Latest / The In-Between Tech and Trust Podcast / WEF26: The Politics of Tech, AI Agent Systems & Models, Adoption Challenges and Tech Sovereignty (EP 17)

Transcript

WEF26: The Politics of Tech, AI Agent Systems & Models, Adoption Challenges and Tech Sovereignty (EP 17) · The In-Between Tech and Trust Podcast

• 0:14
  Hello, this is Eva and you're listening to the In Between Tech
• 0:18
  and Trust podcast. This episode gets quite well,
• 0:21
  personal and special because it's the first solo episode that
• 0:26
  I'm going to record. And I thought it's a good moment
• 0:29
  in time because I just come back from a week in Davos.
• 0:32
  I've connected with a lot of experts on tech, politics, AI
• 0:38
  and also the world in 2026. And so I did want to talk to you
• 0:43
  about what's been in conversations and wanted to map
• 0:48
  out some thoughts that I had afterwards and also some deep
• 0:51
  dives that I thought could be interesting.
• 0:56
  This year's World Economic Forum couldn't have come at a better
• 0:59
  time because getting into 2026, we've already experienced how
• 1:05
  many geopolitical tensions there are.
• 1:08
  Also how collaboration is undermined by distrust and how
• 1:14
  tech convergence impacts the business sector and how we can
• 1:18
  also impact it over the next couple of years.
• 1:22
  And I want to come to some topics that were discussed on
• 1:25
  the ground because next to, you know, all the political speeches
• 1:29
  that you might have heard of, or next to also the arrival of
• 1:33
  Donald Trump, of Elon Musk and all other world leaders.
• 1:36
  For me, there were four specific topics that I wanted to
• 1:41
  highlight to you. And 1st, this was tech, second
• 1:46
  were the politics of tech, Thirdly, it was serenity.
• 1:50
  And then lastly, it was also the topic of collaboration and
• 1:53
  trust. So let's start with the tech
• 1:55
  aspects because the technological progress was, you
• 2:00
  know, with focus on the impact on society, politics and also
• 2:04
  the global collaboration. And it was basically on
• 2:08
  everyone's talking point. And some of those you might have
• 2:11
  heard. So basically in every
• 2:13
  conversation that I attended was talk about agentic systems, AI
• 2:18
  models and also how they are adopted for a time to business
• 2:23
  impact. It's worthwhile to dig into
• 2:26
  those because they will shape of how we also collaborate and
• 2:30
  navigate the next months to come.
• 2:32
  And I wanted to start with the agentic systems aspect because
• 2:36
  they are considered and from the conversations that I had, it's
• 2:39
  really about the transformation that they'll bring to the
• 2:42
  business world. And there are three profound
• 2:45
  reasons of why this transformation is in a way.
• 2:48
  So first, they transform the world for knowledge workers and
• 2:52
  those who use AI to deliver knowledge based output.
• 2:56
  And it means that it can vary from very specific single state
• 3:00
  type of output to whole products that can be provided, which is
• 3:04
  also one of the second topics that I wanted to talk to you
• 3:07
  about because a genetic systems will challenge how we look at
• 3:11
  organizations and collaboration in itself.
• 3:14
  And you might ask yourself, well, why is that?
• 3:17
  And I do think that there are several reasons to it, but the
• 3:20
  first one is because agentic systems only work if they're
• 3:23
  connected via clear processes, that they need to be designed
• 3:28
  towards specific outcomes and they need to be monitored for
• 3:32
  performance and also for up to date models.
• 3:35
  And while I want to talk to you about the model aspect in a
• 3:38
  second, I want to linger a bit with the design part because
• 3:42
  designing agentic systems required to be clear here on
• 3:47
  what they are designed for. And it can be a specific
• 3:51
  outcome. It can be a dedicated way of
• 3:53
  working together, or it can also mean that you want to solve for
• 3:58
  a specific process, a problem or an outcome that an organization
• 4:03
  want to create. Now, if you look into the
• 4:06
  heritage of potentially the organization that you might work
• 4:09
  for, your processes and the way that you collaborate might have
• 4:13
  been designed according to the skills hired according to the
• 4:18
  output that you want to deliver. And you're used to collaborate
• 4:22
  with the strategy department, with a product department, with
• 4:25
  finance teams, HR and so on. And this is a really classical
• 4:29
  way of how teams departments are structured and all of them are
• 4:33
  feeding into like a specific process, but only like they
• 4:38
  cover parts of it with their respective skills.
