This generative AI training explains how language and image models work in principle, where they help and where they do not, and it is aimed at professionals and managers without programming experience. You learn what tokens, context windows and prompts mean, assess use cases from your own work and recognise typical risks such as hallucinations, privacy questions and bias. No technical background is needed. We tailor content and examples to your sector when you enquire.
What you’ll take away.
- Explain how language models work with tokens, training and next-word prediction in plain words
- Distinguish generative models for text, images and code by strengths and weaknesses
- Write prompts with role, context, task and format so that results can be checked
- Evaluate use cases by benefit, risk, data situation and effort
- Name risks such as hallucinations, data leakage, bias and copyright questions and choose countermeasures
Your training in detail
- Who is it for?
- Professionals, managers, project leads, consultants and anyone who has to assess or introduce generative AI at work.
- Prerequisites
- No technical prerequisites. Some experience with a chatbot is an advantage.
- Format & planning
- Suggested format: 1 to 2 days live online or on site, split into modules. Scope depends on prior knowledge.
Agenda: from understanding to application
Our independently developed agenda proposal. Focus areas, lab environment and scope are agreed before booking.
Placing AI, machine learning and generative AI, with a look at the transformer architecture.
Language models in practice: tokens, context windows, temperature, fine-tuning versus prompting.
Image, audio and multimodal models with typical fields of use and quality limits.
Prompting techniques such as roles, examples, step-by-step instructions and structured output.
Evaluating use cases with a matrix of benefit, risk, privacy and review effort.
Risk and rules: hallucinations, data protection, copyright, an overview of the EU AI Act and the NIST Generative AI Profile.
Suggested exercise: You pick three tasks from your daily work, test them with a language model, check each answer against a reliable source and rate the use case in a risk matrix.
Exams, licences & credentials
There is no single authoritative external exam for this topic. Attending alone awards no vendor certificate. Exam fees and licences are only included if the offer confirms them.
Vendor references & review date
Sources checked on 2026-11-03. These sources describe the technology or external exam requirements. They do not imply a partnership with NextNowa.
Your learning content
These objectives describe the training topic. The specific syllabus, duration, dates, trainer and price are agreed before booking.
Frequently asked questions
Why does a language model sometimes invent facts?
A language model predicts likely next words and does not check truth. When knowledge is missing, it produces fluent but wrong statements, known as hallucinations. You reduce the risk by supplying sources, cross-checking answers and not basing important decisions on them unchecked.
May I enter company data into an AI chatbot?
That depends on provider, contract and type of data. Personal or confidential data belongs only in services that clearly regulate data protection, storage and use for training. Clarify this beforehand with your company's data protection and security teams. The training offers a checklist but does not replace legal advice.
Do I need programming skills to use generative AI?
For everyday use, no; a chatbot is enough. Anyone who wants to build their own applications, for example connected to documents or databases, does need Python skills. This training is designed without programming and helps you decide whether and where going deeper technically is worthwhile.
Does Understanding generative AI include a certificate?
There is no single authoritative external exam for this topic. Attending alone awards no vendor certificate. Exam fees and licences are only included if the offer confirms them.