MARC BENIOFF:
So I think we're
beginning to see
the integration
between what's going
on with Salesforce,
And as our customers are
starting to bring that
together, at
the heart of it
You could see in the
example, where certainly
there's a role for
public information
where that public
information is not coming
out of our database,
but we want to augment
But then as we
begin to augment
our own capabilities
inside our app,
we want to be able to
operate at the trust
So to really help us
to understand that
at the next level,
please welcome
Clara Shih is the
CEO of Salesforce AI.
CLARA SHIH:
Thank you, Marc.
And thanks to all of you.
I am thrilled to be
here in my new role,
and to talk to you
about what we're
Not a year from now,
not five years from now,
This month what
we're shipping.
It's really amazing just
to think that a year
ago, most of
the world didn't
know what a large
language model was.
And yet, here we are
today all in together,
with the trust and the
security that you would
We are bringing generative
AI to every application
To our platform, and to
our entire ecosystem.
And we're doing it
in a trusted way.
In a way that's
rooted in the job
to be done, highly
contextual, highly
And the best part is,
from the very beginning
our platform
was architected
for pro code developers,
for low code,
Which means that if you
have a big data science
team and machine
learning engineers,
and you want to train
and tune your own model,
But there's also
a lot of companies
that don't have a
data science team,
and don't have a lot
of machine learning
And those
companies too, can
take advantage of all
of these capabilities.
We're really democratizing
generative AI.
Now, AI Cloud is
powered by Einstein GPT.
And Einstein GPT can
write code for you,
it can generate flows
for you with Flow GPT,
it can write sales
emails, generate
service responses,
marketing landing pages.
A lot of what we've
already heard about
Now the amazing
thing is, if you
think about the
example that Patrick
showed earlier of
the sales email.
That is something
that a sales
person in your
organization
But to get that
quality, it'd
probably be one of your
top salespeople only.
And it might take that
person several hours,
or even several
days to collect
all of the context
about that customer.
Their marketing
interactions,
their service
support issues
that are open,
what they're
doing with their
commerce, what they have
in their commerce basket,
to be able to craft
And with Einstein GPT we
do that in an instant.
And we also do it
for every seller
Every service agent
in your organization.
Every marketing manager
in your organization.
Every commerce manager,
every developer.
Not just the top few that.
Is the power of
Einstein GPT.
Now in March we
unveiled Einstein GPT
as the world's most
trusted AI for CRM.
And since then,
we've been very busy.
Just last week
at Connections
we launched Commerce
GPT, and Marketing GPT.
Last month, we launched
Slack GPT at our New York
World tour, we were here.
And before that Tableau
GPT, and Tableau Pulse
at our event in Las Vegas.
Well, today we're
adding to the mix Sales
GPT, and Service GPT,
bringing generative AI
into the flow
of work where
knowledge workers are
already spending time.
And doing it in a way
that is harmonized
Because of course, first
to succeed with Sales
GPT we need the
context from Marketing,
and from Service, And
from Commerce, and so on.
And as we started
developing these products
with forward
thinking trailblazers
like [? Vasily, ?]
and F1, and Shohreh,
from AAA we'll hear
from in just a moment.
We asked ourselves, what
is the most valuable,
highest impact place
for us to start
What are the
most important
operational challenges,
and bottlenecks,
and opportunities
to drive consistency
and better performance
across each
And as we work
together, we
realized that it was
focusing on those areas
where service
agents struggle.
In service, when there's
a customer waiting to hear
back from you
on the phone.
When they're waiting to
hear back your response
in the chat, it's very
stressful for the service
agent to have to look
up all of these product
And looking through all
these knowledge articles
to find that right
answer for that customer.
The service
agent can focus
on connecting
with the customer
to augmenting his
or her potential,
and really becoming more
than a service agent.
Instead of the
seller trying
to spend all their
time piecing together
that perfect email
to get a prospect
to meet with
them, we automate
that so that sellers can
get out there and meet
Really connect,
and drive value,
AI Cloud drives
productivity,
drives revenue for
every workflow,
every user,
every department,
across every industry,
every segment, and around
And to hear more
about that, I think
we're going to
hear from AAA,
Shohreh Thank you for
being a trailblazer.
Shohreh Abedi is
the EVP, COO and CTO
of AAA, a fantastic
organization
Shohreh, thank you
so much for being
a longtime
Salesforce customer,
SHOHREH ABEDI: Thank you.
It's really
nice to be here.
So I am actually from
AAA, the Auto Club group.
The second largest
club in North America.
CLARA SHIH:
Well, thank you
for being a
longtime Einstein
customer on the
predictive AI side.
Now as we embark
on generative AI,
how are you thinking
about where to start?
