AI is reshaping business, practice and thinking.
Embedding AI strategically, developing proprietary agent systems, reflecting on societal consequences - three layers of one engagement that cannot be separated.
AI in the enterprise
Successful AI projects do not start with the technology, but with the strategy. The questions companies should clarify before adoption - and the conditions that determine success.
Strategic anchoring
AI is not a tool you adopt on the side - it is a strategic decision. Clear objectives, responsibilities and performance metrics must be defined before any implementation, so that investments produce measurable impact.
Data foundation & compliance
AI is only as good as the data it works with. GDPR, EU AI Act, data ethics and trade secrets impose technical and organisational requirements that need to be addressed early in the architecture - not only once a system goes into production.
People and processes
Team acceptance, qualification and adapted workflows decide the success of an AI project more often than the technology itself. As a rule, AI augments - fully replacing human work is the exception, not the rule.
Economic value
Every investment is preceded by a concrete use case with measurable KPIs and a realistic ROI expectation. AI for its own sake produces cost without effect - pilot projects with a short feedback horizon create clarity before a broad rollout.
spectralQ.ai - Digital investigation platform
spectralQ.ai is my AI-driven forensic investigation platform - it reconstructs a coherent picture from scattered digital traces across three dimensions: digital signals, spatial context and temporal sequences. At its core sits the Murder Board, an interactive workspace where more than 20 data sources - from satellite imagery to Telegram, search trends and AIS shipping data - come together on a shared timeline. Specialised AI agents optionally take over board construction, data retrieval and adversarial hypothesis review.
Murder Board
Interactive investigation workspace: data sources, maps, time series and network graphs are synchronised on a shared timeline - clicking on a data point moves maps and curves with it.
5-agent AI system
Question in natural language - the AI builds the board, selects data sources, retrieves data, adversarially checks for reasoning errors and delivers a forensic report with confidence values.
Forensic integrity
Hash-chained audit trail (SHA-256) on every step - manipulation becomes immediately visible. Open-source core, on-premise operation possible, data stays in-house.
ki-agent.org - The Turing test for the classroom
An interactive group game - the modern Turing test for the classroom. Students ask questions and decide whether the answers come from a human or an AI. In a playful way, the format builds media literacy and critical thinking in dealing with AI.
Smartphone-ready
Works on any mobile device - no login, no installation, no account.
No preparation needed
Ready to use directly in class - teachers can launch the game spontaneously and play it through together with the students.
Critical thinking
A playful format for recognising linguistic patterns and reflecting on the relationship between humans and AI.
How AI is changing our lives
Artificial intelligence is not only at work in economy and technology. It shifts how perception is formed, how decisions are made, and how we understand ourselves as thinking beings. This shift leads into a new social order - cognocracy. Understanding it is the subject of my current publishing work. It is an era in which thinking is increasingly handed over to machines, often willingly and with the best intentions - and precisely for that reason, so hard to grasp.
Making AI visible - interactive animations
Everyone talks about artificial intelligence. And yet, even many of the people who use AI every day barely understand what is actually going on inside these systems.
With my interactive animations I want to make the central mechanisms of modern AI intuitively tangible: clearly explained, visually presented and step by step.
No prior knowledge required. But also without hiding the complexity.