2 Critical skills your child needs to prepare for an AI-generative world: Part 1

This article was kindly contributed by our CEO, Neo Zhizhong.

In our new AI-generative world, you can create or think up anything you want on a whim, with the likes of AI models such as ChatGPT, Jasper, Midjourney, and many more. Have an illustrative story you want to publish? Send in your prompts to ChatGPT and Midjourney, and you’re on your way to publishing your book!

There are two vital skills that will shape our children's success in this new normal. In this article, I’ll focus on the first: the ability to create effective inputs. For parents wondering how their child will thrive in this AI-driven future, understanding this skill is an essential starting point.

Creating effective inputs: The cornerstone of AI interactions

Generative AI, as impressive as it is, functions on a fundamental principle: it can generate effective outputs only when effective inputs are received.

Articulating clear objectives

Before any interaction with an AI system, clarity is paramount. Children need to:

• Understand their goals:

  What do they hope to achieve from the AI system or even a particular interaction? Is it information, entertainment, or problem-solving? This understanding forms the basis of how they will use the AI tool. It's equivalent to deciding your destination before you start a journey; knowing where you want to go is the first step.

• Define parameters:

  Building on their goals, children need to understand how to set the parameters for interaction with an AI tool. Parameters are specific instructions or rules that guide how the AI should respond, much like the rules of a game such as chess. Imagine giving your child a chessboard without explaining how the pieces move. They would likely feel frustrated and move the pieces aimlessly. The same holds true when working with AI models such as ChatGPT; without clear parameters, the responses can be inaccurate and random.
  
  Teaching children this skill helps them engage more effectively with technology and fosters disciplined thinking, a necessity that will remain critical and valuable for time to come, regardless of how the world changes.

• Emphasise context:

  In the field of AI, understanding the broader context of a request is crucial for achieving accurate and relevant results. It's as though we’re explaining a story to a child - without understanding the background or setting, the tale may lose its meaning. The custom instruction feature of ChatGPT serves this exact purpose. Once a custom instruction has been added, all prompts will include the context until you turn it off.

Understanding the medium

Inputs can come in various forms, such as text, visuals, or audio - these are the main media for now, and each has its intricacies.

• Textual inputs:

  Teaching a child to interact with AI is similar to guiding them in writing a simple school essay. Both tasks require clarity and precision. Imagine guiding your 12-year-old through a school project about Vietnamese traditions. Start with something specific, not just something generic like, “Vietnam has rich traditions.” Help them craft a detailed opening such as, “The Vietnamese Lunar New Year, known as Tết, is a vibrant celebration that marks the arrival of spring”. This same clarity is needed when one is interacting with AI systems. Instead of asking an AI tool a vague question like, “Tell me about Vietnam”, you would want to spur your child to ask something more targeted, such as, "What are the main customs and traditions celebrated during Tết in Vietnam?"

• Visual inputs:

  Teaching children about pictures and images is a key part of helping them understand how to create quality inputs for computers and AI. Parents can show kids tools such as Midjourney's '/blend' feature, which allows you to combine up to five images into a merged creation. While they may already be familiar with how handy social media image editors are, these editors also have their limitations. For example, AI-driven features like Generative Fill in Photoshop can create a realistic-looking face to expand on an existing image, but any result will likely be generic; without a frame of reference, the possibility of recreating your child's unique face from a back view photo is highly unlikely.
  
  As technology advances, when visual images can be interpreted more effectively into prompts, the use of visual data will become an integral part of generative AI. This integration further emphasises the importance of children understanding how to select and use images as an input wisely in order to produce the desired output.

• Audio inputs:

  Teaching kids to speak clearly is now easier with AI. Parents can use voice tools to help children practise their speech. These tools can understand different accents and even correct pronunciations. AI can also copy a child's unique voice with today’s advancements. Imagine recording your child saying something and an AI tool can make entire speeches in their exact tone and pitch without them actually narrating. This is undoubtedly an exciting way to learn, but parents must also teach their children to use this technology responsibly. It’s important to note that there is a possibility of identity theft through voice cloning, which can be used for illicit purposes. So, do keep in mind to use these tools with discretion.

Conclusion

Preparing our children to navigate an AI-generative world requires critical thinking, clarity, and responsible interaction with ever-advancing technologies. Now that we have delved into the significance of creating effective inputs, another essential skill awaits our exploration in a subsequent article. Keep your eyes peeled on how this second critical ability will shape our children's future in an AI-driven world!