• 4:42
  And now if we translate that into an agentic driven system,
• 4:46
  agents will take either very specific parts of a task or, you
• 4:51
  know, collaborate closely with yourself, for example.
• 4:54
  But they will also be designed as systems overall and will be
• 4:58
  built for several agents to form a system.
• 5:02
  And you can think of them as a an agentic team that collaborate
• 5:06
  and that produce uses and is designed for a dedicated
• 5:08
  outcome. So it is required to know what
• 5:12
  you're aiming for when designing those agents and also what is
• 5:17
  needed to activate it, the conversation on the ground.
• 5:19
  And that was really approached this view on how organizations
• 5:24
  currently struggle with the design part, because it's super
• 5:29
  complex and it also requires quite the clarity of what's
• 5:32
  possible with AI models where human expertise need to be
• 5:36
  catered to, and also how those systems interact and need to be
• 5:40
  organized on an overarching and holistic organizational view.
• 5:44
  So the complexity of what I've just described is what AI
• 5:47
  leaders and levels have referred to when they talked about the
• 5:50
  challenge to adopt AI. And there was this general
• 5:54
  consensus that organizations need to be fast on the
• 5:57
  implementation. Yet most of them feel
• 6:00
  overwhelmed when it comes to the action plan of how that really
• 6:04
  is to be done. And most of the conversation
• 6:06
  that I've been part of during the day is like focused on the
• 6:09
  urgency to prepare for those adoption challenges.
• 6:13
  And it's been clear that an action plan is quite individual
• 6:17
  to the size and the industry and the delivery model the
• 6:20
  respective company is in. Though there has been a clear
• 6:23
  call to action to train employees to also be clear on
• 6:27
  what skills are required and to activate collaboration on an
• 6:32
  organizational level. And interestingly, and this is
• 6:35
  also what Satya Nadella, the CEO of Microsoft stated, he said
• 6:39
  that AI is a platform shift and not an app nor a tool.
• 6:44
  And he described it as the next computing layer.
• 6:47
  He said that organizations need to go from auto complete to
• 6:51
  agent mode to full project autonomy.
• 6:54
  And the point that he made is the winners that adapt the
• 6:58
  fastest aren't model builders, but they're the companies and
• 7:02
  countries that reorganize work around reasoning, which then
• 7:06
  again leads to the fact that an organizational chart becomes a
• 7:10
  software problem. So leaders, organizations and
• 7:14
  also adapting to AI and tech infrastructure need to look into
• 7:19
  the way of how they they approach their collaboration
• 7:23
  within the company and prepare their teams and transform them
• 7:28
  into, you know, the way that software and platforms can then
• 7:33
  be activated. The third part that I wanted to
• 7:37
  talk to you about in terms of tech is the use of different
• 7:40
  kinds of AI models. And it's been widely discussed
• 7:43
  on the ground and also in the past couple of months that an AI
• 7:47
  model provides the intelligence to drive overarching output
• 7:51
  creation. So currently there are only a
• 7:54
  few companies who have the means and the skills to provide that
• 8:00
  to the rest of the world with those models.
• 8:02
  And this particularly fuels the political and the sovereignty
• 8:06
  discussion, which I'll get to you a bit later.
• 8:08
  But for now, the design and the transparency of how those models
• 8:11
  actually work cost some of the conversations because there are
• 8:15
  those who call for world models and those who undermine the
• 8:20
  tendency to specified and small models.
• 8:23
  So I wanted to talk to you a bit about what are world models and
• 8:27
  also what are specified or small model.
• 8:30
  And starting with world models, these are like neural networks
• 8:34
  that understand the dynamic of the real world and also include
• 8:39
  physics and spatial properties. That means that they, you know,
• 8:43
  have a lot of data in terms of also the environment that we're
• 8:46
  in, and they can use input data to generate videos that simulate
• 8:51
  realistic physical environments. The interesting part is that
• 8:55
  within a world model, you can then build them for prediction
• 8:58
  models. You can use them for style
• 9:00
  transfer models or like further reasoning models to begin with.
• 9:05
  But from what was discussed in Davos, we're also just at the
• 9:08
  beginning of how these application contexts are
• 9:12
  evolving. And the benefits are pretty
• 9:14
  clear because they extend AI capabilities with a deep
• 9:19
  understanding of how spatial relationships are being built
• 9:24
  and also how physical behavior in three-dimensional
• 9:28
  environments come into existence.
