Introduction

Setting the Stage and Audience Check

I think I can actually build on what both of you said,

and specifically what you started talking about.

How do you actually bring this into larger organizations?

How do you make that part of teams, routines, etc.?

And I want to share a bit about our own journey.

But before I do that,

I'd really like to understand what type of organizations you're from.

Who's in a company that's, let's say, up to 20 people?

A few.

Whose company, 20 to 50 people?

Okay, 50 to 100?

Okay, more than 100?

Quite a few.

So a good mix.

Okay, cool.

Who We Are: Lufthansa Innovation Hub

Who knows Lufthansa Innovation Hub?

Hey, cool.

Quite a few.

So

Our Mission and Capabilities

we are a bit the, I like to say, Swiss army knife for innovation.

So we do a lot of different

things in our capabilities.

We are 60 people, just right around the corner here in Berlin.

in.

And we do strategic intelligence, foresight, so really trying to understand what's happening

in the broader ecosystem in the market, spotting opportunities.

We do a lot of corporate

venturing, venture building, venture clienting, but we also have our own transformation team,

which is trying to really look at how can we support the transformation of Lufthansa towards

a more innovative company.

And that ranges from culture to technology.

AI is a big part of that

as well.

My Role: Bridging Hub and Group

So my role is a bit two -folded.

I have a role within LIH, really trying to constantly

push ourselves forward in terms of what we do, how we operate, the skills, the expertise

that we have.

Again, AI, a big part of that.

But I also have a role towards the Lufthansa

group, trying to kind of push the larger mothership forward.

And that's what I'd like to share

today what we're doing there and how we're going about and I always love to

kind of stand here and just share that we have it all figured out we don't it's

a significantly difficult journey to get people to do this because it's not a

technology game it's a people game it's an adoption game and we've been part of

your journey where you've started actually and helped us get on this

journey already last year and since then we've been kind of going through this I

The Pace of Change: Technology vs. Adoption

I always like to start, actually, I had this slide up for a long time now, and I think it really captures what we're looking at right now.

Near-Term Hype and Long-Term Underdogs

We always overestimate the change that will occur in the next two years and underestimate change that will occur in the next 10,

which really speaks to our struggle with really understanding kind of the compounding effects that specifically technological progress right now has.

and I actually would change it because we think also underestimate when it comes to technology

the progress that's happening in two years even if you look down kind of where we were six months

ago with respect to a lot of the different capabilities that all of these tools some of

which were demonstrated today have to today it's incredible where it really diverges is the pace

that technology has and the pace that us as people as as organizations have in adopting all of this

which is kind of the the conundrum we're facing and these are actually a couple of slides that

we sometimes use to talk to our Lufthansa group colleagues sometimes internally but we are I think

From Pre-AI to Agentic AI

that was also mentioned before really thinking this is a fundamental change in how organizations

will work how we will work and I mean we're trying to be very descriptive here from kind of pre -AI

where it was all about people when you wanted to scale you had to scale people to AI really as an

efficiency layer when it comes to gen AI which I think we probably are still in

right now really augmenting the cognitive work that we're we're doing to

really then in the future agentic AI where it is not only about assistance

but really delegation so really that is going to fundamentally in the next

couple of years change how we operate in organizations and I haven't brought it

but there's a really cool picture from that I saw on LinkedIn recently from the

entrepreneur professor from VHU, the German University, and he depicted his

team as five humans and three robots.

Because they started to really build

agents and they start to look at them as complement to their team.

And this is

what we're gonna be looking at in the next couple of years.

Really that

obviously changes also how people operate, how leaders operate, how we

manage teams in an organisation.

What’s Already Possible Today

This is kind of the narrative more on a broader level.

And we're already starting to see all of that.

You kindly demonstrated what is possible today

with connecting a couple of different tools, which I'd actually like to know how it works.

But you all have those, kind of whether it's your typical chat GPTs, enhanced intelligence,

automation and agents, whether it's AI coding, no coding, all of these tools that you can

use today, creative augmentation, and I'm really going to go over a couple of slides

really quickly because I want to get to the meat of how do we actually do it.