SHOHREH ABEDI:
So the way we're
thinking about
it, honestly,
when all of this
hype started,
the first thing
that happened
is my CISO rushing
into my office spooked.
The look of
spook in his eyes
saying, we have to
shut everything down.
You have to shut
everything down.
And I said, well,
wait a minute.
I get it, this thing is
pretty scary in terms
of a tool like this
in the hands of just
But at the same
time, we've
got to be able
to look at this
and see how can
we leverage it
for our benefit,
but sticking
to what's core to AAA with
safety and trust in mind.
So I didn't want
to stifle all
of the creativity
of our people,
because I had people
from underwriting,
from all kinds of
departments coming up
with 150 different
great ideas
that they could
do with Chat GPT.
So that was the
first thing.
But the areas
we're looking at
is three different areas.
One, is customer service,
just like you mentioned.
Just to be able to give
our agents more time
to actually
listen, and hear
the customer versus
searching for things.
Another is in the
middle for support.
I can reduce all of
that processing time,
bring down the
cost, make it more
The third layer is
really for my developers
Having to identify,
troubleshoot,
and also increase our
testing capability.
CLARA SHIH: It's
just incredible.
So forward leaning in
all of these areas.
So you mentioned
earlier, hype.
And there is a lot of
hype when it comes to AI.
What advice would
you have for everyone
in this room, and who's
listening on streaming,
on how to cut
through the hype
and really focus
on results?
SHOHREH ABEDI: So again,
from my perspective
to calm my CISO down, I
said, look, I get this,
but at the same
time you've got just
about anybody knocking
down my door right
now saying we've
got the best model,
we've got this AI
capability or another.
What our approach is,
partnering with partners
that we have in
our collection
such as Salesforce,
and other partners.
Because what we want to
do is, we don't believe--
I don't believe
it's the right thing
to take another
layer and component,
but rather fold
it into the fabric
And that's really how
we're moving forward.
You want to be
able to partner
with folks that have
the wherewithal that
can go shoulder to
shoulder with you.
And that's really
what I'm looking for.
Not the quickest
spin off, or someone
out there that has
the greatest thing,
but then leaves
my back door open.
Because that is hugely
important for a company
CLARA SHIH:
Well, thank you
for your shared
value of your trust,
your collaboration
for your business.
SHOHREH ABEDI: Thank you.
CLARA SHIH: Now, one of
the most powerful things
about AI is
that it learns.
It continuously improves,
or if it doesn't
have the right feedback,
it could get worse
Thinking about model
and how it gets better.
And this is really where
us being the number
one AI CRM really makes
a huge difference,
because there's two
aspects of this learning.
It's called
reinforcement learning.
So the top
section is a term
that probably
many of you have
heard in the industry
called reinforcement
learning from
human feedback.
When a sales email is
generated, when a service
agent response
is suggested,
Do they make
significant Changes
how does that differ
across other members
So us having all of those
workers in Salesforce day
in and day out, allows
us to figure out
what is the
reinforcement learning
How do we use that to
tailor, and improve
the next set
of suggestions
that we give to
salespeople and service
It's a big part of
reinforcement learning.
Well, there's another part
of reinforcement learning
that's very important too.
And that's
reinforcement learning
based on objective
business outcomes.
Because in addition to
being a place where--
Salesforce where we
store your customer data.
We also store your
customer outcomes.
And so we start to ask
ourselves, and create
data pipelines around, did
that sales email actually
cause the sales
opportunity
And ultimately, did
it become closed one?
Did that marketing
landing page,
did that commerce
product description,
did that concierge
experience
cause the customer
to convert and buy
the items in their
shopping cart?
This reinforcement
learning
from human feedback, and
from business outcomes
will ensure that every
customer using AI cloud
will develop
the best models.
Not just today,
but over time.
For every use
case specific
to their industry,
specific to their task,
specific to
their department.
And it's really
exciting area
that our research team,
our engineering teams
have spent a
lot of time on.
And so with
that, let's bring
this all to life with
another customer story.
We heard from
AAA just now.
Let's hear from
Rossignol Anyone
who skis or likes
Alpine sports has
heard of this amazing
company based in France.
And let's hear from their
CEO, real trailblazer,
Vincent Wauters
and his team
on how they're
using AI cloud.
And the combination
of AI, plus data,
GABRIEL AUTHIER:
What I love
the most about
the mountains
is the energy that I
get every time I go up.
You get a feeling of
excitement, of discovery.
Every step you take
opens up a new horizon,
and I think that
is beautiful.
VINCENT WOUTERS: We are
very clear on our greater
GABRIEL AUTHIER:
We want to inspire
people to spend more
time in the mountains.