• 9:30
  On the contrary, there is the rise of specified and small
• 9:34
  models that are available for a particular application.
• 9:38
  If I talk about particular applications, you can now think
• 9:42
  it's for example, defence tech, it's state related, it's also
• 9:46
  potentially government driven. And they are designed for narrow
• 9:51
  and predefined tasks that are classified or that also generate
• 9:56
  particular outcome. And they then On the contrary to
• 10:00
  the world models, they use statistical correlations and
• 10:03
  they also are specified models looking into the way of so how
• 10:07
  they are focused on particular and high accuracy tasks because
• 10:12
  that is where you do not want to be dependent on this one
• 10:16
  specific, you know, large scale infrastructure that potentially
• 10:19
  doesn't serve in the way that you want it to.
• 10:22
  And everyone was very clear that going forward, the
• 10:26
  implementation of I will consist of both worlds.
• 10:29
  And we do need to decide the context that those models are
• 10:34
  applied in and the impact that they should have.
• 10:37
  So it's essential to integrate them in the different kinds of
• 10:41
  facilities or workflows that we look at.
• 10:45
  And that it's also orchestrated how their impact on business,
• 10:49
  state, society are being monitored.
• 10:53
  And it became clear that leaders and organizations are
• 10:56
  responsible to put together the most efficient orchestration for
• 11:00
  them and that on a state level, it's required to also dive into
• 11:06
  the enablement of exactly this orchestration.
• 11:09
  While on a societal level, there are still a lot of trust and
• 11:13
  ethical unclarities that haven't been solved and that also now
• 11:18
  get the more and more into contradiction of how we will
• 11:21
  handle it. But if we linger a bit with the
• 11:22
  business sides of it, it became quite clear that the future of
• 11:26
  AI model application for businesses is about how they
• 11:30
  select and orchestrate those models in their specific
• 11:34
  context. And for organizations, it means
• 11:37
  that they will have to choose from several different kinds of
• 11:42
  specific and different model setups.
• 11:44
  And also for organizations, it will impact their operating
• 11:47
  model. So they need to stay flexible
• 11:49
  and clear on the outcome. They will have to navigate
• 11:52
  constant model updates and they will also need to still provide
• 11:57
  a stable infrastructure so that it allows some sort of a degree
• 12:01
  of flexibility, which then again also needs to be translated in a
• 12:06
  whole probably other level of delivery.
• 12:09
  But it's also required to think about what do we deliver
• 12:12
  towards. And interestingly, and in this
• 12:14
  context, it was Alex Karp, the CEO of Palantir, who reflected
• 12:19
  upon that as a stress test for organizations and for reality,
• 12:24
  because he made the point that it's really about institutions
• 12:27
  and that real environments are adversarial, which means that
• 12:32
  there is no connectivity, There is still messy data.
• 12:35
  We need to talk about permissions.
• 12:38
  Where is ethics anchored though AI doesn't fix that, but it
• 12:42
  exposes it and essential for organization that we need to
• 12:47
  provide an and reliable and auditable layer in terms of, you
• 12:52
  know, infrastructure. And when we talk about
• 12:54
  infrastructure, this was exactly a recurring topic that came up
• 12:57
  as well because everyone was pointing towards the requirement
• 13:01
  for it to be interoperable, particularly when it comes to an
• 13:05
  organizational level. But if we look in to the
• 13:08
  business aspects of interoperability, it means that
• 13:11
  tools or AI models can interact with each other and ideally they
• 13:16
  can do so seamlessly. Currently, it's one of the
• 13:18
  biggest challenges for organization and particularly
• 13:21
  also for tech companies. Why?
• 13:23
  Because most of the tools work in their respective context or
• 13:27
  environment and it's hard for them to be integrated with one
• 13:31
  another because there are mechanisms and also how they
• 13:35
  deliver an output might not work together, also might not be able
• 13:40
  to be activated the way that a person, an organization works
• 13:45
  with different kinds of infrastructure.
• 13:47
  So currently there are several startups and also in Davos,
• 13:51
  there were several ones pitching for an interoperable
• 13:55
  infrastructure layer that want to provide exactly this kind of
• 14:00
  integration of different kinds of tools, of different kinds of
• 14:03
  ambitions, of different kinds of models to also serve as a
• 14:07
  general connector and to also provide clarity for
• 14:11
  organizations of how they can then orchestrate their AI tools
• 14:16
  and models and integrate it into the way of how they produce
• 14:20
  output. All of this leads into a larger
• 14:22
  conversation that still lingers and that is the increasingly
• 14:26
  need for an adoption rate that becomes faster in its time to
• 14:32
  business impact. And a lot of organizations still
• 14:35
  do not see critical business value generated.
