OpenAI ChatGPT introductory training
The OpenAI ChatGPT introductory training is designed for companies that want to give their people a shared understanding of what ChatGPT can do and when to reach for each of its features.
We work through ChatGPT’s main techniques and features in one coherent flow, so that participants leave with a clear sense of what each one is for: choosing a model and a thinking level, files and Library, web search, Deep Research, Memory, instructions, Projects, voice and images, Skills, data analysis, Work, scheduled tasks, and Plugins and Apps.
What you'll be able to do
- understand the principles, limitations and main risks of generative AI
- know ChatGPT’s main features and which one to use when
- tell apart the situations that call for an ordinary chat, web search, Deep Research or Work mode
- know how to use files, Library, Projects, Memory and instructions to build the context for a piece of work
- try out one clear prompting technique and build a personal Skill for a recurring task
- understand how scheduled tasks, Plugins and Apps extend ChatGPT — automating recurring work and connecting it to other tools and data sources
Who this course is not for
For an advanced user the material will be largely familiar and high-level. This is an informative overview of the features rather than in-depth skills training — deepening specific working skills and redesigning company processes belong to a later stage, in smaller groups.
Course structure
The fundamentals of generative AI
- What generative AI is, and how a language model works in principle
- Where AI genuinely helps: finding information, producing content, analysis, and making recurring tasks simpler
- Limitations and risks: hallucinations, outdated information, excessive agreement, and the user’s own responsibility
OpenAI ChatGPT
- What sets ChatGPT apart from other AI tools, and how the ChatGPT environment is laid out
- How to use ChatGPT safely, and what to keep in mind in an organisation’s workspace
- ChatGPT’s main features, and how to choose the right approach for the task at hand
Prompting
- What makes a prompt clear and effective
- A group exercise: comparing a poor prompt with a good one
- How to give ChatGPT enough context, the goal, the constraints, and the output you want
- When a longer, detailed brief is worth writing, and when a plain question will do
Models and the level of thinking
- How to pick the balance of speed and thoroughness that fits the complexity of the task
- When a fast answer is enough, and when more thinking is worth it
- How the depth of thinking affects the speed and quality of the answer, and how much of your allowance it uses
Files, Library, web search and Deep Research
- Uploading and analysing documents, spreadsheets, presentations, images and other files
- Library, for finding and reusing files
- Web search, for finding and verifying up-to-date public information
- Deep Research, for more complex multi-source, multi-step investigation
- When to use an ordinary chat, when web search, and when Deep Research
Memory, instructions and Projects
- Memory and earlier context: what ChatGPT can remember, and how to manage it
- Account-wide instructions and personal preferences
- Project-specific instructions and files
- Projects, for keeping the context of a recurring topic, client or workflow together
Voice, images and multimodal work
- Voice mode, for thinking out loud, talking a question through, and working hands-free
- Analysing images and screenshots, and answering questions from visual information
- Creating and editing images with ChatGPT
We started to see where AI could actually be of help.
Even though AI has existed for over a decade, it's only now becoming a mainstream tool. We were aware of it, but our use was sporadic and, at times, skeptical. The greatest value of working with Eneli and her team was that we began to see where AI could truly help us and how to use this tool to achieve various goals. This journey is just beginning, but we can confidently say that our entire team is now on this accelerating train, rather than waiting on the platform. It's a good feeling because the next train won't be coming anytime soon.
JOEL VOLKOV
Creative Agency TANK,
Partner and Creative Director