DATA, AI & CRITICAL THINKING
The Data-Literate
Mind
Searching for Truth in the Age of Data and AI
DATA LITERACY
The same data. A different story.
Photos: Nigel (winnu), CC BY 2.0; Thomas Fuhrmann (Snowmanstudios), CC BY-SA 4.0, via Wikimedia Commons. Plotted coordinates are illustrative.
DATA & REALITY
Data is a choice.
Data is a selective representation of reality.
What it leaves out can matter as much as what it shows.
Every dataset begins with choices about what to notice, measure, and record. Those choices shape the conclusions that follow.
AI can magnify this problem because it learns from existing data and can reproduce its omissions and biases at enormous scale.
THE DANGER OF AVERAGES
The average can hide the climb.
Two routes can have the same average steepness and feel completely different. One may be a steady climb, while the other is mostly gentle with one steep section. The average is identical, but what you experience along the way is not.
The averages match. The climbs do not.
If we make decisions based on the average alone, we may overlook what matters and choose the wrong route.
WHEN DATA BECOMES A PICTURE
The picture is not the numbers.
Which is larger?
30.4 million km²
2.17 million km²
Africa ≈ 14× Greenland
AI LITERACY
The easier question is not always the right one.
“Does this presentation actually provide convincing evidence that the strategy will work?”
“Is the presenter confident and entertaining and the slides polished?”
CRITICAL THINKING IS A SKILL
Good judgment needs more than good intentions.
Critical thinking is not simply about being skeptical or paying attention. It depends on knowing what to question, how to question it, and what to look for.
Six questions worth asking
Of data, stories, and machines.
- DefinitionWhat exactly is being claimed?
- AssumptionWhat must be true for this claim to hold?
- EvidenceWhat evidence supports it, and what is missing?
- AlternativeWhat else could explain the same pattern?
- ConsequenceWhat follows if I act on it?
- SourceWhere did this come from, and can it be traced back?
“Think not those faithful who praise all thy words and actions, but those who kindly reprove thy faults.”
LLMs can sound confident, coherent, and agreeable. That is part of their usefulness, but also part of the problem. Sycophancy can make an answer reinforce what we already believe. Recognizing when an answer is plausible, but wrong, requires enough understanding of the subject, but also data, and AI to question the machine.
FROM DATA TO AI LITERACY
What happens between reality and the answer?
AI literacy starts before AI.
Before a machine produces an answer, reality has already passed through human choices: what to capture, what to measure, and how to interpret it. Data literacy helps us see those choices and understand what AI is building on.
Reality
Answer
WHEN AI FILLS THE GAPS
What you leave unsaid, AI has to fill in.
AI always has to make assumptions when your request leaves something unspecified. Those assumptions may be wrong.
The more clearly you specify what you want, the less AI has to assume. And the more control you have over the result.
THE BOOK
A clearer way to think about our data-driven world.
The Data-Literate Mind is a guide to thinking more clearly in a world shaped by data and AI. It helps you see what others miss, ask better questions, and make wiser decisions.
Available in paperback and Kindle.
ABOUT THE AUTHOR
Making data and AI literacy accessible to everyone.
- Technical University of MunichTeaching & researchSince 2010
- SiemensData & AI in practiceSince 2017 · Data & AI literacy · AI for non-coders
- Enterprise AI platformCloud-based platform
50,000users
- Teaching & enablementWorkshops · training · real-world use casesThousands of people reached
- The Data-Literate MindMaking that knowledge accessible
I have spent much of my career teaching and working with data and artificial intelligence. I began teaching at the Technical University of Munich and later continued this work at Siemens, where my focus has increasingly been on data and AI literacy, including AI for non-coders.
Through teaching, workshops, training, and real-world use cases, I have worked with thousands of people. Along the way, I realized that the questions I was teaching were no longer relevant only to data scientists, analysts, or technology specialists. They were becoming relevant to everyone.
Data and AI are now part of everyday life. They influence what we see, what we are offered, how decisions are made, and how we understand the world around us, often without us noticing. For much of history, literacy itself was accessible mainly to the educated few. Reading and writing eventually became accessible to everyone because they became essential for participating in society. Data and AI literacy now need to follow the same path.
The Data-Literate Mind is my attempt to make that knowledge accessible, not to turn everyone into a data scientist, but to give everyone the tools to question, understand, and navigate a world increasingly shaped by data and AI.
CONTACT