• 14:37
  There's still, you know, up to 50% of organizations on the
• 14:42
  ground that said that for them, it's super hard to implement AI
• 14:46
  and to navigate it in a valuable way.
• 14:49
  And I think this is due to two topics.
• 14:52
  First, it's like the design aspect that we've already dove
• 14:56
  into earlier and it's also the lack of focus on the design
• 15:00
  process and on organizational architecture required for AI to
• 15:06
  really improve and provide impact.
• 15:08
  And the second is like access to skill improvement for employees.
• 15:12
  And the latter is one of the critical ones.
• 15:15
  And there are two aspects of it. On the one hand side, it's the
• 15:20
  provision of time and education material for employees
• 15:24
  respectively, for people to adopt and learn to their
• 15:27
  required background information. However, it also comes down, on
• 15:31
  the other hand, to clear responsibilities.
• 15:33
  And one of the main key takeaways that I had during
• 15:36
  Davos is that it isn't really clear who provides also the
• 15:40
  incentives for people to actually follow through.
• 15:44
  And I'm not talking about an organizational level because
• 15:46
  every, you know, organization does have and provide a lot of
• 15:50
  effort for their employees. It is more about the holistic
• 15:53
  view of it because tech companies somewhat point to
• 15:58
  government to provide more frames, regulations or also
• 16:01
  clear guardrails of how future models will be incorporated and
• 16:06
  how then again, employees make use of it.
• 16:08
  And because due to the lack of a unified approach, also how to,
• 16:13
  you know, distribute the responsibilities when it comes
• 16:16
  to implementing AI in organizations.
• 16:19
  The transparency aspect is so important, but also the fact of
• 16:24
  what actually gets implemented respectively, what output we can
• 16:28
  expect. So there's a lot of, you know,
• 16:30
  navigation to be done and also distrust to be addressed.
• 16:35
  And the these are some of the most critical conversations to
• 16:39
  be solved in the foreseeable future.
• 16:41
  And who is it going to be solved by?
• 16:43
  Well, This is why politics, and it's the second overarching
• 16:48
  topic that I wanted to talk to you about, got tied into tech in
• 16:52
  so many conversations. Either it was on the back of
• 16:55
  geopolitical tensions and due to the large scale resource needs,
• 17:00
  but also in the context of whether tech is up or downstream
• 17:03
  of politics. And if I think that tech is up
• 17:07
  or downstream of politics, you might ask, well, that's this
• 17:10
  sounds complicated. What do you actually mean?
• 17:12
  Well, at the core, the distinction is about direction.
• 17:15
  And more explicitly upstream is when the things are defined,
• 17:18
  designed and made possible. So you can think about this as
• 17:22
  strategy part, the standardization, infrastructure,
• 17:25
  resources, data generation and so on.
• 17:27
  And downstream is where things are applied, distributed and
• 17:31
  experienced. So you can think about it as
• 17:34
  products, services, markets, users and social impact.
• 17:38
  And so when it comes to tech, that means that chips, cloud
• 17:43
  infrastructure, protocols, training data, open source
• 17:46
  libraries and so on, these are upstream, while apps, platforms,
• 17:52
  consumer devices, workplace tools are downstream functions.
• 17:57
  Now you might have to come to the same conclusion and
• 18:01
  technologies both upstream and downstream of politics.
• 18:04
  And the direction flip depends on different kinds of layers and
• 18:08
  also how we're looking into it. And this is at the core.
• 18:11
  And I can tell you it was visible so many times of what
• 18:15
  was experienced and why tech leaders were so present at
• 18:19
  Devils, which particularly also so was shaped by politicians,
• 18:22
  right? It became clear that the
• 18:24
  interwovenness of how tech companies currently are both
• 18:28
  upstream and define how tech impacts us as well as it is
• 18:32
  downstream and shapes the application side of it, is why
• 18:36
  we're in, you know, those discussions.
• 18:38
  At the center of it all, and ideally on what everyone keeps
• 18:41
  asking for is that politics, it's upstream of tech and that
• 18:45
  it shapes what kind of technology gets built and by
• 18:48
  whom, for example, through regulations, geopolitics, public
• 18:51
  funding or standards, laws, etcetera.
