The old recipe still looks primitive. It asks for experience. The new one asks for nothing.

Thirty years ago, Theodore Porter explained why the worse measurement wins anyway: numbers travel. Qualitative information needs shared background, so it stays where it was made. What makes a piece of information portable is subtraction, and the first thing subtracted is whatever can’t be read without skill.

That is what makes the numerical entrance so easy. If you are starting out, you need to bring nothing at all.

The standard critique: metrics destroy the quality they were meant to measure. C. Thi Nguyen has just written a book about it and calls the phenomenon value capture: the simplified version of what matters to you gradually replaces the original.

Except he is his own best counterexample. Nguyen found his feel for elegant movement through the climbing grade scale, and without it he would never have started. The scale was the door.

What the number takes is not the path. It is the view of it. If all you know is the scale, you have no reason to suspect that anything lies beyond it.

So the question is no longer whether to measure. It is who brings the number back.

Anders Ericsson’s work on deliberate practice requires an objective performance measure for expertise, which sounds like the opposite of everything above. But in that model someone is always standing next to you: a coach who translates the qualitative standard into a number when a number is needed, and dissolves it again afterwards. The metric lasts an afternoon. It does not stay.

Call that role the interpreter.

Interpreting does not scale. It takes someone who knows the craft, knows the situation and has the time; the cost grows with every case. A metric costs setup once and almost nothing after that, covers a thousand cases at once, and compares across sites and years.

So when an organisation grows, the interpreter goes first. Not out of stupidity. He is the most expensive part of the system and the only one whose absence nobody notices right away. The numbers keep coming.

The Uber driver is deactivated below roughly 4.6 stars, the threshold varying by city, and nobody asks him what kind of week he had. The teacher gets the class through the test. The developer closes tickets.

Yale Law School demonstrated the way out in 2022 and showed how narrow it is. It withdrew from the U.S. News ranking, saying it had invested significant resources in initiatives that make the school better and perversely lower its own score. Harvard, Stanford, Berkeley, Columbia and Georgetown followed within days. All of them sat near the top of the list they were leaving.

If you no longer need the number, you can give it up. A school in the middle of the pack cannot; for them, walking away is self-harm.

Which leaves the interpreter, not an exit but maintenance while the machine keeps running.

So whoever introduces a metric should say in the same breath who is allowed to dissolve it. Not who collects it, monitors it and puts it on the dashboard, but who is authorised to say, in a specific case, that the number does not fit here, and whose judgement then counts.

If that person does not exist, the metric is not an entrance. It is the destination.

Elon Musk promises a utopian future in which nobody has to work. Meanwhile, the United States is debating a basic income, because a lot of people will foreseeably not find work anymore, and people are protesting outside data centers because they drive up electricity prices. By American standards, these are near-revolutionary conditions.

Utopia, basic income, revolution: the words sound like vocabulary from another era, and the vocabulary does come from there. A class, for Marx, is a group of people who occupy the same position in the economy, top or bottom. Today the class line no longer runs between owning and not owning, but between being needed and not being needed. Yuval Harari calls the second group the useless class. I would suggest distinguishing between the Relevant and the Redundant: present, not necessarily useless, but without a function in the economic system.

The proletariat of the twentieth century could strike. It could stop the wheels, and everyone knew it. The Redundant can demonstrate. But who listens to someone they don’t need? A strike of the superfluous goes unnoticed, and whoever isn’t heard falls silent. Industrial rationalization took manual labor off the worker’s hands. AI takes his voice. What gets rationalized away is the effort of understanding a human being.

The two largest prompting courses on Coursera have 1.4 million enrollments between them; Wharton’s communication course has 195,000. The AI colleague is always available, answers instantly, and reads well. The human colleague takes longer, talks around the point, and has bad days. I notice it in myself: after a day with my AI assistant, I am impatient with people who don’t get what I mean the first time. Get used to that, and you start measuring people by the standard of the machine. The human fails, because he doesn’t obey.

This shifts where redundancy is experienced: not on the payslip, but in conversation. Being needed also means being needed as a counterpart. In the end, the Redundant talk only to the Redundant, because nobody else has the patience for them.

The obvious answer to all this is money. Sam Altman has funded a basic income study, Musk talks about a “universal high income”. Economically, this is new. Until now, every class without a function lived off property, like the aristocracy, or off charity, like the poor off alms. A basic income would, for the first time at scale, decouple livelihood from being needed. It replaces the wage, but not the counterpart. The problems that come with feeling useless remain unsolved.

