1. The AI tools ecosystem – how to find your way in it
Across the AI landscape, tools differ not only by brand but by role in the workflow. This section organizes the ecosystem into key categories, explains the differences between chatbots, assistants, generators and specialist solutions, and helps assess free, paid, general-purpose and dedicated options.
- the most important categories of AI tools,
- differences between chatbots, assistants, generators and specialist tools,
- general-purpose models versus dedicated solutions,
- free and paid versions of tools,
- how to quickly assess what a given tool will work best for.
2. Overview of the most popular generative AI tools
ChatGPT, Gemini, Claude, Grok, Microsoft Copilot and related tools are compared through the way they support real work. The section examines their main capabilities, strengths and limits, differences in handling text, data, files and web access, and when using more than one tool makes sense.
- ChatGPT, Gemini, Claude, Grok, Microsoft Copilot and other solutions,
- the most important capabilities of individual tools,
- strengths and limitations,
- differences in the way of working with text, data, files and Internet,
- when it is worth using more than one tool.
3. How to choose an AI tool for a specific task
Choosing the right AI tool begins with defining the task, not with picking the most popular platform. The section maps solutions to work with text and communication, documents, data and spreadsheets, research, presentations and visuals, as well as work organization and productivity.
- defining the goal before choosing a solution,
- selection of AI for creating texts and communication,
- AI for analysis of information and documents,
- AI for working with data and spreadsheets,
- AI for research and searching for information,
- AI for presentations, graphics and visual materials,
- AI for work organization and increasing productivity.
4. AI for search and research
When research relies on AI, the key issue is not only speed but also source quality and answer verification. Participants compare traditional and AI-supported search, use models to gather and synthesize information, evaluate sources, and reduce the risk of outdated or incorrect conclusions.
- traditional search engines vs. AI-supported search engines,
- searching for information using the use of AI models,
- analysis i comparison of sources,
- creating summaries of a larger amount of information,
- verification of answers generated by AI,
- reducing the risk of incorrect or outdated information.
5. AI in working with documents and content
Document-heavy work often requires extracting, rewriting, and adapting content efficiently. This part focuses on summarizing reports, finding information in large materials, comparing documents, creating new text from existing sources, editing and simplifying language, and translating for specific audiences.
- summarizing documents and reports,
- searching for information in large materials,
- comparing documents,
- creating new content based on existing materials,
- editing, shortening and simplifying texts,
- translating and adapting content to the audience.
6. AI as support for everyday work
Daily operational tasks are often the fastest area for useful AI support. The material includes drafting messages and replies, preparing notes, summaries and task lists, outlining action plans, generating ideas and alternatives, organizing information, and assembling materials for meetings.
- preparing messages and responses,
- creating notes, summaries and task lists,
- preparing action plans,
- generating ideas and solution variants,
- organizing andinformation,
- preparing materials for meetings.
7. Prompting – how to communicate effectively with AI
Effective prompting is based on giving AI a clear objective, enough context, and an explicit expected outcome. Participants practice defining role, audience, and response format, refining instructions iteratively, and improving answers through follow-up interaction instead of restarting each time.
- what a good prompt is,
- defining the goal, context and the expected result,
- indicating the role, recipient and response format,
- clarifying instructions,
- iterative work with AI,
- improving responses instead of starting from scratch.
8. Prompting techniques increasing the quality of responses
Better responses come from structuring prompts with intent rather than asking once and accepting the first result. The section develops prompts step by step, adds examples, criteria and constraints, requests multiple variants, compares outputs across tools, and builds reusable prompt templates.
- creating prompts step by step,
- using examples,
- asking AI additional criteria and constraints,
- asking for several solution variants,
- comparing results from different tools,
- building your own prompt templates.
9. One tool or several? Building your own AI Toolbox
Instead of collecting too many overlapping apps, the goal is to assemble a small set that fits everyday work. Participants define primary and supporting tools, learn how to move outputs between them, combine functions across applications, and design a simple workflow matched to their tasks.
- selection of a set of tools for everyday work,
- defining the main and supporting AI tool,
- combining the capabilities of different applications,
- transferring results between tools,
- avoiding duplication of functions and an excess of applications,
- creating a simple AI workflow AI tailored to one's own tasks.
10. Multimodal AI – text, image, voice and files
Text is only one input channel in modern AI workflows. This section covers working with images and photos, analyzing PDFs and presentations, using voice in interactions, generating visual materials, and understanding the practical capabilities and constraints of multimodal models.
- working with images and photos,
- analysis of PDF documents and presentations,
- use of voice in working with AI,
- generation of graphics and visual materials,
- capabilities and limitations of multimodal models.
11. AI for automation and more advanced applications
Automation becomes relevant when repetitive work can be delegated beyond a simple chat interaction. Here the focus is on the distinction between chatbots, assistants, and AI agents, cooperation with other applications, and judging when an automated flow is justified and when a basic chatbot is enough.
- the difference between a chatbot, an assistant and an AI agent,
- automating repetitive tasks,
- AI working with other applications,
- the basics of AI agents and automation based on AI,
- when automation makes sense, and when an ordinary chatbot is enough.
12. Safety and responsible use of AI tools
Safe use of AI starts with deciding what information can be shared and what must stay outside the tool. The section addresses confidential, personal, and company data, differences between private and business accounts, hallucination risk, and a limited-trust approach to generated content.
- what data can be provided to AI,
- and confidential information, personal data and company data,
- differences between private accounts and business solutions,
- the risk of hallucinations and incorrect answers,
- the principle of limited trust in generated content,
- best practices for safe use of AI.
13. How to evaluate new AI tools
Before adopting another AI product, it is worth checking whether it solves a real problem rather than only looking convincing in marketing materials. This part focuses on assessing output quality, usability, integration potential, security, data handling, and cost versus actual value.
- whether the new tool actually solves the problem,
- functionality versus attractive marketing,
- quality of results and ease of use,
- the possibility of integration with the current work environment,
- security and the data processing model,
- cost in relation to actual value.
14. Effective work with AI – good practices
Efficient AI use depends on knowing where it genuinely reduces effort and where human judgment should remain primary. Participants work on quality control, combining domain expertise with model output, and setting internal standards for consistent day-to-day use of AI tools.
- when AI actually saves time,
- tasks that are not worth delegating to AI,
- controlling the quality of results,
- combining human knowledge with the capabilities of AI,
- creating your own standards for working with AI tools.
15. Building your own AI Toolbox – summary
Your final toolbox is organized around the tasks you handle most often, with clear roles for core and supporting solutions. The section also includes defining use cases, assembling a reusable prompt set, and outlining practical next steps for developing AI skills over time.
- tool map tailored to the most common tasks,
- selection of basic and supplementary tools,
- defining applications for individual solutions,
- preparing your own set of prompts and good practices,
- plan for further development of AI competencies.