Circles, Nested
Nested Circles in Computer Science
with LLMs as a starting point
Nested Circles in Computer Science
with LLMs as a starting point
Nested circles in Computer Science,
with LLMs as a starting point.
Figure created by Martin Ågren with help from ChatGPT.
CS - Computer Science:
"Datavetenskap" - Includes: AI, software dev, cybersecurity, etc.
AI - Artificial Intelligence:
- Many definitions
◦ IBM: "Artificial intelligence (AI) is technology that enables computers and machines to simulate human learning, comprehension, problem solving, decision making, creativity and autonomy."
- AI systems can be ML-based or rule-based (e.g. Expert Systems).
ML - Machine Learning:
ML-based AI model learns patterns & relationships from training data to produce output (e.g. classification or generation) without being explicitly programmed for every possible input.
NN - Neural Network:
Type of ML model that adjusts connection weights between artificial neurons (nodes) as it learns patterns from training data.
DL - Deep Learning:
- DL is a type of NN-based ML.
- DL is based on deep NN. "Deep" meaning many layers of nodes.
GenAI - Generative AI:
Modern GenAI uses DL-based AI models to generate new content e.g. code (Codex), images (Midjourney), video (Veo), audio (Elevenlabs), music (Suno).
LLMs - Large Language Models:
- A type of AI model that understands & generates language.
- AI tools based on LLMs: ChatGPT, Claude, MS Copilot, Gemini.
Figure INCLUDING examples
Created by Martin Ågren with help from ChatGPT.
Written by Martin Ågren - September 28, 2026
Swedish: Datavetenskap
Computer Science (CS) is the overarching field that includes areas such as Artificial Intelligence (AI).
Examples of Computer Science that are NOT AI:
Example 1: Web development, where you write code to create a website or web application (and with no AI chatbot on the site).
Example 2: Programs that do NOT exhibit intelligent behavior.
Artificial Intelligence (AI) is a subfield of Computer Science (CS).
Example of AI systems that are NOT ML-based:
Expert systems (rule-based systems without ML) that exhibit human-like intelligence.
Machine Learning (ML) is a subfield of AI.
ML is today often an important technology for developing AI.
How ML works: An “ML-based AI model” learns patterns and relationships from training data to produce output (for example classification/categorization or generation of text/image) without being explicitly programmed for every possible input.
Example of an ML-based AI model that does NOT use NN = Decision trees (neural networks are NOT used in decision trees).
Synonyms / abbreviations:
Artificial neural network
Neural network (assumed artificial)
ANN
NN
A neural network is a type of ML model that adjusts weights, the strength of connections between artificial neurons (nodes), as it learns patterns from training data.
Example of NN without DL = Simple classification, e.g. predicting customer behavior, such as whether a customer is likely to click on a specific link or not.
DL-based ML is a type of NN-based ML. DL uses deep neural networks. "Deep" means that there are many layers of artificial neurons.
Example of DL-based AI model that are NOT GenAI:
Advanced classification, e.g. X-ray analysis.
Modern GenAI uses AI models based on deep learning to generate new content such as text, code, images, video, audio.
Example of GenAI models that are NOT LLMs:
Text-to-image model (an AI model that can generate images).
LLMs are a type of GenAI model that is based on DL.
Examples of LLM-based AI tools: AI chatbots such as:
ChatGPT, Gemini, Microsoft Copilot, Claude.