Why leadership now determines AI success
A conversation with Tobias Piegeler (consultant for digital strategy, transformation and innovation), Henning Sander, partner banking, and Kai Bättenhausen, senior manager software (HAGER Executive Consulting).
The disruption caused by technology has never been greater. The way companies deal with artificial intelligence is already a key factor in determining their competitiveness. The willingness to consciously shape technological change rather than just going with the flow, the courage to respond decisively to change, and the ability to embed AI deeply in the operating model are becoming critical success factors for executives and companies.
Translated with DeepL.com (free version)
This is precisely the starting point that Henning Sander and Kai Bättenhausen from HAGER Executive Consulting discussed with Tobias Piegeler, a consultant who supports companies with digital transformation and AI integration. The conversation focuses on economic impact, governance issues and leadership requirements in the age of AI. The conversation will be published in three parts. The first part focuses on why the current AI transformation is structurally different – and why it is changing people, systems and organisations at the same time.
The crucial question is not: Do we have an AI strategy? But rather: Do we have the leadership to implement this strategy?
Why this change is different
Henning Sander:
Tobias, technological changes have always been around. Why does the current transformation feel so radical?
Tobias Piegeler:
Because speed and scope have a structurally different effect this time. AI is changing markets, business models, decision-making logic and role models. Simultaneously, not sequentially. This is not iterative change. It is a parallel shock on several levels.
In regulated industries such as banking and insurance, this dynamic is coming up against established structures, rigid governance and high liability. However, AI does not follow traditional project logic. It develops faster than decision-making processes.
Anyone who responds to this with classic project thinking – concept, pilot, review – has not understood the problem. The question is not: Do we use AI? The question is rather: How do we guide people and organisations through constant change and achieve economic impact in the process?
Three ways AI is changing work
1. Humans: Knowledge asymmetry tips the balance
Henning Sander:
What specific changes occur in humans?
Tobias Piegeler:
AI fundamentally shifts the knowledge asymmetry in organisations. Today, employees have access to knowledge, analyses and decision-making bases at the touch of a button that used to take entire teams several weeks to compile. At first glance, this sounds like efficiency, but in reality it is a critical management task.
When employees suddenly have a better basis for decision-making than their superiors used to have, when content silos are diluted and automation disrupts existing role profiles, hierarchy loses its traditional legitimacy. In future, leadership will be legitimised not by knowledge advantage, but by orientation, context and decision-making maturity. Those who fail to internalise this will gradually lose their creativity and competitiveness. Faster than many organisations expect. The first effects are already evident in everyday management, but will be most noticeable with the next generation of managers.
2. Systems & Tools: From Feature to Business Model Question
Kai Bättenhausen:
In the software industry, we are seeing that AI is not just another feature, but is fundamentally changing product strategy. This presents companies with questions that no one would have asked just a few years ago (or even a few months ago).
Tobias Piegeler:
Exactly. Systems are becoming interactive, automated and increasingly AI-native. This is not only changing working methods, but is already noticeably shifting market positions. At the same time, developments in recent months have significantly increased the momentum once again: new AI-supported development environments such as Claude Code or Codex-based approaches, as well as increasingly autonomous coding agents, show how software can already be co-developed by AI and, in some cases, generated independently. This is significantly shortening innovation cycles.
This also fundamentally shifts the classic make-or-buy logic. Now that tailor-made solutions can be developed in significantly less time and companies often use only a fraction of standard software, it will be necessary to reassess which software will still need to be sourced externally in the future – and where in-house, AI-supported development creates strategic advantages.
The key strategic questions are:
- Is AI a distinguishing feature for us or just a hygiene factor?
- Do we develop ourselves or do we buy in – and how dependent do we make ourselves on individual providers or foundation models?
- How is AI changing our pricing logic – and thus our margins?
- Which parts of our software and value chain do we need to rethink in the future?
In the software industry, AI is increasingly determining margins, product architecture and competitiveness. Not as an add-on, but as a central lever along the value chain. This is noticeably increasing competitive pressure and will also change role profiles and capacity requirements in the developer job market in the future.
3. Organisation: From pilot to scaling
Henning Sander:
In banking in particular, we see many AI pilot projects but little scaling. Why is that?
Tobias Piegeler:
Because AI is often treated as a standalone innovation project rather than a company-wide transformation task. Successful companies differ in three ways: they prioritise a few economically relevant use cases; they anchor responsibility at board level and cascade it consistently throughout the organisation down to the operational units; and they consistently measure business impact, not just activity.
Many initiatives get stuck in the proof-of-concept stage because the operating model and governance are not adapted. The laboratory is not the problem. Often, the mandate is formally in place. However, there is a lack of consistent implementation, organisational anchoring and the necessary learning and scaling structures.
AI does not need a lab – it needs a mandate. And leadership that lives up to this mandate.
The second part of the conversation focuses on where AI actually creates economic value – and what role governance plays in this.