VINCENT WOUTERS:
We want to carve
movements of
sustainability
GABRIEL AUTHIER:
There's always
a very strong
emotional connection
Nostalgia, anticipation,
excitement.
VINCENT WOUTERS:
But if you
want to create
and foster loyalty
with our consumer, we
need to personalize
Salesforce is a
very precious power
GABRIEL AUTHIER: This
is a unique turning
The seasons are
getting shorter,
and it's going to
become more challenging.
VINCENT WOUTERS: We
are at a crossroads.
We are progressing
to a new phase
where we need to deploy
the full emotional
connection with
our consumers
GABRIEL AUTHIER:
This brand
is really the oldest,
most recognized ski brand
VINCENT WOUTERS: Rossignol
represents the beating
GABRIEL AUTHIER: We
make skis for beginners.
We are making skis for the
highest level athletes.
When you have
such a wide range,
it is a curse
and a blessing.
How do we reach our
consumer with the product
that they actually
need in that moment?
Customer 360 is really
about understanding
the desires and needs
of our audiences,
and how Rossignol
can deliver a better
VINCENT WOUTERS:
When that is clear
we express ourselves
with more authenticity.
And through authenticity
we create an inspiration.
GABRIEL AUTHIER: The
world is changing fast.
VINCENT WOUTERS:
And now we
have almost an infinite
amount of data coming
from all the touchpoints.
Salesforce help us collect
those data, analyze them,
create efficiency,
create personalization.
GABRIEL AUTHIER: And now
with AI with Salesforce,
we are able to really
engage with our consumers
seamlessly across seasons.
They're going to get
the proposed the right
product that
inspires them.
Our CRM, our data,
together with AI
is going to create a
circle, a virtuous circle
that is going to elevate
the consumer experience.
VINCENT WOUTERS:
The partnership
with Salesforce
started 12 years ago,
and now we have an
exciting new chapter.
GABRIEL AUTHIER:
This is going
to be a new
era for skiers,
and snowboarders, and
bikers around the world.
VINCENT WOUTERS: I believe
that we can be agile
Salesforce will
help us to adapt
and thrive in an
ever evolving world.
CLARA SHIH:
Incredible story.
Sanjna, please help
us bring it to life
by showing us how
Rossignol uses AI cloud.
SANJNA PARULEKAR:
All right.
Back for another
amazing demo
of how Rossignol
is using AI
cloud to deliver these
incredible experiences.
Now Rossignol cares deeply
about their brand voice,
and tone, and
bringing that
to every personalized
recommendation
they bring to
their customers.
And they're thinking
about launching
this new heretic
bike, which
they know will be a hit
for their customers.
So it all starts
within Commerce GPT.
Now, this helpful
Guidance Center
here with the
next best action
is going to help me with
that checklist of what
I need to launch a
new product in market.
And it looks like
we're actually missing
We're missing some
product descriptions.
And just like I said
in the previous demo,
creating these
descriptions,
or creating a marketing
email, a sales email,
this takes a lot of time.
And it's no
different here.
But Einstein GPT is here
to help by automatically
generating product
descriptions
that are not
just descriptive
of the product, but
that are actually
in the brand
voice and tone
that Rossignol wants
to bring to their end
Now there's a catch here.
Rossignol is also
a global company.
So it's not enough
to have these product
descriptions
generated, they also
need to be localized
for the languages that
are most relevant to
their customer base.
In this case, we want
to start with Spanish.
So we can translate all
of these descriptions
with Einstein GPT in
the click of a button.
Now normally, this would
take a lot of time.
Creating the
descriptions, getting them
to the local teams,
getting them translated,
and then taking it that
extra mile of inserting
the voice and tone
of that language
It sounds painful,
because it is.
But within a few
clicks, Einstein GPT
can help them
bring this bike
to market with this
commerce experience.
Now once this
bike is launched
and it's out there, it's
flying off the shelf.
People are so into
this new bike.
And all of our
brand managers
globally want to
make sure that they
have the right
understanding of how
their KPIs are being
affected by all
Now, these powerful
visualizations
you see here
would normally
take a lot of data
analysts, a lot of work
behind the
scenes to create
every time you
want it to refactor
some of the elements
of the dashboard.
But with the new
Tableau pulse,
this is available
for business users
to keep track of their
business as it happens.
It's keeping up
with the speed
of how things change,
all in a trusted manner.
And even if this
dashboard isn't enough,
that business
user doesn't need
to go back to the data
analysts for changes.
Einstein has your
back here as well.
You can actually
just ask a question
in natural language
here, and Einstein GPT
will generate this
additive visualization
here with a description,
answering your questions.
So you can get
on with your day.