• 18:54
  So they define incentives, constraints, and the strategic
• 18:58
  direction at the same time. And what we're witnessing and
• 19:02
  what we've been also discussing in the past couple of months or
• 19:05
  years is that tech sits upstream of politics and also at the
• 19:10
  political inaugurations. If you just remember last year
• 19:13
  when we saw inauguration of President Donald Trump, it
• 19:16
  reshapes how politics itself functions through algorithms,
• 19:20
  surveillance, decision systems and the finance of it all.
• 19:25
  So this is how political behaviour gets shaped and also
• 19:29
  how institutional capacities and power asymmetries come to view.
• 19:35
  And so of course, there is this feedback loop that is defined by
• 19:39
  a recursive relationship. Power roles and power structures
• 19:44
  shape upstream tech decisions. And tech reshapes society and
• 19:48
  politics itself. And politics again reacts
• 19:50
  oftentimes too late to complicated or too far away from
• 19:54
  operational reality regulations and redirective elements.
• 19:59
  But that then again impacts how tech continues to apply and be
• 20:04
  applied. And the discussion of upstream
• 20:05
  and downstream tech mostly resonated in devils in the
• 20:09
  conversations on energy supply data transparency.
• 20:12
  To provision talent enablement and so on.
• 20:15
  But interestingly and with the tech firms being so huge
• 20:19
  economical conglomerates already, they now also shape the
• 20:23
  labor markets, influence political processes and also
• 20:26
  concentrate power in a few firms or states, which then again push
• 20:32
  tech downstream into political outcomes.
• 20:34
  And that again undermines why endeavors, the political
• 20:37
  presence and the high attendance of tech CEO's was so omnipresent
• 20:41
  this year specifically and interestingly.
• 20:43
  And if you look a bit deeper into those who are now first
• 20:47
  movers in tech and apply their advantage, they define the
• 20:51
  standards and they also shape user behavior and they lock in
• 20:54
  dependencies which are hard to reverse them again.
• 20:57
  And that is what drives the narratives that are framed as
• 21:01
  efficiency driven, providing safety or also how they designed
• 21:04
  for user experience. And while these are all, and I
• 21:08
  really don't want to downplay it, I do think it's important to
• 21:11
  deliver exactly that and to also optimize for growth and
• 21:14
  innovation. The conclusion is when tech
• 21:17
  shapes political decisions as we've just arrived, there is a
• 21:20
  certain amount of transparency needed on how those decisions
• 21:24
  are determined. So tech needs to move fast and
• 21:27
  at the same time, they also need to be hold accountable of how
• 21:30
  they get there. And this then needs to be fed
• 21:33
  back into the responsibility conversation we touched upon
• 21:36
  earlier and the Nevos. This became essentially the core
• 21:40
  but underlying non explicit conversation because tech
• 21:43
  companies affect billions of users and like the companies and
• 21:47
  revenues and so on. And at the same time their
• 21:49
  systems are hard to inspect while only a few firms control
• 21:54
  core infrastructure. And that then, shapes the
• 21:56
  following. The dependency aspect on how
• 22:00
  states rely on private tech for public functions and how they
• 22:05
  rely on tech for public functions centers one aspect,
• 22:08
  sovereignty. And well, sovereignty for all of
• 22:12
  us is quite like the huge word. And in itself, it has been, you
• 22:18
  know, a top topic and also in every other conversation for the
• 22:22
  past two to three years or so. And on a political aspect, I
• 22:25
  mean, this is some, this is where tensions on also state
• 22:30
  aspects come into play. But from a tech perspective,
• 22:33
  this gets really interesting because in the one hand side and
• 22:36
  particularly from a European perspective, the need for
• 22:40
  technological progress based out of Europe is like essential to
• 22:46
  stay sovereign. On the other side, it was also
• 22:49
  Satya Nadella again, who said the topic least talk about where
• 22:53
  we most talk about this year. And it's the sovereignty of a
• 22:56
  firm. But it inherently became a
• 22:59
  statement about competitiveness, value and control.
• 23:02
  And you might ask, well, why do we talk about it and why now?
