Get AI Ready (2nd edition)
A lot has changed in AI since we launched Get AI Ready in 2025. This new edition brings the course up to date with recent AI developments and how people use it at work.
New to Get AI Ready? We always recommend taking the latest edition.
Already completed the previous edition? This page highlights the main updates. If retaking the full course isn't an option right now, you can always skip the familiar parts and just focus on the new concepts and activities.
Learn it at a glance
What do people mean when they say “AI”?
Artificial intelligence is a broad field of technology, much like “chemistry” or “medicine” describes a broad field rather than one specific thing.
In everyday work, however, when people talk about “using AI,” they often mean tools such as AI assistants and chatbots that can help them write, analyze, create, research, and solve problems.
So AI is the broad field; AI assistants are one way we interact with it.
What is multimodal AI?
Early generative AI tools were often built around one type of information, such as text.
Today, many AI systems are multimodal. This means they can work with several types of information — such as text, images, audio, and video — and increasingly move between them.
For example, you might give an AI an image and ask questions about it, speak to it instead of typing, or ask it to create an image from written instructions.
Where do LLMs fit in?
A large language model (LLM) is an AI model designed to understand and generate language.
LLMs still power much of the interaction behind today's AI assistants. But modern AI systems increasingly combine language models with other capabilities so they can work with images, audio, video, tools, and other types of information.
LLMs remain an important engine behind AI assistants, but the experience around them has become much broader.
AI models are improving quickly
AI capabilities have advanced rapidly in recent years. Newer systems can work with more kinds of information, handle much larger amounts of context and complete increasingly complex tasks.
That also means individual features and tools can change quickly.
Rather than trying to memorize what every AI tool can do today, it is more useful to understand the principles for working with AI effectively — principles that will likely remain useful as the technology develops.
What makes a good prompt?
You don't need a perfect formula or special vocabulary to prompt AI effectively.
A useful prompt usually makes three things clear:
Task — What do you want AI to do?
Context — What does AI need to know to do it well?
Requirements — What should the result include, avoid or look like?
For simple tasks, your prompt might only be a sentence. For more complex work, giving AI relevant context and clear requirements can make a big difference.
What is a context window?
The context window is the amount of information an AI model can consider at one time.
That context can include your prompt, earlier messages in the conversation, documents or other information you provide, and information the AI system adds itself.
Modern AI systems can work with much more context than earlier ones. This makes it possible to give AI larger documents, longer conversations, and more background information.
But more context isn't automatically better. The most useful context is information that is relevant to the task.
What does transparent AI labeling mean in the EU?
As AI-generated content becomes harder to distinguish from human-created content, transparency about its use is becoming more important.
Under the EU AI Act, transparency requirements apply to certain uses of AI-generated or manipulated content. For example, some deepfakes and certain AI-generated material intended to inform the public need to be disclosed as artificially generated or manipulated. AI system providers also have requirements around making generated content detectable as AI-generated.
For organizations, the practical takeaway is simple: think about when people should know that AI has been used to create or significantly alter something.
Your organization's AI policy can set clear expectations for when and how employees should disclose AI use.