From January 20 to 24, Codelearn centers will host the first thematic week of the school year, which this year will focus on generative artificial intelligence tools. Throughout the week, we will talk about the origins of artificial intelligence, its evolution, and its applications today, discover some free applications, and use them to generate different types of content. To ensure we have understood the key concepts, we will also do a Kahoot at the centers on this topic.
What is Artificial Intelligence?
Artificial Intelligence (AI) is a branch of computer science dedicated to developing systems and algorithms capable of performing tasks that require intelligence. This technology aims to replicate the human process of thinking, learning, and decision-making through computational models capable of solving problems autonomously.
The applications of AI are varied and extend to many fields, such as pattern recognition, decision-making, solving complex problems, process automation, computer vision, and natural language processing, among others. Today, AI is used in many sectors to optimize operational efficiency and enhance problem-solving capabilities in various situations.
AI has its origins in the 1950s, when some scientists began exploring how machines could mimic human intelligence. Since then, AI has evolved and achieved significant advances that have led to various beneficial applications in areas such as healthcare, industry, and technology.
What is Generative AI?
Generative AI refers to a type of artificial intelligence designed to create or generate new content (texts, images, music, videos, or other data). To train a generative AI, a large volume of data (which can be of any type) is used so that it can learn the hidden patterns in that data. This is how AI can generate new, coherent content based on what it has learned from the data it was trained on.
To create and train this type of AI, the first step is usually data collection, gathering a lot of data of the type the AI will handle. Then, the data is preprocessed, meaning the data is prepared and processed so the AI can understand and handle it. Next, the model is created by defining the AI system’s architecture (e.g., GPT). After that, pretraining begins, and the model is trained with a large volume of data. The next step is fine-tuning, as in some cases, the model needs to be adjusted to better adapt to specific data or tasks. At this stage, the model generates content, and its ability to produce coherent content based on what it has learned is tested. Next, the overall results are evaluated, and if necessary, the model is retrained or parameters are adjusted. Finally, when the model is considered robust and reliable, it is deployed and made available for public use.
When we talk about generative AI tools, we often think of ChatGPT, an advanced chatbot capable of natural conversation. However, there are an increasing number of applications available designed to generate content with the help of artificial intelligence: Canva or Krita are examples of graphic design and digital illustration tools; CapCut or Descript have video and audio editing features that include automatic subtitles and other AI effects; Copy-ai or Grammarly help generate creative texts and ensure grammatical and stylistic correctness; Speechnotes can quickly transcribe notes, and DeepL is an advanced and precise translator; Boomy creates AI-generated music in minutes for personal or commercial projects, and the list of both free and paid applications keeps growing, meaning this week at our centers, we will only get to know a small part of them, but we encourage you to explore the options available in the area you like the most.