• 23:07
  Well, given the interwoveness of tech and politics as we've just
• 23:12
  dove into, and also as the economical impact of AI on
• 23:16
  society became quite clear and quite the central theme that we
• 23:20
  need to also disentangle. And before we do that, we need
• 23:24
  to talk about what sovereignty in itself and particularly what
• 23:27
  it means in tech and AI. And sovereignty means being and
• 23:32
  having the ultimate authority to the side and to make or to say a
• 23:38
  final decision, set binding rules and also define the
• 23:42
  impact. There are different
• 23:44
  perspectives, different actors and also different aims to be
• 23:47
  sovereign and it can be territorial, legal, political or
• 23:52
  economical. But when it comes to tech and
• 23:54
  AI, and in my opinion, there are several layers to it.
• 23:57
  The compute energy sovereignty and so the ability to secure
• 24:02
  economically viable reliable energy for AI compute over time.
• 24:07
  So this is where it gets mostly interwoven with territorial
• 24:11
  sovereignty and paired with the mining and raw material aspects
• 24:14
  because only a few countries currently have the means to
• 24:17
  power and the infrastructure needed for the future AI growth.
• 24:21
  And this is also why the current conversation of energy supply
• 24:25
  through alternative aspects get so important.
• 24:28
  Second, infrastructure and platform sovereignty, which is
• 24:32
  one of the most complex ones because it's about dependency
• 24:35
  depth and some sub layers to look into it.
• 24:39
  We talk about compute sovereignty.
• 24:41
  How do you get access to GPUs and cloud contracts?
• 24:44
  Then you talk about platform server entities.
• 24:48
  Which cloud and ML OPS stack do organizations use?
• 24:51
  And then there is also the network sovereignty, which
• 24:54
  addresses the latency, routing and resilience of AI to work
• 24:58
  with. And these layers particularly
• 25:00
  are steered by only a few companies in the world, which
• 25:03
  get even more specific when compared to the vast amount of
• 25:07
  organizations they provide their infrastructure too.
• 25:10
  So this is also the conversation you might have followed along
• 25:13
  and most shaped by the term the Magnificent 7 with Alphabet,
• 25:17
  Amazon, Apple, Meta, Microsoft, NVIDIA and Tesla and focus of
• 25:21
  providing critical infrastructure to it.
• 25:25
  And then looking into third, the data server and aspect, which is
• 25:30
  the ability to control, protect and strategically also provide
• 25:34
  continuous data flows generated by AI deployment and the
• 25:39
  securance of ongoing data generations as well as the
• 25:42
  control of feedback loops. And looking into the data aspect
• 25:46
  a bit more. This is where is this underlying
• 25:48
  need for transparency and regulation because it plays into
• 25:52
  the way of how they do not have sufficiently the possible
• 25:55
  ability to control, protect and provide continuous data flows
• 25:59
  for them to be secured in the aspects and respective
• 26:03
  environments without being dependent on the tech firm
• 26:06
  overall. 4th, we talk about operational sovereignty, which
• 26:10
  means that we can audit AI decisions, we can intervene in
• 26:14
  real time and we can avoid single point of AI failure,
• 26:18
  which is specifically critical and regulated or safety critical
• 26:22
  industries. Here again, you could think
• 26:24
  about the defense second environment and you can also
• 26:26
  look, look into the geopolitical tensions.
• 26:29
  But This is why it becomes one of the areas to watch and also
• 26:34
  where we need to decide on how to move ahead, particularly from
• 26:38
  a European perspective because this serenity currently is not
• 26:42
  given. And this is like the main focus
• 26:44
  areas of a lot of organizations of where they want to push into
• 26:47
  to have this operational focus and ability to steer it in their
• 26:54
  respective views and not being dependent on like a third party
• 26:58
  vendor. There is the design and
• 27:00
  behavioral sovereignty and this implies the the ability to
• 27:03
  define what AI optimizes for and to also influence model
• 27:09
  behaviour systematically and define the control over the
• 27:13
  evaluation matrix. This is specifically the part of
• 27:16
  dependency that all companies have on the models that they use
• 27:20
  because this is also the value aspect the AI and tech companies
• 27:24
  try to secure to ensure their level of intelligence and the
• 27:28
  progress their respective large language models to keep and also
• 27:32
  to make the race for general artificial intelligence.