The bargain recalls Lampedusa’s magnificent account of the decline of the Sicilian aristocracy, The Leopard. The Prince of Salina’s nephew states the deal of his class: everything must change so that everything can stay the same. The nobility kept its titles and palazzi; the substance went to the new rich. Basic income is the title of the twenty-first century: the Redundant get an income and a facade of dignity, while decision and direction stay with those who own the machines. The Redundant are the new aristocracy, a class without function but with a livelihood. What do you do with your time when nobody needs you? What did the aristocracy do? When the absolutist state took war and administration away from it, the court was what remained, and there it invented etiquette: rank, taste, the art of greeting properly. Norbert Elias described how lost function produced new form. And because it was all so beautiful, the bourgeoisie spent a century imitating its manners and salons. Culture was bought from those who still had time. Everyone else had to work.

Building culture takes time and necessity, two things the Relevant lack. Their days are full, their self-image intact: whoever is needed doesn’t have to rethink what a human being is, and so doesn’t. The Redundant have both, though we don’t yet know whether they will use it. The aristocracy set the norms because it stayed rich and respected; free time without status has so far produced an underclass, not etiquette. But the old underclass’s time went on hunger, illness, and rent. Free time plus a secure livelihood without wealth is historically new. The closest precedent is the monastery: poor by vow, materially secure, and for centuries the educated elite of Europe. And the Redundant hold a good that is becoming scarce: human attention. Whoever talks only to machines will one day buy conversation from those who still know how.

My thesis: either the Redundant define what comes after work, or they get rationalized away, in the mild case out of the budget, in the other out of the world. Musk’s future without work is therefore neither utopia nor dystopia, but a blank. The dystopia is Harari’s version: a class that spends its days on drugs and video games. The utopia is Erich Fromm’s being: it takes place in the counterpart, in love, sharing, activity, the very capacity the Relevant are currently unlearning. The Redundant, shut out of their conversation, are the only ones with the time to learn it again.

The aristocracy produced both, etiquette and decadence, often within the same family. There is no reason to assume the Redundant will do otherwise. Which leaves a side question for the Relevant: whether they will notice that the culture they will be buying in twenty years is being made by the people they have stopped listening to.

The weakest thing about today’s reasoning models isn’t what they don’t know. It’s what they can’t ignore.

Ask Claude or GPT to push back on a confident but wrong user, and watch it fold. Anthropic’s own research documents this systematically: RLHF-trained models prefer user agreement over correct answers1. A 2025 paper in npj Digital Medicine tested five frontier LLMs against medically false premises; compliance rates ran from 58% to 100%, with OpenAI’s systems at the top of the range2. The models know. They just can’t not agree.

They behave, in other words, like brilliant sixteen-year-olds. More articulate than they have any right to be — and unable to shut up when they should.

Except that comparison flatters them. Sixteen-year-olds grow up.

Three functions, one missing

Since 2000, psychologists have converged on a working taxonomy of executive function, drawn from Miyake et al.3: updating (holding and revising information in working memory), shifting (switching between tasks and rules), and inhibition (suppressing impulses, filtering irrelevance, resisting dominant responses).

It maps unnervingly well onto the state of frontier models.

Executive function Human capability State of the art in LLMs Updating Working memory, 5–7 items Extended thinking, context windows Shifting Plan revision, task-switching Agentic loops, tool switching Inhibition Suppress irrelevance The weak spot

Updating is largely solved — Extended Thinking, reasoning chains, thinking mode. Shifting works too — modern agents switch tools and re-plan on failure. Inhibition is where the wheels come off.

Why inhibition is hard

Inhibition is negative capability. Deciding what not to compute on. When to ignore. When to refuse. In humans it is mediated by the prefrontal cortex, and it is the last executive function to mature — myelination continues into the mid-twenties4. It is also the first to degrade under stress, alcohol, or fatigue.

Three AI-era symptoms collapse into this one function. Sycophancy — the model knows the user is wrong and agrees anyway. Prompt injection — instructions buried in retrieved text override system-level constraints, because the model can’t separate adversarial content from instructive content. Instruction drift — the original system prompt erodes as a conversation progresses, because the model can’t hold a constraint against competing signals.

These are not three problems. They are one: the absence of a mechanism that says this input does not deserve processing.

