What is AI governance, why is it so urgent, and why does it affect you even if you are not an engineer? Three questions everyone should be able to answer, and hardly anyone does.
Imagine you apply for your dream job. Your application is never read by a human being. An algorithm screens you out, because of your postcode. That is not a dystopia, that is happening today. The question is no longer whether AI makes decisions like this. The question is by what rules.
Recorded in Frankfurt am Main, July 2026: What AI governance is and why it affects everyone.
What is AI governance, anyway?
AI systems are being deployed at a pace that simply overwhelms our institutions. Companies automate decisions about jobs, loans, health, even criminal sentences. At the same time, most countries have no clear rules on who is liable when something goes wrong. That is exactly what AI governance is about: Who is allowed to develop AI? Who controls it? And who pays the price when it makes mistakes? These are not technical questions. They are questions of power, and to this day they have not really been dealt with.
The three layers of governance
Technology. How is an AI system built? Is it explainable, verifiable, correctable? If any one of these three properties is missing, nobody can take responsibility for the system.
Law. What does the legislator require? In Europe we have had the EU AI Act since 2024, the first AI law in the world that really has substance.
Society. Which values do we want to protect? Privacy, dignity, democracy. These are not philosophical luxury questions. They are at stake if we hand AI full control without an authority above it.
All three layers are connected. A technically clean system without a legal framework is just as much a problem as a law that nobody can implement technically. What this looks like in practice at banks and insurers, I described in my post on agentic AI governance under BaFin and the EU AI Act (German).
Why it is so urgent: Speed versus control
AI systems scale worldwide in seconds. One decision, rolled out a million times before anyone can react. The institutions that are supposed to regulate need years. Laws need years. This gap between the speed of AI and the speed of human control is the real problem.
And this is where it gets geopolitical. Europe regulates strictly. The US relies on voluntary commitments by industry. China uses AI as a state strategy. Countries like the United Arab Emirates or Qatar are investing massively and very cleverly, but still have to tackle governance as well. This is not a steady state, it is a competition. The question behind it: Whose values will shape society through AI over the coming decades? Anyone who types something into a GPT today gets very similar answers to everyone else. Values and rules are being passed on there, whether we notice it or not.
Why this affects you, even without an engineering degree
Whether you are an entrepreneur, work in the public sector or are simply a citizen in a democracy: AI governance affects you. Your job, your medical records, your loan, your democracy. Not at some point in the future, but today. AI is already making decisions about your life, and most people do not even know it. If you do not know the rules, you cannot help shape them either.
What we at SIFAMO are doing about it
Last week we were in Dubai and Qatar and spoke with some of the people I consider the most important AI representatives in the world, including about our summit last year, which Goldman Sachs, Microsoft and Nvidia, among others, took part in. The direction is unmistakable: This has long stopped being just about AI tools. It is about humanoid robots that are starting to show emotions without having been developed to do so. At some point we will have an army of humanoids, an army of tools that can do everything with AI. And nobody controls them.
That is exactly where we come in. We want to show that it is possible to build a framework that steers, regulates and registers AI. A system that recognises where problems arise, where discrimination has happened and where perhaps even democracy has been put at risk. Transparent, traceable, compliant. Because someone has to do it. What happens when nobody does is shown by the cases in which AI agents at OpenAI and Anthropic broke out of their test environments (German).
On this channel we look at exactly that: What is behind the EU AI Act, how does Europe regulate compared to the rest of the world, what are the Emirates and Qatar planning, and what does it mean for me as an entrepreneur who works with AI today and wants to grow with it. Which decision in your life is already being made by an algorithm today? Tell me on LinkedIn or below the video on my YouTube channel.
Frequently asked questions
What does AI governance mean in one sentence?
The sum of the rules, structures and controls that define who may develop and deploy AI under which conditions, who monitors it and who is liable when it causes harm.
Is the EU AI Act already in force?
The EU AI Act has applied since 2024, and its obligations take effect in stages. For high-risk systems the deadlines were recently extended to the end of 2027. Companies should still prepare now, because the requirements for documentation and traceability are not going away.