• 27:36
  The last serenity aspect then again is from a talent and
• 27:40
  organizational perspective because organizations required
• 27:43
  to the internal capabilities that provide an understanding
• 27:48
  and operations and also an involvement with AI systems
• 27:53
  without external gatekeepers. If you've now heard the seven
• 27:57
  different kinds of types of sovereignty, you might get a
• 28:00
  glimpse into why this is so important, but also why it's so
• 28:03
  complex to well sort it because we have 7 different kinds of
• 28:09
  ways of where we currently are dependent on non European
• 28:13
  platforms. We're also looking into the way
• 28:17
  of how to enable politicians or to enable also financial flow
• 28:22
  across our, you know, or European environment.
• 28:24
  When we pair this to the criticality of what we've
• 28:28
  earlier looked into and that is the influence of tech companies
• 28:31
  on politics, you realize that knowing Texas upstream politics,
• 28:38
  knowing that tech also benefits from autocratic tendencies.
• 28:44
  This is going to become one of the most crucial conversations
• 28:47
  that we're going to have in the next couple of months and years.
• 28:50
  It's been core of the discussion in Devils and it's also going to
• 28:53
  be a core of the discussions that are going to be move ahead.
• 28:56
  And interestingly, on the ground, there have been several
• 29:00
  solutions tied to it. On the one hand side, it's been
• 29:04
  said to focus on making sure that the auditing and
• 29:08
  transformation of organization. And when it comes to the state
• 29:12
  of infrastructure, it needs to be very clear of where big tech
• 29:15
  companies currently have access to, of how they also are
• 29:19
  dependent on the work that the different kinds of organizations
• 29:22
  do, which tools and platforms could be good alternatives that
• 29:26
  are being provided based out of Europe.
• 29:28
  Second, it's all about the development of talent and making
• 29:31
  sure that Europeans know and have like a respective skill set
• 29:35
  to enable an infrastructure shift.
• 29:38
  And this is a point which looks into how education systems are
• 29:42
  set up. And then thirdly, it's also
• 29:45
  about the unification of a financial landscape and how we
• 29:49
  enable investment influx because currently there is a lot of
• 29:53
  money that flows into the US and that is also enabling US based
• 29:59
  company. Now you might have heard about
• 30:01
  the EU Inc proposition that was a lot underline has proposed.
• 30:06
  It's like one of the best steps into an implementation part and
• 30:10
  also into the way of accumulate more financial means to enable
• 30:16
  this infrastructure shift and to also attract talent in the mid
• 30:20
  or long term, to sum it up. And I think sovereignty aspect
• 30:24
  of like one of the most challenging ones to follow
• 30:27
  through. And it also remains to be seen
• 30:30
  of how we'll establish as European a mindset again that
• 30:34
  provides us with a strong foundation.
• 30:36
  And then lastly, 2 topics that I still wanted to talk to you
• 30:40
  about this collaboration and trust.
• 30:42
  And interestingly, both of them are quite tied to each other.
• 30:45
  And that's also why I wanted to connect them with each other.
• 30:48
  And when it came to collaboration on a global basis,
• 30:52
  but then again, also on an organizational or team basis, I
• 30:56
  sometimes got the feeling when I talked to, you know, experts
• 30:59
  that there was this unspoken question tying into it.
• 31:03
  So what does collaboration even still mean?
• 31:06
  So can we trust each other? And we also look into a
• 31:10
  commitment that we want to collaborate on a global basis.
• 31:13
  And while I know and while I experienced a lot of commitment
• 31:17
  to enable global scale collaboration, particularly in
• 31:21
  tech, the requirements to cater to individual needs became clear
• 31:26
  in many ways. Because if we continue to, you
• 31:30
  know, provide this commitment from mutual collaboration and
• 31:33
  trust, and also on a diplomatic and business level, we need to
• 31:38
  enable it through action. And with that, I wanted to round
• 31:41
  up a lot of topics that we touched.
• 31:44
  We talked about the impact of tech, the interwovenness of tech
• 31:48
  and politics, why sovereignty is so crucial but also so complex
• 31:53
  to solve, and how we going to steer ahead with collaboration
• 31:57
  and trust. And it's now end of January
• 32:00
  2026. What I come home with is to be
• 32:03
  very clear that this is a current conversation that needs
• 32:06
  to to a context and commitment, being very specific of how we
• 32:11
  want to shape the world ahead and making sure that the values
• 32:16
  that we have, the ambitions that we want to drive forward are
• 32:19
  enabled and not predefined by an infrastructure that we might
• 32:23
  use. I am very curious to hear what
• 32:27
  that came to your head, but until then, thanks so much and I
• 32:31
  hear you next week.