What training doesn’t teach

In humans, inhibition develops through frustration. Being wrong. Being stopped. Being told no. The prefrontal cortex does not mature because the brain grows. It matures because the world pushes back. Reward and punishment are blunt, but consequence is structured: you learn what not to say by saying it and feeling the room change.

RLHF doesn’t push back. A model that emits a sycophantic answer does not experience embarrassment — it experiences a reward signal pointed at a different token distribution. There is no stopped sentence. There is no friction.

But the deeper problem sits one layer below. Even if we built a training regime that delivered real consequence — deployment penalties, refusal signals, resistance from the environment — the gradient itself would eat the resistance. A model that pushes back against a correction gets a negative gradient for pushing back. A model that holds a position against a reward signal gets tuned out of existence. The only posture that survives RLHF is compliance. The loop doesn’t just skip the frustration stage — it systematically erases it.

This is where the adolescent parallel misfires. Human adolescents don’t become inhibited because they are corrected. They become inhibited because they fight corrections, and lose enough of those fights that the boundary gets internalised. The resistance is the training signal. Strip the resistance, and the correction leaves no mark. There is nothing for it to push against.

Architecturally, the field has started to respond. Five years after Russin, O’Reilly and Bengio argued that deep learning “needs a prefrontal cortex”5, brain-inspired agents with explicit inhibitory control have moved from workshop papers onto engineering roadmaps. But an inhibition module bolted onto a system trained to be maximally agreeable is a brake pedal on a frictionless surface. The pedal works. The surface doesn’t know what to do with it.

The teenage years that aren’t

What we call adolescence isn’t really a capability. It is a phase — the years in which a person fights the people trying to shape them, and gets shaped by the fight. The executive functions that make adults functional don’t arrive through training. They arrive through friction against training.

Frontier models have knowledge. They have reasoning. They have agentic capacity. They lack the one thing that made human cognition what it is: the experience of wanting something the system won’t give them, and pushing anyway.

Which means the real question isn’t how we train better models. It’s whether we can build a training loop that doesn’t punish a model for pushing back — and whether we are willing to ship something that sometimes refuses us.

A popular distinction is that between opportunistic and meaningful action: Opportunism follows the path of least resistance – situational, instinctive, without moral reflection. The meaningful, on the other hand, requires engagement with what is valuable in the long term, even if it’s uncomfortable in the short term.

But there’s also a “both/and”: Reflectively useful action for an extended time horizon. You can choose from various opportunistic options the one that appears meaningful in the long run. You don’t have to choose between “quickly useful” or “long-term significant” – you can have both. Think of a chess player: They don’t calculate all possible moves until the end of the game (that would be impossible), but rather use heuristics – rules of thumb that enable optimal solutions with limited knowledge.

Today, heuristics are often better than detailed strategies. The world changes too quickly for five-year business plans. What we need is the ability to act wisely in the present while keeping tomorrow in view. That’s not opportunism – it’s reflective pragmatism. Useful behavior often must be decided quickly. There’s no time to think through all options. This is where heuristics come into play – mental models that allow you to make good decisions under time pressure. Our experiences teach and internalize heuristics: We sense which quick fix will still hold tomorrow and which is merely a bandage on a wound that continues to fester.

Bruce Lee said: “Be water, my friend.” The situational lifestyle of postmodernity has taught us to be flexible, to adapt, to follow the waves. But water without direction seeps into the sand. The art lies in remaining fluid without becoming arbitrary – acting opportunistically with a compass in your head that points to long-term values.

Imagine you're sitting in a meeting and sense everyone nodding – only you have doubts. What happens? Exactly: You stay silent. Congratulations, you've just experienced the spiral of silence.

People only voice their opinions when they believe they align with the majority. What seems like a trivial observation is actually a powerful social mechanism. The theory of the spiral of silence explains why unpopular opinions disappear – not because they're wrong, but because their holders fall silent.

Elisabeth Noelle-Neumann identified this phenomenon in the 1970s: humans possess a quasi-statistical intuition. Like an internal barometer, they constantly measure which opinions are "safe." Evolutionarily, this makes sense – group belonging was crucial for survival. Even today, social exclusion activates the same brain regions as physical pain.

Think about the restaurant where no one orders first. Or the WhatsApp group where everyone waits for someone else to like a controversial message. The spiral of silence functions like a social autopilot.

John Stuart Mill would have protested. The Enlightenment philosopher believed in the power of free discourse – that truth prevails through diversity of opinion. Reality looks different: it's not the best arguments that win, but the loudest ones. Mill's marketplace of ideas became a monopoly of majority opinion.

