Why we are rethinking AI at M2 – and what Dr. Faruch Amini is doing to reinforce this
AI 2025 – the status quo in practice
By 2025, we will hardly see any AI that does not work technically.
What we will see is AI that has not been thought through organizationally.
In projects, architecture reviews, and customer meetings, we encounter powerful models, fast PoCs, and convincing demos. What is missing is rarely intelligence—but rather responsibility: for operation, integration, and further development.
AI delivers results.
But it only has an impact when it is built as part of a system.
Based on this experience, we at M2 are refocusing our view of AI. Not as an isolated project, not as a playground for innovation – but as a system capability that works in the long term.
Our goal is clear:
to consciously develop PoCs in such a way that they are designed for the long term – and to consistently integrate them into existing organizations, processes, and system landscapes.
This text is therefore not about a short-term impulse.
It is about consciously developing our attitude toward AI – and the consequences we draw from it.
Chapter 1: The real turning point
AI has arrived in many organizations.
But rarely where it actually changes work.
What we are seeing is not a lack of ideas or technology. There are enough pilots, demos, and initial successes. What is missing is commitment.
AI projects get off to a quick start, generate attention, and deliver results. But as soon as integration, operation, or clear responsibilities are required, things come to a standstill. Decisions are postponed, responsibilities remain vague. The next pilot project is launched faster than the previous one has been properly completed.
This creates momentum—but no substance.
And above all, no reliability in everyday life.
The real turning point therefore does not lie in the next stage of innovation.
It lies in an uncomfortable question:
Are organizations prepared to treat AI as productive infrastructure—with clear responsibilities, ongoing costs, and permanent operation?
This is precisely where the pilot project diverges from practice.
Chapter 2: When attitude becomes direction
Against this backdrop, we are consistently developing our view of AI at M2. Not as a change of course, but as a conscious evolution: away from isolated experiments and toward the targeted integration of AI systems into corporate value creation.
We are not concerned with producing more ideas.
Rather, we want to continue developing good ideas – beyond the initial success.
Not leaving AI where it attracts attention, but bringing it to where it becomes part of everyday life.
Important to note:
This further development is not a result of an increase in personnel.
It is a strategic decision by M2 as an organization.
To consistently implement this approach, we are strengthening it in a targeted manner – among other things, with the help of Dr. Faruch Amini. His background combines scientific depth in strategic foresight and intelligence analysis with years of execution experience in regulated and security-critical environments.
Faruch thinks about AI not in terms of models, but in terms of the overall system: from data and architecture to governance and operation to the question of what makes a solution sustainable in the long term.
He is not a soloist, but part of a claim that M2 as an organization supports:
Not only making AI operational – but also permanently usable.
Chapter 3: What remains is crucial
AI must not only function.
It must be permanently integrated, deliver reliable data, and withstand everyday use.
In practice, this means a change of perspective: away from individual models and toward the interaction of data, systems, and operations. What matters is not what is convincing in a demo, but what runs stably over time, remains traceable, and can be further developed.
Especially in demanding environments—public administration, regulated industries, security-critical contexts—relevance is not demonstrated at the moment of go-live, but in ongoing operation: under regulation, under supervision, and with real users.
Anyone who talks about AI without considering integration and operation is not talking about productive solutions—but about experiments.
Chapter 4: No system, no everyday life
What does that mean specifically?
Solutions do not take effect alongside the organization, but only when they become part of it: embedded in processes, decisions, and existing systems—so that they function stably in everyday life without requiring ongoing support.
To achieve this, operation, responsibility, and control must be considered from the outset. Not as a safeguard after the fact, but as a prerequisite for genuine use.
An isolated use case is not progress.
It is a risk.
The real lever is not in the interaction with the model.
It lies in the system we build around it – or don't build.
Chapter 5: Responsibility, consistently thought through
With this further development, we at M2 are not opening a loud, disruptive chapter.
We are sharpening one that already exists.
M2 continues to stand for technical substance, delivery standards, and responsibility for feasibility. What is new is the consistency with which we think about AI systemically—and specifically translate it into resilient, scalable operating models.
The goal remains clear:
AI should relieve people, improve decisions, and make organizations more capable of action.
No more. But above all, no less.