Artificial intelligence
Language models, machine learning, agents, ethics and responsible use of AI.
AI and the future of work: changing tasks and the skills you need
How AI is reshaping work: why tasks change more than whole jobs, which activities are most exposed and which skills are worth developing right now.
AI and software development: using coding assistants wisely
How to use artificial intelligence in software development: where coding assistants truly help, which errors they introduce and how to review their output.
AI and defensive cybersecurity: detecting anomalies and threats
How AI supports cyber defence: anomaly detection, anti-phishing filters, log analysis and help for security analysts, along with its real limits.
AI for small businesses: where to start without wasting resources
A practical path for bringing AI into a small business: concrete use cases, costs, internal rules, staff training and measuring the first results.
AI hallucinations: why a model makes things up and how to cope
Why language models sometimes invent facts, sources and quotations, which requests carry the most risk and how to check answers before relying on them.
Algorithmic bias: where it comes from and how it can be reduced
What bias in AI systems is, how it enters data and models, how it is measured and which practices genuinely help to reduce it in real projects.
Automation with AI: which processes to hand over to machines
How to decide which business processes to automate with AI, how to design the workflow, where to keep human oversight and what you should measure.
Context window and memory: why AI sometimes forgets
What a language model's context window is, why long conversations lose details, how chatbot memory works and how to manage both well in your daily work.
Embeddings and semantic search: how AI captures meaning
What embeddings are, how they turn text into vectors, how semantic search works and where it is used: archives, support desks, recommendations.
Ethics of artificial intelligence: responsibility and human control
The practical ethical questions raised by AI: who answers for errors, transparency, human autonomy, environmental impact and copyright issues.
The EU AI Act: key concepts of the risk-based approach
The core concepts of the AI Act, Regulation (EU) 2024/1689: risk levels, prohibited practices, high-risk systems, transparency duties and roles.
How to evaluate an AI service before adopting it: a checklist
A practical checklist for evaluating an AI service: quality on your real cases, data handling, security, costs, contract terms and an exit plan.
Fine-tuning or prompting? How to adapt a model to your needs
When good instructions are enough, when you need RAG and when fine-tuning a language model is worth it: costs, data required, pros and cons.
Generative AI for audio: how speech synthesis and music work
How AI generates voice and music: digital audio, spectrograms, the speech synthesis pipeline, neural vocoders, audio codecs and music models.
How a language model works: tokens, probability and context
What happens when you type to a chatbot: tokens, next-word prediction, context and training explained without formulas and with practical examples.
How AI generates images: diffusion models explained
How AI image generators work: noise, diffusion models, latent space, text guidance and the main parameters that shape the final picture you get.
Computer vision: how a machine recognises objects
How computer vision works: pixels, convolutional networks, classification, object detection and segmentation, with practical applications and limits.
Local or cloud AI: advantages, limits and hardware requirements
Running an AI model on your own computer or through a cloud service: privacy, performance, costs, quantisation and the memory you actually need.
Machine learning explained simply: learning from examples
What machine learning is, how it differs from traditional programming and the main types of learning: supervised, unsupervised and reinforcement learning.
Neural networks explained: artificial neurons, weights and layers
How an artificial neural network works: neurons, weights, activation functions, layers and backpropagation of error, explained with everyday examples.
Privacy and AI: what you should never paste into a chatbot
Which data not to share with chatbots, how to read the settings on conversation retention and which habits protect your privacy when you use AI services.
RAG: connecting a language model to your company documents
How Retrieval-Augmented Generation works: document retrieval, answers, citations, typical errors and the checks a company assistant needs before launch.
Training data: why quality matters more than quantity
Where AI training data comes from, how it is cleaned and labelled, and which dataset flaws make a model unreliable once it is put to work in practice.
AI agents: what they are and how they differ from a chatbot
What sets an AI agent apart from a simple chatbot: tools, planning, memory and autonomy, with use cases, concrete risks and rules for using agents safely.
Effective prompts: how to ask AI for what you actually need
How to write effective prompts for language models: context, goal, format, examples and constraints, with reusable templates and the most common mistakes.