:59}}
incumbents and a lot of upstarts, and
it's been really exciting to see some

{{00:04:02}}
of the creativity around leveraging the
code, like generative AI code within

{{00:04:07}}
the context of a paradigm that already
respects the rules and regulations.

{{00:04:11}}
So you can actually say instead of the
allowing the hallucination of any message,

{{00:04:16}}
you can have the parameters around
what's allowed, like from the SEC in

{{00:04:20}}
communication with customers, for example.

{{00:04:22}}
That can be also within the context
of that code, so that every message

{{00:04:26}}
has the wrapper of what that,
what is regulatorily allowed.

{{00:04:30}}
But I would say this part of the world
is very new on, on figuring all this out.

{{00:04:35}}
So we're still in
experimentation phase, right?

{{00:04:37}}
And so most financial services
companies that I'm talking to, they

{{00:04:40}}
have built the models internally, right?

{{00:04:42}}
They're playing around with them,
but very few things are actually

{{00:04:45}}
in production because they're
figuring out what these edge cases

{{00:04:47}}
are and how to properly implement
them within the regulatory context.

{{00:04:51}}
William Mills: That's
what I'm hearing as well.

{{00:04:52}}
Everybody wants to do it, but
nobody's sure where to start.

{{00:04:56}}
Sarah Hinkfuss: I think
you start anywhere.

{{00:04:58}}
I guess I would say on that, that
it's the, most important thing is that

{{00:05:01}}
you start, and so figuring out, like
having a SWAT team internally, right?

{{00:05:06}}
That is getting the data and starting
to put it against different use cases.

{{00:05:09}}
Or a lot of companies
do hackathons, right?

{{00:05:12}}
So really bringing everyone together.

{{00:05:14}}
Having no bad idea, putting stuff
on a whiteboard and then letting

{{00:05:17}}
people go for 24 hours on it.

{{00:05:19}}
I think there's a lot
of excitement around it.

{{00:05:20}}
And so the worst place to
start is nowhere, and the best

{{00:05:23}}
place to start is anywhere.

{{00:05:25}}
William Mills: So in closing, five years
from now, what do you think will be

{{00:05:29}}
commonplace AI use cases that we're not
seeing today or just being talked about?

{{00:05:34}}
Sarah Hinkfuss: Yeah, so the area
I'm most excited about is that most

{{00:05:38}}
people experience generative AI
today within a chatbot form, right?

{{00:05:43}}
Because it became so popular
when OpenAI released ChatGPT.

{{00:05:47}}
And so we're used to asking a question
and then having the chatbot answer,

{{00:05:50}}
but that's just one form factor to
expose the underlying models, right?

{{00:05:54}}
And so I'm really excited about
thinking through the other different

{{00:05:58}}
form factors that'll be relevant within
this tech stack of financial services.

{{00:06:02}}
And so I think it'll be coming an
ingredient in the way we experience

{{00:06:05}}
financial services and won't even
realize the ways that it's combined.

{{00:06:09}}
William Mills: Sarah, really appreciate
your time, your contribution to Finovate

{{00:06:13}}
and fascinating topic and looking
forward to hearing more in the future.

{{00:06:17}}
Thank.

{{00:06:17}}
Hi, I'm William Mills with Sarah Hinkfuss
here at FinovateSpring in San Francisco.

I'm here at FinovateSpring
in San Francisco.

I'm pleased to be joined by Sarah Hinkfuss
who's a partner at Bain Capital Ventures.

Listen, I saw your fireside chat
this morning on generative AI,

For those folks that weren't
at that panel, what was the key

Sarah Hinkfuss: Yeah,
it was for 20 minutes.

We packed a lot in, so it was
a great discussion and I was

So Alex, who's from vic.ai and
Lawrence from Slope Pay, who were

really on the forefront of thinking
about how to integrate generative

I would say the main takeaway, like
the most important thing that people

should know, is that generative AI
is a concept that is an ingredient to

But in and of itself, it
doesn't solve all the problem.

William Mills: That's
just part of the solution.

Sarah Hinkfuss: Exactly
one of the many tools.

And I think that matters because
there's a lot of hope for it can be

a platform shift, which I think is
right, but that's only in combination

with great software, with other
types of more traditional AI models.

And so really seeing the entire
workflow and how you get your

entire company to embrace it is key
to the success of leveraging it.

William Mills: That's what I
was going to ask you about.

It's so important for us to
be paying attention to now.

Sarah Hinkfuss: Yeah, so it
hasn't been possible until now.

So just in terms of why now there's
all of this technological change

And the power of it is
absolutely incredible.

And so I would almost turn the question
back to you and say, for anyone

who doesn't think that they need
to care about it, their competitors

are going to, and disruptors in
their industry are going to as well.

And so for every company to actually have
a strategy and a way to integrate it,

because if they don't, other people will.

William Mills: So in what
areas do you see generative AI

So I, I wrote a blog post about this
actually, around the 10 applications

But maybe taking a step back from that,
if you think about what are the use cases

It creates new content, so
generates content, right?

It enables much more efficient search,
so you can look through a lot of

documents and find exactly what you want.

And then it also enables summarization
of many documents as well.

Reasons to be or, types of
work that it does, right?

You can think about those three
applications within the context

So financial services is about
distributing financial products,

manufacturing financial products,
and servicing financial products, and

Content generation
searching and summarization.

And so it's pretty transformative when it
comes to both the ability to generate new

products as well as the cost underlying
the provision of those financial services.

Being in the financial industry
and understand the importance of

How is this gonna, how is a AI going
to influence or change the relationship

with the regulations that more
financial institutions have to follow?

Sarah Hinkfuss: Totally, and this,
is one of the reasons why I think

it's been a more complex topic
within financial services as well.

So if you think about marketing
applications or gaming or

Those areas are, those domains are really
ahead of the curve on generative ai.

But within financial services,
there's a lot of concern around

How will it change my
relationship with customers?

There's a really big cost to
being wrong and sometimes it's

And so for that reason,
there's more reluctance.

So I, think a few things
I would say to that.

One is in speaking in, in my role as
an investor, I get to speak to a lot of

incumbents and a lot of upstarts, and
it's been really exciting to see some

of the creativity around leveraging the
code, like generative AI code within

the context of a paradigm that already
respects the rules and regulations.

So you can actually say instead of the
allowing the hallucination of any message,

you can have the parameters around
what's allowed, like from the SEC in

communication with customers, for example.

That can be also within the context
of that code, so that every message

has the wrapper of what that,
what is regulatorily allowed.

But I would say this part of the world
is very new on, on figuring all this out.

So we're still in
experimentation phase, right?

And so most financial services
companies that I'm talking to, they

They're playing around with them,
but very few things are actually

in production because they're
figuring out what these edge cases

are and how to properly implement
them within the regulatory context.

William Mills: That's
what I'm hearing as well.

FinovateSpring

Best Practice

Trend

Sarah Hinkfuss, Partner, Bain Capital Ventures, breaks down the outlook on generative AI and shares her vision for its place within financial services. 

Stay up to date with the community on https://finovate.com/ 

Find out more about FinovateSpring on http://www.finovatespring.com/ 