10808_AI_Bussiness_School_EXT_FINAL_v3
0 (0 Likes / 0 Dislikes)
[AI Business School]
>> Norm: In talking to
customers around the world,
different industries,
different sizes,
different geographies,
there's one common
question that comes up,
which was,
"How do I start?"
"Where do I start?"
And looking at
all that information,
looking at all those
customers of different sizes,
we've synthesized this back
into an AI maturity model,
which is a framework
to do the analysis
and more deeply understand
where you are
as a customer today
and how do you
get to the next step?
This model is
a phenomenal tool
really is an incredible tool
for you to understand,
as a company,
your level of maturity,
Where you
should be successful,
where you
need to develop,
what are the
kinds of applications
you should
be developing?
It talks about
organizational model,
it talks about governance,
but the most
important lesson for us
is the four pillars
that the model is built on,
where only one of them
is about technology
and the other three are about
strategy,
organization and culture.
It's about engagement
from business leadership
and really
understanding how AI
can actually help
transform a business
and either
transform a process
or in fact transform to
new businesses of new types.
It's a phenomenal tool
to understand and benchmark
where you are and what are
the things you need to do next.
How to build
your organization,
what roles that you need,
what readiness you need,
what are the characterizations
of the applications
that you're running
at those different levels?
And if you understand
what level you're on,
what are the next things
that you have to do?
And so we step back
and we look at, well,
what are the
pillars of this model
that we're
trying to look at,
and we identified
four key pillars.
One is about capability.
That's really about
the technical content.
The technology, the
capabilities of the engines,
the IT architecture
to support and operate it.
But what you quickly see
is that there are
actually three other pillars
about strategy,
culture and organization,
and when we see
companies being successful,
it's where they're excelling
not only in
the technical capability,
but they're excelling
in these other areas
of strategy over all.
Of organization,
of talent management,
of governance,
of testing and learning,
of agility,
to actually take
a customer-centered view
of your data.
A core element of this
would be linkage
to the business strategy.
Can we actually take the
underlying business hypothesis
and tie that to the
work that's being done?
When we look
at these four pillars,
we can then build the core
model around AI maturity,
which starts at the bottom
and which is foundational,
which is customers
either not using AI
or questioning,
"Can AI even help me?"
Where they
might have tried something
and ended up at
a point of disappointment.
They typically will have
a low level of digitization,
a low level of
data-driven decision making.
They probably
are running some BI,
but it's much more
around the reporting
than it is
deep analytics.
When you
move up to the next step,
these are the folks who try.
These are the folks who
are attempting to do things.
They probably
built some bots already,
they've learned about
what a bot should do,
what a bot shouldn't do.
The notion that a bot
actually has persona
is a reflection of
your brand, of your company,
and it's not just
a simple chit-chat,
asking how to
change your password
or what's my balance.
But parallel with that,
we find that
these are companies
that are actually starting to
digitize their process as well.
The third tier,
what we call aspirational,
these are companies
that are being successful.
They've experimented with AI,
and they're
starting to build AI
into the core
of their processes;
in fact, they probably
are changing processes
because of what they've
they've able to do with AI.
They also have a
pretty deep data culture,
but much more than data,
they're starting
to build knowledge
of their customers,
their services,
their products,
maybe even their employees,
and really
trying to understand
this notion of moving from
a data-driven business
into a
knowledge-driven business.
And at the high end,
it's just a
small number of companies
that are very
mature in what they're doing.
They have
core capabilities
around machine learning,
around business understanding.
They probably have
a central organization
that does
strategy and governance
and then within each
one of their business units
have a team of
excellent engineers,
data scientists,
building solutions.
But one of the
observations that we've made
is building that community
from the
central governance organization
into the silos,
that turns out
to be the important step.
It's not that you just
have this distributed model,
but you have this community.
Because what
you learn in one side
is valuable to the other side.
But more interestingly,
what we start to see
is data actually evolving
and being
developed in one silo
can be incredibly useful
to another organization.
If it's between retail
banking and commercial banking,
or it's in a
manufacturing organization
between supply chain,
the actual manufacturing,
and defect detection.
So being able to
navigate across the data
of those silos,
that's when we start to see
this degree of maturity
that we really would expect.
In sort of interesting ways.
So taking this core model,
we then went back
and looked at some data
that we'd got from
270 countries in Europe.
And we looked at their
behaviors against the model.
We found this
amazing distribution.
What you'll find is
about 30% of the companies
sit at that lower tier,
which is that struggling
to understand how AI can help.
The flip side of that is about
4% are in the super mature.
And those
tend to be banks
or manufacturing
companies of some kind,
and in this
clustering in the middle,
of another two-thirds
that actually is either
approaching or aspirational.
The question is,
how do you move along?
What are the things
that you should be doing
to take you from
one tier to the next?
How do you actually
build the capability?
What are the
behaviors that we should see?
What are the examples of
systems that we're producing?
And so what we're
trying to do today
is to actually develop tools
to remove the
friction at that lower end
so that you actually can
adopt SaaS-based solutions
for customer care,
for example,
that allow companies
to actually build a bot
or actually get
into a call center
and to quickly be able
to have an AI application
up and running
and use that application
to augment human ingenuity to
make people better.
[Where do I start?]
So one of the key
observations that we see
to answer this question of,
"Where do I start?"
is, who's engaged?
And one of the
things that we find
is that in any company,
if there is air cover,
conversation and participation
from senior leadership,
it could be the CEO,
it could be
the sales leader,
the marketing leader,
the operations leader.
When they're engaged,
that's when we start
to see greater success.
They're engaged when we start
to see a business hypothesis
being proposed.
In other words,
How can AI help me
understand internal?
How can AI help me
understand
customer demographics
or understand
a new product offering?
That starts with
business-level engagement.
[Start writing your
AI manifesto today]
Every company,
no matter who you are,
needs to develop
an AI manifesto,
which starts at
your senior management
and comes all the way through
the businesses and technology.
This manifesto
is an articulation
about what you
believe about AI,
what you will do with AI,
how you will use the data,
what you will do,
and more importantly,
what you won't do.
And every company needs to have
this framed in some way
before they start moving
up that curve too fast.
That is one of the gateways,
but more importantly,
is actually to take the step
to evaluate your maturity.
Evaluate where you are
and what are the
next steps that you can take.
This all comes back,
in reality, to this
basic definition of AI.
Responsible AI to augment and
amplify human ingenuity.
AI is about
making people better
and making the systems
that people use better,
and to put humans
at the center of AI.
Thank you.