Right, so far this morning we have heard
about rapid progress in AI, the future world of agents
and ways in which businesses can use AI
to realize meaningful value
across more than just the bottom line.
However, we also know that for many organizations,
AI projects have simply not delivered desired returns.
Some never get beyond pilot stage
and many struggle to scale.
With this in mind, our penultimate session
of the day is going to deep dive
into a successful AI implementation,
highlighting what it really took
to overcome barriers, achieve success and scale AI value.
Joining us to host the session is Wiem Sabbagh,
Vice President and Global Microsoft Consult Leader
at Kyndryl.
Wiem played a key role
in launching Kyndryl’s AI readiness program
in collaboration with Microsoft,
supporting organizations in building strong data foundations
and deploying enterprise grade generative AI.
Wiem is joined by Anders Bjornrud,
CTO of Care Safety Innovations,
and Kate Rosenshine, global Technology Director
of strategic partnerships at Microsoft.
Over the next 20 minutes, Wiem, Anders
and Kate will be bringing to life
how Care Safety Innovations with support from Microsoft
and Kyndryl turned ambition into real business value
and we’ll be sharing what we can all learn
from that journey, over to you Wiem.
Thank you Catherine.
Kate, Anders, a big and warm welcome to you both.
I’m super excited to be here with you.
Today’s session is all about
how we can turn Gen AI from big ideas
to real world impacts and results
and with the right vision, with the right technology
and the right partnerships, we can make that happen
between Care Safety Innovations, Microsoft and Kyndryl.
We have a fantastic story to share, so let’s dive in.
Perhaps let’s start with you Anders.
Can you tell us more about Care Safety Innovations
and what were the business challenges
you were looking to solve and how did it lend itself to AI?
That’s a good question and thank you for having me.
Care safety evaluations is focusing
on a really big challenge
and that is care in homes in the UK alone,
it’s 1.7 million care takers in the UK alone
and we have even less
and the scale is going down as well
for the people who handles that care.
So we are focusing on the challenge
on how to deliver safe care, more efficient care
and focus on the human part.
So the big question we asked ourselves
is how can we adopt to that and help them do that?
Amazing, a lot of organizations today talk about AI,
but not so many make that leap from idea
and proof of concept into production
and Care Safety Innovations managed to do so
and you have an up and running
AI enabled solution implemented.
So what were the factors that helped you do so?
First of all, we started off by thinking
that AI should be part of the foundation,
not something that just should come in step three
of our application.
On top of that, we need partners to do so.
So we figured what would it take to get to that point.
So we started with the business side.
We started with the main pain point
for the care providers in the UK.
We partnered up with a few of them
and specifically one down in Bournemouth.
We got a lot of input and then we configured
and developed a platform
and from there, we attacked the market
to be able to deliver to even more.
But the key thing is to focus on the challenge
and on top of that we partnered with Kyndryl
to have the competence to do so
and build this platform with AI
to take it from, you know, pilot to production
and even Microsoft with the Azure platform
so we can build scale and have it secure
and compliance all the way.
So the key thing is to have partners
and do it the correct way.
Thank you Anders, Kate, over to you.
Microsoft clearly played a crucial role
into these partnerships.
Can you share more about that and share how did Microsoft
and Kyndryl helped Care Safety Innovations in scaling AI?
Yeah, sure, thank you for having me.
What I love about this story is that you came
with a very clear business objective,
a clear problem statement
and an idea of how advanced technology
such as AI could really plug into that.
And what was great is, you know, from the Kyndryl side,
you bring in that consulting,
that sort of longer term business vision if you will,
and not just treating it as you know,
a POC that will end up going nowhere,
and the graveyard of many POCs,
but really being thoughtful about it
and looking at it as a long-term business outcome
and not just a technology project.
Where we came in is sort of powering it
with the Azure platform
and really working together
to make it an enterprise grade solution,
especially in an area that’s quite sensitive, right,
there are data considerations,
there’s infrastructure considerations using AI responsibly
as well, and bringing that all together,
you know, hopefully the results speak for themselves.
But it really starts with that idea, you know,
we can power with the technology,
but if you don’t have a clear idea of what that ROI is
and what your business objectives are
and what your customers want ultimately
or your end users,
if you think about it slightly differently,
it’s not gonna be very successful in in the longer term.