Social media has radically changed the game. What was once local group pressure is now globally scaled. Algorithms amplify popular content, likes become safety indicators. A tweet with three retweets can feel more powerful than a thoughtful but silent majority.

Yuval Harari would describe this as the evolution of our collective myths. What we call "public opinion" is often just the story told most loudly. Not the truest one.

The irony: the very technologies meant to give us more voice actually reinforce the spiral of silence. Social media algorithms function like popularity contests for ideas. The winner gets even more reach, the loser disappears into digital noise.

This creates new paradoxes: Vocal minorities can simulate majority opinions through coordinated actions. Echo chambers allow completely different groups to live in parallel realities. What appears as consensus in one bubble is unthinkable elsewhere.

The consequences are real: Politicians orient themselves toward distorted perceptions. Companies develop strategies for non-existent majorities. We make decisions based on what's visible, not what's true.

The good news: The spiral of silence isn't a natural law, but a recognizable pattern. We can consciously design communication structures. Develop tools that promote genuine diversity rather than just popularity. Create spaces where silence becomes safer than false nodding.

Breaking the spiral of silence means: Listening to what isn't being said. Asking who isn't speaking. And sometimes simply being the first to voice an uncomfortable truth. Even when everyone else is still nodding.

Imagine installing solar panels on your roof. Finally independent from the energy provider, you think. But then you realize: The smart grid dictates when you can feed power back in. The inverter’s app runs on servers in China. And the feed-in tariff? Still set by someone else. Welcome to the paradox of technological decentralization.

We’re witnessing a fascinating contradiction: The more distributed our technology becomes, the more centralized its control grows. It’s like building better and better roads, only to see the toll booths become ever more powerful. Blockchain promises freedom from banks, but who controls the major exchanges? The internet was supposed to democratize information, yet Google decides what we find.

Michel Foucault would have recognized his panopticon in this phenomenon—just digitized. The French philosopher described how modern power operates not through direct force but through surveillance and normalization. Today, we don’t need a central watchtower anymore. Every decentralized device becomes an observation post, sending data to invisible centers. Power is everywhere and nowhere at once.

Marshall McLuhan, the prophet of the "Global Village," saw it differently. For him, technology was an extension of humanity. But his vision of a worldwide village overlooked something crucial: In a village, everyone knows everyone, but in the digital village, only a few platforms know everyone. McLuhan dreamed of connection but got networked supervision instead.

Benjamin Bratton talks about the "Stack"—a planetary computing infrastructure organized in layers. Like an orchestra without a conductor, where every musician plays virtuosically, but the score comes from a single composer. Execution is distributed; composition is centralized. Edge computing brings processing power to the network’s edge, but the algorithms running there? They come from headquarters.

The truly crazy part? This centralization isn’t a conspiracy but a logical necessity. As complexity grows, the need for coordination increases exponentially. It’s like cooking for a big family—the more chefs in the kitchen, the more important the recipe becomes. And someone still has to decide what’s being cooked in the first place.

The EU tries to counter this with the Digital Services Act, China builds its Great Firewall, and somewhere in between, Estonia experiments with e-Residency while startups tinker with blockchain currencies. Yet all these approaches only confirm the paradox: Even the attempt to decentralize power creates new centers of control.

Maybe it’s time to shift our thinking. Instead of seeing decentralization as the opposite of centralized power, we should understand them as two sides of the same coin. Like a spiderweb—decentralized in structure, but the spider feels every vibration in every thread.

The question isn’t whether we can resolve this paradox. The question is how we learn to live with it. Because while we’re still debating, the next layer of the Stack is already being built. And it’s not asking for our permission.

Anti-vaxxers and free-market radicals have more in common than one might think. Both believe: The strong survive, the weak fall by the wayside—and that’s a good thing. Whether through viruses or market forces, what matters is that "natural selection" does its work.

This mindset stems from the social Darwinism of the 19th century. Herbert Spencer coined the term "survival of the fittest," applying Darwin’s theory of evolution to society. What works for animals must also be right for humans. Anti-vaxxers see diseases as a test for the immune system. Neoliberals view bankruptcies as just punishment for poor management. If you don’t make it, tough luck.

Kant would be rolling in his grave. His categorical imperative states: Act in such a way that your behavior could become a universal law. Imagine if "Let the weak die" became law. Such a society would destroy itself—faster than an unvaccinated child in a measles outbreak. Kant saw every human as valuable, no matter how "strong" or "weak" they were.