Shifting Skills and Roles

And this is going to change roles, competences, skills, people, whether it's kind of, I think

in the, already today, and I think that's what Ferdinand, you mentioned, never start

at zero, I think that was what it was called, there's going to be always -on collaboration.

You're never going to not collaborate.

collaborate, maybe not with a human but with some sort of tool.

Accelerated task proficiency

specifically for more junior roles with the help of all of these different tools, you're

going to be changing skill profiles.

Everyone can be a builder.

You're not going to be able

to build enterprise ready applications if you don't know how to properly code, but you

can still do much, much more than you could six months ago, 12 months ago, 18 months ago.

go.

And that obviously changes organisations.

So this is just a set -up of some of the things

that we believe in.

I specifically spared you from the latest predictions from McKinsey

on adoption statistics or anything or impact of AI.

But really this is some of the perspectives

that we have.

Turning the Lens Inward: Our Activities

And then with that in mind we obviously looked at ourselves.

So this is

a bit of some of the activities that we

have.

I talked about our

intelligence team, analysts, research

that's the left side column.

Intelligence, Venturing, and Transformation

Really create research reports

insights, re -curate them in

newsletters, share them

widely externally

but also within our Lufthansa group

environment.

We have our corporate venturing

realizing the next

so partnerships with startups

and other tech companies

building new digital innovations

investing less so currently and then we have our transformation team which really

Looks to enable the next which is a lot of kind of engagement around the topic of innovation learning formats trainings

etc.

So when we

We looked all of that.

Where AI Augmentation Fits—and Where It Doesn’t

There's a lot of stuff that lends itself to some sort of augmentation

Whether it's ideation on all kinds of levels

built, hypothesis testing, MVP building, coding, marketing communications, training design,

maybe facilitation unless we have humanoid robots is the one thing that doesn't lend

itself to it, but otherwise there's a lot of stuff that actually lends itself to it.

Why Individual Wins Didn’t Scale

But what we've also seen, adoption has been often driven by individuals.

individuals who are, we have so many really curious exploratory people, but we didn't

really have a strategic approach to how we manage that.

So we've had people who build

AI -based, no -code solutions to build an intelligence platform.

So pulling together all kinds of

different data sources, putting a platform layer on top of it, really experimenting a

lot.

We have people who are very, very sophisticated in our research team on prompting and really

pulling together information from different perplexity, Claude, Gemini,

whatever, into really powerful, powerful tools and speeding up their process of

research, synthesis, documentation or document writing, etc.

We never got this

to really scale across the company.

And where's those 60 people?

So it gets

infinitely harder if you kind of look at 10 ,000 people.

But I think we serve as a

good example and there's multiple reasons for that and hope I think you

might know them so operational workloads it just takes time to get some of those

things working you have to kind of dedicate time to get an invest return

investment of your view on your time sometimes it's being overwhelmed of all

of these different tools who are constantly changing so more on an

individual level sometimes on a more team or organization level it's about

Do you kind of have the team routines to actually talk about this?

And I always like to talk to ask the question like who has in the last two weeks spoken about AI with within your team

Okay, cool.

That's why you're here at the event

From Ad Hoc to Strategy

So we like some sort of a strategic approach

How do we start the journey actually you were part of it

that was already last year but after that actually it didn't really take off we did not all build

these types of fancy agents and use of our llms were still sporadic and there were a couple of

reasons for that and then earlier this year we started basically our approach a bit more in a

bit more structured way to really augment our organization and give a lot more attention and

and drive to this, because we believe it really

fits our strategic narrative.

So we are an innovation hub.

Number one, we should be using these tools

and should be at the forefront.

Number two, we want to be lean as part of our strategy.

And obviously, these tools help you

to do more with the same amount of resources.

And then there were a couple of other reasons.

Adopting a Deploy–Reshape–Invent Framework

And we really shamelessly stole this from BCG,

who had this really nice approach of,

how do you go about actually implementing

that in your organization?

And they talked about deploy, reshape, invent.