So that’s where I think it sort of tied together
really well and each company brought forward
their capabilities to achieve that.
So full AI stack
and strong platforms with the right vision,
with the right skills, we managed to do so.
We’ve mentioned already how hard for some organization
to go past that PEC stage and rapid prototypes
and even idea in theory.
Anders, as you were moving to production,
what were the challenges that you faced,
whether organizational, cultural
or from a technical standpoint
that you faced a little bit for that like stage
to go past that and implementation?
Yeah, so first and foremost we have been talking
a lot today about adopting to AI and how to use it
and how to use it correctly.
And the other part is behind the scenes.
So how can we create that platform,
what competence do we need?
Do we have the right people, do we have the right talent?
All that to build the platform
and then being able to adopt our users to it.
And secondly, it’s all about winning the minds
because we are met with a lot of carers
and also other generic IT people which is using AI
that says, are you going to replace me?
That’s the big question and we are saying no,
we will not replace you
because we would like for you to have more human focus
inside the care visits so we can speak
and they can speak back to the carer
instead of the carer having to go, you know,
all the digital tools all the way.
So we prefer to think ourselves as helping with that.
So, but again, it’s a lot of effort to to get there,
but when you get there, it pays off.
[person laughing]
Absolutely, there is like the design,
there is the change management, the culture,
it’s not just the the technology.
Correct.
Kate, Anders spoke about the importance
of having the right foundations in place.
Where is Microsoft providing companies
to help them scale AI?
Yeah, so first of all, I love that sort of human element
of technology and I think ultimately,
like that is the goal, right?
So how can people work with technology better
to get outcomes that allow them to do things
that are inherently human, such as social connections,
judgment, et cetera.
I think from a Microsoft perspective,
we come at it from different angles.
So first of all there’s the skilling
and I think someone touched on it before,
I don’t remember who, but we were doing a lot
in this space, obviously not just for companies
but for greater society because ultimately,
these are the people that will then go
and work for these companies in one way or another.
So really thinking about it more holistically,
there’s the business side where we partner
with, you know, partners such as Kyndryl
to really tie that together, right?
The capabilities, the skills, the thoughtfulness,
and finally is the technology piece, right?
So how do you think about technology
in a fast evolving pace, which you know,
not just from an AI perspective,
but all the capabilities that need to sit around that.
So I think each company tends to be unique,
but ultimately it comes down to the skills
and the understanding and not just technology skills
but understanding how
and not just the what, it’s the business requirements
and finally it’s the technology itself
and really thinking about it more in the longer term.
Love that, it’s really holistic.
One of the things that I keep hearing a lot
from leaders like yourself Anders
when I speak to our customers is the ROI question, right?
Whatever there is an AI investment,
the question is, where is the ROI?
So within Care Safety Innovations,
where did you see the value of AI
and how did you unlock that value?
And more importantly as well
is how do you measure the value of AI?
That’s a huge question.
Yeah, absolutely.
[Speaker] The million dollar question.
And we’re still developing how to do the KPIs
for measuring ROI,
but I think first of all we need to think
about the efficiency out there in the field
because we are operating in a healthcare sector
in the UK, it’s really low profit business
so we are measuring what they are doing,
how much time they can use to do something else
that is revenue stream for them
instead of doing administration.
So we have seen examples
where carers actually use 40% less time
for their tasks with our tool, the Vita Care Companion,
we call it, the VCC.
And secondly, we also measure
the time for the carers to actually,
in every care visit they have to produce
something called care note
and in the end it’s a legal document
if something were to happen.
And that’s what we are doing with the AI platform
because we are giving that an objective sort of way
instead of the carer writing whatever they want.
And that has taken the time down from five minutes
even more, down to one minute, maybe one and a half minutes.
So that’s a big, big thing for them.
So freeing up time for the caregivers
to focus on the human part,
which is very important, I love that.
I would love to also add the third thing
because we are actually being transparent.
So we are giving access, we are not giving access
basically because ESI
and Care Safety Innovations can’t see anything
but the providers are giving families access
and maybe other third parties as councils
so they can see everything inside the family portal
instead of having to call for every little bit
of thing they want over information from that care visit.
So more transparency as well.
Yes.
Love this.
We’ve heard earlier from Azeem but also Patrick
and Ismail, the true evolution of AI
and that’s like really impressive
how fast moving this market is.