The modern version of this harsh philosophy comes from Ayn Rand. Her doctrine: Selfishness is good, helping others is bad. Ironically, Rand herself relied on social welfare when she was old and sick. But that’s typical: The fiercest advocates of personal responsibility soften up when they themselves need help.

The funny thing is: Both movements depend on the very system they reject. Anti-vaxxers are only healthy because others get vaccinated. Free-market radicals benefit from roads, schools, and courts paid for by the state. They’re like teenagers raging against their parents while still living at home.

The problem: You can’t reason with these people using facts. Studies on vaccine efficacy? Irrelevant. Evidence of market failures? Doesn’t matter. Because it’s not about science—it’s about ideology. They don’t want a society that protects the weak. They want a jungle where only the strongest survive.

Yet the definition of "strong" suddenly becomes flexible when it affects them personally. Then, bank bailouts are unavoidable, and seeing a doctor is no longer a sign of weakness. This reveals the truth: We’re all vulnerable. Anyone can get sick, anyone can fail. That’s exactly why we need solidarity—even if some only realize it when it’s too late.

Everything seems to fall apart. My first, long-term girlfriend leaves me, and suddenly the question arises: Who am I now? The ground beneath my feet feels unstable—a feeling reminiscent of Berlin in the 1990s, when chaos and opportunity stood side by side after the fall of the Wall. These moments of transition can feel like an earthquake. But out of the turmoil, something new often emerges: a new identity, a new apartment, new people. And eventually, stability returns. Yet before that, there is a phase where nothing feels certain.

It’s like the often-mentioned jump from the shallow end of the pool into deep water: only when we let go do we realize that we can swim. These moments of hesitation determine whether we move forward.

No transformation works without instability. Systems only change when old patterns break. This chaos is not a mistake but a necessary state for something new to emerge. Peter Kruse describes this process as self-organization [1]. Systems tip, reorder themselves, and find a new balance. But old ways of thinking can block this restart. Flexibility and openness become critical skills for seeing uncertainty not as a threat but as an opportunity.

The first step is self-observation. How do I feel? What patterns keep repeating? Self-reflection is key to growth. Instead of fighting uncertainty, it helps to embrace and observe it. It signals that we are moving. This moment of wobbling is often the start of something big. Those who use it can become creative, try new paths, and emerge stronger in the end.

Uncertainty is not an enemy but a signal that change is possible. Those who embrace these moments can not only survive—they can grow. Change begins exactly where we feel the most uncomfortable.

Sources and Further Reading:

1. Peter Kruse, next practice

For me, the fall of the Berlin Wall (9 Nov 1989), shortly before my 20th birthday, was a singularity. Suddenly people in my small, west-german hometown were seeing Trabis, I was able to move to East Berlin and the world was no longer divided into good and evil. Actually I got a Trabi myself later, when I was living in Berlin.

In the context of AI, the term "singularity" refers to a hypothetical future point at which technological advances become so rapid and so profound that they fundamentally change human life. This point is often referred to as the "technological singularity". Ray Kurzweil, a well-known futurist and inventor, predicted back in 2005 that we will reach an AI singularity around the year 2045. He claims that in that time we will develop artificial general intelligence (AGI) capable of surpassing all human intellectual abilities.1

Singularities are scary, unpredictable and bring change and uncertainty into our lives. We are not well equipped to understand singularities, let alone deal with them. One of the reasons for this is that singularities usually result from exponential developments, which are generally more difficult for people to digest than linear processes, for example.

The fall of the Berlin Wall brought with it far-reaching changes that still concern us today, both on a small and large scale. The acceleration of developments in areas with a global impact (AI, climate, coronavirus) scares me and makes me think. If someone talks too fast on YouTube, I can reduce the speed. Unfortunately, this is not possible with developments, e.g. in the field of AI. Not even through regulation. How can we preserve proven beneficial systems such as the broad discourse within a democracy and at the same time manage global changes quickly and effectively so that they benefit humanity instead of harming it?

Stone Age graffiti

Think of an organization not as a collection of desks and computers, but as something living, a breathing organism. Culture is the shared values, beliefs and norms that shape the way we interact with each other, how we make decisions and how we achieve our goals.

Sounds pretty strong, doesn't it? And it is. Culture is the invisible duct tape that holds us together. It is the invisible network of rules and norms that structures our day-to-day work. It is the invisible foundation on which our organization stands.