No need to meet with
those data analysts
to generate this,
and wait a long time
Now, another element that
Rossignol cares deeply
about is the way that
they service their end
And there are two
key reasons here.
One, they want to make
sure that their service
time, their case
resolution time
is extremely short
for every case.
And two, they
want to make sure
they have happier
customers.
So this is where Einstein
GPT within Service
As this company
question is coming in
through the
service terminal,
Einstein is providing
these helpful
recommendations for
me as I go along
Now, the service
manager is always
They can accept,
edit, or decline
There's always a
human in the loop.
But just the
sheer fact that
these recommendations
are in my place of work
as a service manager
is a huge time saver.
It's keeping me
from chair swiveling
between different
knowledge articles,
searching through
my whole knowledge
base for that perfect
answer for the customer.
And as the
conversation goes on,
I even have Next
Best Action giving me
those relevant,
personalized product
recommendations for
that end customer.
And the cherry
on the top here
actually, is that as this
conversation is going on,
once I close the
case, I can easily
summarize it and
share it with the rest
So for example, if
our field service reps
are out there
in the field,
they're not
behind a laptop.
They can access
this best practice
in the form of a
knowledge article
that is shared amongst
the whole team.
Now, I hope between
these two stories,
these simple stories,
it's got your imagination
thinking a little bit on
how you can get started
The possibilities
are really
endless when you bring
AI Data and CRM together.
CLARA SHIH:
Thank you Sanjna.
So so many of
you have been
asking how you can get
started with AI Cloud?
We talked to
our customers.
And we sat down
with Gucci, and AAA,
and Rossignol, and the
Royal Bank of Canada.
And we realized that the
world's leading, best
in class, AI first
companies were
using a number of
solutions together.
Having AI in the
flow of work.
Where salespeople, where
service agents work.
Marketing managers,
commerce managers,
developers,
business analysts.
Both for the reinforcement
learning, and also
so that we can automate
the prompt engineering
for their specific
job to be done,
for their industry, for
that specific customer
Then we saw that the
best in class customers
are using Slack,
because it's
Once we have those
generated outputs
it's not about
individuals.
It's about teammates,
coworkers coming together
and saying, how do we take
what the AI has produced
and make the
landing page better?
How do we make the
commerce description
Let's work together
across teams.
How do we use
generative AI
to summarize all of
these conversations, all
this knowledge
that's sitting inside
of Slack that represents
our organization's brain?
Then we see these
organizations
using Tableau
and saying, well,
if I want to dig deeper
and really understand
the data that's being used
to train, and fine tune,
and ground my prompts, how
do I explore that data?
How do I have
a conversation
with that data through
natural language?
And of course,
Einstein, we've
been talking about both
the predictive aspects
of Einstein, as well as
now the generative aspect
And then very
important is the data.
How do we get
that data from
Both data that's in the
cloud as well as data
that is on premise,
and use MuleSoft
And use Data Cloud
to unify the data,
to harmonize the data,
to create data pipelines
that can be used
in any model
But it's not just
the technology,
it's also a tremendous
change management
And that's where
Salesforce services
Able to partner
with you to skill up
And I'll talk
about what we're
doing with our
trailblazers
But everything that
you need to get
to a proof of technology.
To make sure
that it complies
with your company
governance,
with industry regulation.
And then to go from
there to coming up
Which use cases are
you going to start?
Will it be similar to
what Shohreh is doing?
Maybe it's different
for your organization.
We want to help you
figure that out.
And so Salesforce
services is here to help.
As well as amazing AI
first partners like Julie
and her team at Accenture.
Our partner at
Deloitte, PWC, and more
So we're excited to
help you get started.
And as I mentioned
earlier, all of this
Our engineering, and
product and research
teams have been
busy at work.
They've been burning
the midnight oil
This is going to be
the summer of AI.
It's going to
be even better
And we're going to have
sales email generation.
We're already
piloting that
We're going to be
shipping service reply
recommendations to augment
your service agents,
just like [? Vasily's ?]
team at Gucci.
We're going to be
shipping summarization.
You just saw that
demo from Sanjna.
Every cloud busy at work.
We've mobilized the entire
team, the entire company
And helping ship
that, and get that
into the hands
of our customers.
And of course, any
change management
It starts with
your employees.
It starts with
our employees.
And so we're also
thrilled to be launching
today, a new
series of courses
It's our free online
learning platform
that any of you, any
of your employees
can start signing
up for today.
To start earning, we
have over 35 AI badges
Everything from
the very basics of,
How do you approach AI
in a responsible, ethical
way that honors data
security and data
All the way to the
most advanced courses
around training, and
tuning, and prompt
Some of the more
advanced concepts
So please check this out.
This is this is going
to take all of us,
and all of our employees
to reskill, and learn