If you were to give one piece of advice for the leaders,
for the companies that are looking to really go beyond
that rapid prototype and ideas and theory
to the skill and implementation, what would that be?
I think I touched on it earlier,
but I think start with the challenge
instead of the technology because you need to figure out
what are keeping people up at night
and then you need to figure out how you can build a solution
to help assist on that problem.
Love this, any other tip you wanna add?
I have a lot.
I said one.
You said one, I probably have more,
but it’s doing it the right way.
So instead of figuring out you need this today
and don’t think about tomorrow,
start thinking about tomorrow as well
because when you get to tomorrow,
you would probably not like to do the rework, so.
Absolutely.
Do the steps one at a time.
Love this, maybe now for you, Kate, again,
in this fast moving market, where is your advice
for companies to stay ahead of the curve?
Yeah, so I think I’m just gonna sort of echo
what you were saying
and it is really understanding your business,
your customers, your future customers as well
and how in the longer term technology plugs into that
because the space is evolving so fast, like super fast.
So I’d love to be a futurist
that can predict like what’s gonna happen
in even six months time,
but thinking about it more holistically
and being prepared for the evolutions that come
and also being thoughtful about it.
Not every single part of the organization
is a good candidate for plugging in generative AI.
I think there’s other tools that are applicable.
So really understanding how to use it in the right way
and thinking about it long term
are the two elements along with, you know,
treating it like any sort of change, right?
So the skills, the capabilities,
the training that you have to give your employees
but also, you know, wider society in that sense.
Thank you Kate.
Thank you both for sharing these fascinating insights.
I think we have some time
to take some questions from the audience.
So first question is, where is your position
regarding this phrase?
Your job is not at risk from AI,
your job is at risk from a person
who knows how to use AI, Anders, Kate?
Should I start?
Yeah, you can start.
So I think our focus should be
on using AI to help us do our jobs
instead of having AI do the job
because there will always be this human part of it.
Yeah, and I think the good news is
it’s becoming a lot more accessible.
So not to age myself,
but I learned how to code on a piece of paper
and that would be marked by a teacher.
Now there are all these tools
that really unlock that creativity
and you don’t need to, like you need
to sort of understand some elements of it,
but you know, people can get into it
and adopt a lot quicker than before.
So I do think it’s easier to get started.
The interface is more linguistic,
it’s more interactive than what computer systems were,
I mean even 10 years ago.
So yeah, that’s what I would say to that.
So get your hands dirty
and to be honest with all the low-code,
no-code capabilities as you mentioned,
it’s so easy now to really ramp up
and get to use that and know how to use it
and become native whatever you are in the organization.
The second question is, what does it take to move out
from the POC phase and move to the production?
My enterprise clients are already well aware
of the potential of Gen AI
and now they expect consultants to help with the next phase.
What do you say to that Anders?
I think the really important thing
to keep focusing all the way from idea
into proof of concept,
into pilot projects and then into production
because if you lose focus then you’ll not be able
to put that POC into production.
I think focus, strategy
and having the processes is really important.
Perfect, there is actually another question
for you Anders.
Can you tell us more about your solution
and what it does?
Is it a physical bot, a screen, a voice?
Where is the interaction with the patient?
Yeah, so we are not trying to have any communication
directly to the caretaker from our platform.
We have some sort of a mobile device hanging on the chest
and we have the back office and the family portal
and all that stuff in the web as well.
The thing is that we are going into home care visit
and we are documenting everything.
So we are recording a video once we go inside
to catch whatever the status,
how are you all that kind of stuff.
We are documenting with video as well.
The medicine administration,
we are documenting if something critical were to happen,
if they have some kind of wound
or need something in particular.
And also the carer are able to connect
with the back office directly
so they can push this button called online assistance
and then if they are unsure what to do,
they can actually ask the back office, what should I do now?
On top of that, we’re also documenting that last stage.
So when the carer leaves the care visit,
we document, how are you now?
Everything happened as it should.
And then we make a transcription of that
into the objective care note being the legal document
that is required by law.
Thank you for sharing, I actually saw the video on this.
It’s pretty impressive.
So congratulations on that.
Anders, Kate, thank you so much for joining us
and sharing all this exciting insights on this partnership.
I’m gonna now hand it over to Catherine.
Thank you.
Thank you. Thank you.