So really getting AI and Gen AI into your everyday work.

So the simple stuff, reshaping your processes, your capabilities, and then in the last stage, also inventing.

So really using all of that to build new products, new services, et cetera.

And we believe that fit really well.

So we said, hey, cool, we have an approach.

Let's do it.

But it still didn't take off.

Why Prescribing Doesn’t Work: Learning by Doing

because you can't just prescribe people now go and figure this out because a lot has to do with

I think what some of you also said it's really hard and I love this quote from I think the

general council of Bayer when it comes to AI and technology it's all about learning by doing

you won't figure everything out right away but the more you engage with it the more opportunities you

you see.

I think that's very true.

The Four Stages of Competence

To put it in a bit more fancy 2x2 matrix, has anyone

heard of the four stages of competence?

Okay.

Fairly simple.

You move from unconscious incompetence

to conscious incompetence to conscious competence, yes, and unconscious competence.

So, really,

in the beginning, you don't know what you don't know.

You have no freaking clue how

how powerful these AI tools are, you're unaware of what they can do.

So how should you kind of go and figure out how to put that in your process if you don't

know what they're all capable of?

Then second stage, you've seen them, you've seen demonstrations of these fancy tools,

you know nothing.

So conscious incompetence.

And the rest, I think, is clear.

So you're starting to kind of more actively use in this case AI or gen AI

you still

Require quite an effort.

It's not kind of it doesn't come naturally

so this is more of a bit of a strategic framework that we used with our

corporate to help them explain how we you can go through and then we also

Talked about how could we can kind of make that a bit more real?

Making It Real: Our Enablement Toolbox

So and we kind of moving through the quadrant so you need

Awareness and Learning

need to bring awareness to people and these types of events are perfect.

We have events, exchanges that we organise.

We had 11Labs last week helping all of our teams build agents, voice -based agents.

We had events on AI in venture building bringing together a lot of different companies talking

about how we can use AI in venture building and so on.

So we offered self -paced learning to our Lufthansa group colleagues, creating kind of video content, but also for ourselves.

Performance Support and Prompt Craft

We provided performance supports.

One of the best ones that we had was this Chrome plug -in that you provide.

I think you still do, because the Chrome plug -in would always give you feedback on your prompts.

That was the best thing.

everything prompt libraries everything that you can give to people to actually help them in their

Budgets, Trainings, and Team Workshops

day -to -day job we started or we actually have that as li as lih learning budgets so really just go

and figure this out go on online courses upskill yourself we have trainings we had team workshops

where it really is about starting a conversation with your team

and trying to figure out what are the use cases,

where can we deploy AI.

Embedding AI in the Company Narrative

And lastly, we also made it part of our company narrative

and kind of the structures and how we operate.

I think that was then really where it changed our game a bit.

And I have a couple of examples of what we did.

What We Changed: Concrete Initiatives

So what we then started with across all of the different teams

Launching an AI Practice

is launch an AI practice.

Three things.

Number one,

figure out who are the relevant people

at Lufthansa Group

and how do we build the bridge.

Number two,

develop own thought leadership

and perspectives on AI.

So we as Innovation Hub

should have understanding

of how is it going to impact organizations,

but how is it going to impact

the value chain,

industry dynamics,

how is it going to affect agents,

the booking process.

So we should be at the forefront.

And number three,

also help us figure this out.

So that was one of the game changers.

AI Stipend and Tool Experimentation

We then started offering an AI stipend.

So we pre -approved a couple of tools

and we said, here you have, I think, 30 euros per month.

Go and test them.

Switch between Cloud and ChatGPT.

Use other tools

because they're going to constantly change capabilities.

So you need to experiment and kind of see what works for you.

I think I talked about AI events and upskilling.

Communication and Product Redesign

We started to really, and sometimes very, very simple, regular AI Slack digests.

So really just keeping AI and developments somewhere in the company.

We have a couple of, obviously, products towards Lufthansa Group.

One is, for example, an eight -week innovation training that we do with Lufthansa Group colleagues and employees.