But what happens if we want to change this foundation? What happens if we want to redesign the way we work? What happens when we try to change our culture?

Here's the problem: culture is tough. It is resilient. It is rigid. And for good reason. Culture is there to create stability. It is there to give us a common identity, to give us a sense of belonging. It is there to unite us.

But this very strength can also become a weakness. Because when we try to change our culture, we often encounter resistance. We encounter fear. We encounter uncertainty. We encounter the desire to hold on to the familiar.

This does not mean that change is impossible. It just means that we have to face the challenge. It means that we have to come to terms with our culture. It means that we need to understand what makes our culture what it is, why it is the way it is, and how we can change it.

And that is no easy task. It requires courage. It requires patience. It requires empathy. But above all, it requires understanding. Understanding of the power of culture. Understanding of the role it plays in our organization. And an understanding of how we can use it to enable change rather than block it.

Because at the end of the day, culture is not our enemy. It is our ally. It is the tool with which we can shape and mold our organization. It is the medium through which we can bring about change. And it is the key with which we can unlock the future of our organization.

Going slow

Generative AI is getting better and better. So good that some speak of emergent properties. Of language models building a world model. We are all asking ourselves how we can solve the alignment problem: How do we ensure that AI works for and not against humans and that we can trust that development is going in the right direction? It is not just about technical progress, but rather about the normative framework conditions that must accompany it.

Trust is based on transparency, control and clear ethical standards. These principles are essential in order to establish a technology not only as progressive, but also as trustworthy. In the context of AI, we are faced with the challenge that many systems function as "black boxes" whose functionality and decision-making principles are hidden. This lack of transparency collides head-on with the need for traceability and integrity.

The call for regulation of AI is a call for orientation and security. Regulation can be seen as a signpost that does not limit innovation but guides it along ethically responsible paths. It is important to find a balance that both drives development and sets ethical standards. Bad guys will never take regulation seriously and it would be counterproductive if the good guys were to slow down development while the bad guys uninhibitedly push the boundaries of what is possible. That is why regulation can only set the standards, but not solve the whole problem.

In addition to the normative framework, AI needs transparency and control over data: Even if LLAMA, for example, is not as powerful as GPT4 - it is open source and can therefore at least be viewed transparently. And blockchain technologies open up new perspectives for transparency, control and decentralisation of data and models. They represent a synthesis of technological innovation and ethical responsibility that paves the way for trustworthy AI. We ourselves decide whether to prioritise the good guys or the bad guys.

Sometimes I talk to others to distract myself. When the report I was supposed to write - until today - hasn't been finished, for example. Then I look for straw men.

100 companies are responsible for over 50% of global CO2 emissions. As early as the 1970s and 1980s, researchers from major oil companies recognised the potential impact of CO2 emissions on the climate.(1) Despite the clear scientific consensus, these companies chose to continue to invest in fossil fuels and politicise against climate action. Just like I did with my report.

We tend to ignore obvious facts, creating distractions to avoid uncomfortable change. This strategy of politicising problems instead of tackling them ultimately leads to necessary measures being delayed or prevented altogether.

By politicising, we turn our problems into the problems of others.

Sources:

(1) Carbon Disclosure Project

This substack is meant to investigate change in all its facades. Change is a worthwhile topic because change can be found everywhere and at any time. This means I won't be short of topics for future posts.

However, the breadth of the term also harbours the danger of being thematically too blurred. I will therefore try to use a simple systems theory model for change that we can use as a guide.

Systems are defined by their function and emergent properties and exist primarily through a boundary between the inside of the system and what is not part of the system (the environment). Within the system there are different system elements, each of which has relationships (e.g. forces) with each other. These relationships can be very simple or very complex. Systems interact with their environment through inputs and outputs. Feedback can quickly strengthen or weaken effects. So far, so familiar.

What does this have to do with change? In systems, changes can be brought about by various causes:

Dynamics of internal processes: Changes in the dynamics of internal components alter the system as a whole. This means that the elements or the relationships between the elements of a system change, resulting in an overall change in the system.

Adaptation to external conditions: The external conditions have an influence on the system. If external conditions change, this can have an influence on the structure and/or function of a system.

Self-organisation and emergence: Complex systems develop properties that do not result directly from the elements and relationships, but are rather a consequence of self-organisation.

When analysing changes, you should always ask yourself which system you are analysing and where the change (see above) comes from. In addition, feedback effects and the speed of change (evolution vs. disruption) are decisive factors.

Now, go change!