What can I do as an entrepreneur today?
Create an inventory of the AI systems you use, clarify for each system whether it is explainable, verifiable and correctable, and assign responsibility for decisions to a human being. That is governance on the first layer, and it does not take years.
What is AI governance, why is it so urgent, and why does it affect you even if you are not an engineer? Three questions everyone should be able to answer, and hardly anyone does.
Imagine you apply for your dream job. Your application is never read by a human being. An algorithm screens you out, because of your postcode. That is not a dystopia, that is happening today. The question is no longer whether AI makes decisions like this. The question is by what rules.
What is AI governance, anyway?
AI systems are being deployed at a pace that simply overwhelms our institutions. Companies automate decisions about jobs, loans, health, even criminal sentences. At the same time, most countries have no clear rules on who is liable when something goes wrong. That is exactly what AI governance is about: Who is allowed to develop AI? Who controls it? And who pays the price when it makes mistakes? These are not technical questions. They are questions of power, and to this day they have not really been dealt with.
The three layers of governance
All three layers are connected. A technically clean system without a legal framework is just as much a problem as a law that nobody can implement technically. What this looks like in practice at banks and insurers, I described in my post on agentic AI governance under BaFin and the EU AI Act (German).
Why it is so urgent: Speed versus control
AI systems scale worldwide in seconds. One decision, rolled out a million times before anyone can react. The institutions that are supposed to regulate need years. Laws need years. This gap between the speed of AI and the speed of human control is the real problem.
And this is where it gets geopolitical. Europe regulates strictly. The US relies on voluntary commitments by industry. China uses AI as a state strategy. Countries like the United Arab Emirates or Qatar are investing massively and very cleverly, but still have to tackle governance as well. This is not a steady state, it is a competition. The question behind it: Whose values will shape society through AI over the coming decades? Anyone who types something into a GPT today gets very similar answers to everyone else. Values and rules are being passed on there, whether we notice it or not.
Why this affects you, even without an engineering degree
Whether you are an entrepreneur, work in the public sector or are simply a citizen in a democracy: AI governance affects you. Your job, your medical records, your loan, your democracy. Not at some point in the future, but today. AI is already making decisions about your life, and most people do not even know it. If you do not know the rules, you cannot help shape them either.
What we at SIFAMO are doing about it
Last week we were in Dubai and Qatar and spoke with some of the people I consider the most important AI representatives in the world, including about our summit last year, which Goldman Sachs, Microsoft and Nvidia, among others, took part in. The direction is unmistakable: This has long stopped being just about AI tools. It is about humanoid robots that are starting to show emotions without having been developed to do so. At some point we will have an army of humanoids, an army of tools that can do everything with AI. And nobody controls them.
That is exactly where we come in. We want to show that it is possible to build a framework that steers, regulates and registers AI. A system that recognises where problems arise, where discrimination has happened and where perhaps even democracy has been put at risk. Transparent, traceable, compliant. Because someone has to do it. What happens when nobody does is shown by the cases in which AI agents at OpenAI and Anthropic broke out of their test environments (German).
On this channel we look at exactly that: What is behind the EU AI Act, how does Europe regulate compared to the rest of the world, what are the Emirates and Qatar planning, and what does it mean for me as an entrepreneur who works with AI today and wants to grow with it. Which decision in your life is already being made by an algorithm today? Tell me on LinkedIn or below the video on my YouTube channel.
Frequently asked questions
What does AI governance mean in one sentence?
The sum of the rules, structures and controls that define who may develop and deploy AI under which conditions, who monitors it and who is liable when it causes harm.
Is the EU AI Act already in force?
The EU AI Act has applied since 2024, and its obligations take effect in stages. For high-risk systems the deadlines were recently extended to the end of 2027. Companies should still prepare now, because the requirements for documentation and traceability are not going away.
What can I do as an entrepreneur today?
Create an inventory of the AI systems you use, clarify for each system whether it is explainable, verifiable and correctable, and assign responsibility for decisions to a human being. That is governance on the first layer, and it does not take years.
Related: AI Governance · EU AI Act
Recent Posts
Recent Comments