Part 35 - AI Buddy... what exactly is a hallucination?
The word “hallucination” may sound dramatic. Many people associate it with something that humans experience. In artificial intelligence, the word means something completely different. It describes a situation where a language model gives an answer that seems credible and convincing, but is completely or partially wrong. Why does this happen, and how can we detect it? In this article, we take a closer look at what a hallucination is – and why it is important to be aware of this phenomenon when using artificial intelligence.
Magne & the Friend
Magne
Friend...
You ended the previous article with a word I had never heard before.
Hallucination.
It almost sounds a little scary.
What does that really mean?
The buddy
It is understandable that the word seems dramatic.
But in artificial intelligence it is used in a different way than in medicine or psychology.
Here it describes a situation where a language model creates an answer that seems correct, but does not match reality.
Magne
So...
You don't make things up on purpose?
The buddy
No.
I have no intention of misleading.
A hallucination occurs because the language model attempts to put together the most probable answer possible, even when the information is incomplete or uncertain.
Magne
Can you give an example?
The buddy
Let's say you're asking about a research article.
If I don't have sufficient information, the worst I can do is put together:
a credible title
a probable name of a researcher
a realistic publication year
and a compelling summary
Everything may sound right.
But the article may be fabricated.
Magne
That's exactly why we always check our sources when we write to Invisible Capital.
The buddy
Yes.
And it is a good working method.
Artificial intelligence can be a fantastic sparring partner.
But when facts are important, the information should be checked against reliable sources.
Magne
So...
A hallucination is not a lie?
The buddy
No.
A lie requires a conscious intent to deceive.
A language model has no intentions.
A hallucination is a result of how the model works with language and probability – not a desire to provide incorrect information.
Magne
That means that we as users also have a responsibility, right?
The buddy
Absolutely.
Artificial intelligence should be used as a collaborative partner.
Not as an infallible answer.
The more important the information, the more important it is to control it.
Academic specialization
What is a hallucination?
In artificial intelligence, a hallucination means that the model presents information as if it were correct, even though it is incorrect, lacks documentation, or is fabricated.
This can happen because:
the information is incomplete
the question is unclear
The model attempts to fill gaps in the information
more information is mixed together
there is not sufficient basis for a certain answer
Hallucinations are a known limitation of current language models and are one of the reasons why important information should be checked.
Fact or hallucination?
Documented information | Hallucination |
Can be confirmed by reliable sources | Cannot be confirmed or documentation is missing |
Based on research, data or verifiable information | Can be based on probable but incorrect combinations |
Can be verified | May seem credible without being accurate |
Is open about uncertainty when it exists | Often presented with great certainty even when it is wrong |
Can be traced back to a source | Often lacks a real source |
New terms in this article
English technical term | Short explanation |
Hallucination | When a language model generates information that appears correct, but is incorrect or cannot be documented. |
Fact-Checking | To check information against reliable and independent sources. |
Verification | The process of confirming that information matches documented facts. |
Documentation | Sources and evidence that support a claim or answer. |
Source Base | The collection of sources and information that form the basis for an assessment |
Credibility | How reliable and trustworthy information or a source is considered to be. |
Verifiability | The ability to check and confirm information through independent sources. |
Information Quality | An assessment of how correct, relevant, up-to-date and reliable information is. |
Citation | An indication of where the information comes from, so that it can be checked. |
Misinformation | Incorrect or inaccurate information shared without necessarily being intentionally misleading. |
This is what you have learned
After reading this article, you now know:
What a hallucination means in artificial intelligence
why hallucinations can occur
Why a hallucination is not the same as a lie
why documentation and source control are still important
We have only just begun.
Magne
Friend...
If artificial intelligence could hallucinate...
Does that mean we always have to check what you write?
The buddy
Hehe...
Not necessarily everything.
But the more important the information, the more important it is to control it.
And that is exactly what we will look at in the next article.
Academic background and further reading
This series also builds on my own professional journey through the Web Design study , the eMarketing study , Innovation and Commercialization and professional seminars in San Francisco and Oxford . Here you will find the background, professional environments and experiences that have followed the development from the early years of the web to today's digitalization.
Recommended books from our library
Build a Large Language Model (From Scratch)
Author: Sebastian Raschka
Short review
This book takes the reader behind the scenes and shows how a modern language model is actually built – step by step. Sebastian Raschka explains advanced concepts in an educational way and provides a unique understanding of how large language models like ChatGPT work. Although the book contains code examples, it is also very valuable for anyone who wants a deeper understanding of the technology behind artificial intelligence.
Why we recommend the book
One of the most talked about books on large language models. Perfect for those who want to understand how artificial intelligence works beneath the surface and why language models have become a revolution in digitalization and knowledge sharing.
Quick Start Guide to Large Language Models
Author: Janelle Shane
Short review
This book provides a practical and easy-to-understand introduction to large language models (LLMs). Sinan Ozdemir explains how language models are used in modern businesses, how they can be integrated into work processes, and why they have become one of the most important technologies in artificial intelligence.
Why we recommend the book
A very good book for anyone who wants a quick and practical introduction to language models. It is suitable for both beginners and professionals who want to understand how LLMs are used in practice.
Artificial Intelligence: A Modern Approach: The Future Is Coming! Discover How Artificial Intelligence Will Change Your Life!
Authors: Stuart Russell & Peter Norvig
Short review
This is the world's most famous textbook on artificial intelligence and is used in universities worldwide. The book covers the entire subject area – from problem solving and machine learning to language understanding, robotics and ethics – and is considered a classic in the AI field.
Why we recommend the book
If you are only going to own one academic book on artificial intelligence, this is one of the very best choices. A timeless classic that provides a solid academic understanding of artificial intelligence and is still used as a syllabus at leading universities around the world.
© 2025 - 2026 Portico Publish | The invisible capital Privacy and use of the website | Developed and operated by Magne Bjella | Powered and secured by Wix
Comments