So we thought, so again, from idea to MVP, prototyping, so the typical innovation funnel,

let's completely redesign this thing with all the AI tools that we know.

So really, you go all the way to lovable for prototyping, which obviously then requires

us to figure out how this works.

It's a mix.

It's a mix.

So for example, this product elevation, our products towards the Lufthansa group, but

to obviously infuse AI into them, we have to figure out how AI works.

A lot of use case workshops.

Venture Process and Continuous Showcasing

We're more and more building it also in our venture process.

And really, again, not having individuals kind of do it,

but really bringing it to a company level,

continuous showcasing of, hey,

what applications have we kind of prototyped with Lovable?

How are we using AI in testing hypothesis

hypothesis, or synthetic customers, et cetera, in our validation.

Experimenting with AI Video

We're trying to jump at every opportunity to experiment.

For some reason, we had a couple of requests now for AI video generation.

Nice that colleagues kind of think about us as Innovation Hub, but we had no clue.

So can you create a two -minute video on this topic?

There's AI for that.

Can you do it?

So we asked our marketing team before they directly declined it, go and figure it out.

What can we do?

And then we experimented with all the different tools.

And again, just for video creation, there's probably like 20 different tools that we went through to figure out which one's the best.

It's an effort, but we try to jump at every opportunity to experiment.

Making AI Part of OKRs

Depending on how you're managing yourself, we work with OKRs, objectives and key results as a goal setting tool.

tool, we made it part of that.

So teams were encouraged to create goals or key results.

We're planning to automate X processes and see if we can use AI for that.

It's more of

a technology angle, not a use case angle per se, but that also helped to really kind of

get people to think about it on a quarterly basis.

The last one, and we're a little bit

Building an AI Adoption Platform

bit late in our own timeline, is we're just exploring an AI adoption for ourselves.

It's an

idea that came out of our AI practice, which is still, with all of that, trying to give it more

structures.

So it's about trying to help the teams and the team leads with transparency, a structured

approach, and understanding how far are we actually along this journey.

And we wanted to really

trigger conversations internally.

This is entirely vibe coded in two hours.

We wanted to kind of test it with one or two teams.

I can briefly walk you through

and happy for any conversation or feedback afterwards because this is our latest attempt

in kind of trying to give us a structure and a platform for AI.

So what we're

planning on doing is giving everyone basically a platform where they see their AI usage personally,

team averages, how the company's doing across different products we have, which

have obviously different activities.

We can look across different teams, how are

different teams using, augmenting certain activities with AI.

We're gamifying this

a little bit, so here you can basically anonymously add how you are, for the very

specific activities that you have in your team, using what types of tools, how

how frequently, what use cases you have.

The team lead can view how their team is currently doing,

how many people in their team are using what.

There is a priority matrix where they can see, let's say,

what are the important processes that we have that help us

achieve our business impact, what's the AI potential,

augmentation potential, and where could we kind of find

use cases, et cetera.

So we're kind of constantly, again, experimenting.

We don't know whether this is going to work,

but it's our attempt to figure out how we can kind of

make this in a much more structured way because one thing that we realized over the last couple of months is

this is far more than just an implementation of a new tool where you just send people to some

class and then afterwards they figured out where to click.

This is like a fundamental

change in

how we work that will affect most of the activities that we do.

Bringing It Home: What Leaders Should Do

And with that, this is, I guess, the last slide, and it really starts with bringing

this topic to the forefront.

Keep AI on the Agenda

So a couple of things that I always end with is do you have a decent understanding of the

AI language?

You talked about some of the key terms.

Can you translate from the technology to actually why the heck is this relevant for us?

Do you bring exciting use cases?

Do you constantly kind of look left and right and find those things that you could kind of use to inspire your team?

What routines do you build to actually keep AI

part of the conversation in your in your teams?

Where can you put it on the agenda visibly?

Create Space to Experiment

Early next year, we're currently using our company or planning to use our company off -site to just do an AI hackathon.

And again, where can you encourage experimentation in your team?

And maybe that's a nice theme of all the three.

Closing Thought

So it's all about experimentation.

And that's it.