How did you get
to work today?
By bike maybe, on
foot, on the train?
Well, what if a
driverless car took you?
Autonomous technology
like Waymo's fleet
of driverless cars
is becoming part
And on this episode
of Inside Dreamforce,
we're exploring what
autonomous agents are
and how they
will transform
your company and your
customer relationships.
Autonomous
agents don't need
conversational prompts
to take action.
They can anticipate,
plan, and reason
They work right beside
you as an extension
Agents are the next step
in how your organization
can use AI to achieve
real business value.
Now, you probably have
a lot of questions
After all, this is another
huge technology shift.
Like what is an
autonomous agent?
What does this mean
for your business?
Well, we're joined by
our Salesforce experts,
Jessica Geutner
and Patrick Heinen
But first, let's hear
all about Agentforce
When I think about
Salesforce today,
I think about it in
three layers, maybe four.
First is I think about
for the last 25 years,
We've been automating
every customer
Sales and service and
marketing and commerce
and analytics and
Slack and anywhere
your customers coming
in, we're connecting
and we are making that
part of our Customer 360.
Two, we are also
building a data cloud
so that you have
a data lake that
is able to absorb
all of that customer
connectivity,
including if you have
multiple Salesforce orgs
and federate that data
cloud out to not only
all your Salesforce orgs,
but to other types of
data lakes like snowflake
and databricks and Google
BigQuery and Amazon
Redshift and even
IBM mainframes.
And then we're able to
take all of that data
and bring it down
into the Customer 360.
And this has been our
vision for the last 25
But now, what
we want to do
is wrap that with
agents and make
you even more productive
and more capable.
And it's an extension
of our core platform.
So you can have health
agents and banking
agents and product
agents and ops agents
and sales agents
and support agents
and marketing agents
and customer experience
agents and analytics
agents and finance agents
And it's all built on
this Salesforce platform
with all the trust and
security and ecosystem
And this means that
our architecture now
And this is not what it
looked like the last time
we were all in this room.
It didn't look like this.
This isn't where we were.
We had a different
vision of the future.
Customer 360
and Data Cloud
was our best selling, most
exciting new product ever.
But agents are
really changing us.
They're giving
you that ability
to have something that's
trusted and secure
and scalable and
accurate and easy
to customize with
this built-in AI.
And the Customer
360 is built in,
the metadata and the
data are built in,
the open ecosystem
are built in.
And you buy nothing
else from anybody
It just works
with accuracy
And some of the
hypnosis that we've
had around AI that it
doesn't work exactly
like that or you have to
DIY it, you know, DIY.
Do you all know
what DIY is?
I thought everybody
knew what DIY was,
Maybe it's a San
Francisco term.
DIY means do it yourself.
I was talking to my
friend Neil Young,
and he knew exactly what
that was because he said
But DIY means I'm just
putting it all together
But I don't think
you can DIY this.
I think you want a single,
professionally managed,
secure, reliable,
available platform.
I think you want the
ability to deploy
this Agentforce
capability across all
of these people
that are so
important for your
company and all
of these digital
and human workers.
And the reality is,
we all have struggled
in the last two years with
this vision of Copilots
and LLMs and how are we
putting it all together.
So why are we
doing that when
We can move from
chatbots to Copilots
to this new
Agentforce world.
And it's going to
know your business
and it can plan,
it can reason,
it takes action
on your behalf.
Welcome to the
show, Patrick.
Let's get honest
for a second.
The speed at which
AI has developed
over the last
few years has
There's trust and
privacy for one.
There's proving ROI and
identifying use cases.
And there's the question
how to use it effectively.
How to find the best
ways for humans and an AI
So I would say we
are just in front
of the third
wave, which means
we can get autonomous
agents working for us.
And that's, of course, a
big thing for Salesforce.
We just announced
Agentforce.
And also for the companies
because these agents can
not only make you
more productive,
you can put an
agent 24 by 7
in front of
your customers.
And can you give
us some examples?
Yeah, you can have
multiple agents.
You can have a
sales agent which
is supporting
your sales teams
You can give an agent
an enormous amount
of knowledge and
then the sales teams
can ask the agent
nearly everything.
And an agent can
do all these tasks
for your salespeople,
which they don't like,
which they don't
have time for.
And it works
completely autonomous.
And one important thing
is you can set also
guardrails to an agent
so that an agent is not
doing things which you
don't want them to do.
I like that tasks
I don't like to do.
But you shared what
autonomous agents are
and how you can use them.
But how do Agentforce
agents actually work?
So you need to
understand that there
are five important
steps that you
So first of all, you have
to tell an agent the role.
Are you a support
agent, for example?
And then you need to
tell the agent what data
Of course, we can
connect all the data
And the data can
also be transferred
And once you have
done the data part,
the third thing
is the actions.
Agent, for example,
in a support agent
can look after
your recent orders.
So multiple tasks can
be trained to an agent.
Then it's important
to set the guardrails.
So, for example,
if a support agent
should give a
discount, the agent
needs to know how
much discount it's
And this is a very
important step
because most AIs are
trained to give you
an answer and to
find a solution.
But giving guardrails
is very important
so that you don't give
the wrong answers.
And the last
step, of course,
is the action
and the channels.
So in which channel do
you want to use the agent?
Is it through
your internet
portal where you might
have a short question
and answer agent
running in front
So we are supporting
multiple channels
And these are the
five important steps
I like those
agents already now,
but it's a very important
for organizations
to really understand, what
are the steps they should
take to get started in
implementing autonomous AI
agents in their
organization?
So the good thing
to understand
is that Salesforce is
following its unique path.
So we are focusing
on low code.
We want to get rid of
all the complexity.
Think about you're
building this up
You need to
understand LLMs.
You need to put
multiple LLMs together.
So these very
complex steps
we have put into
our platform
so that our customers can
customize their agents
with low code technology.
So they just need
to point and click.
And then it's
important to focus
on the right use cases
and the business value.
Of course, it's nice to
see what these agents can
do and what different
AI capabilities
you can use today,
but you need
to understand what's
really the benefit
for your customers
or for your company
or for your employees to
support them with agents.
And you need to
start from there.
And then you think about
the right use cases
and then you concentrate
on the technology.
Thanks so much
to our guests
on this episode of
Inside Dreamforce.
And thank you for
sharing your time with us
and joining us
on this journey
If you want to learn
more about Agentforce,
Data Cloud, and
how your company
can benefit from
autonomous agents,
then I've got some
great news for you.
We're heading to cities
across Europe, the Middle
East, and Africa with
a series of events.
They're your chance to
hear about the latest
product innovations,
get hands-on
with in-depth demos,
network with your peers
and Salesforce experts,
and hear from trailblazing
You can find your
nearest event
by scanning the QR
code on screen now.
And for now, tune in
to the next episode.
