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

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

Discerning high-quality output: A prerequisite for navigating an AI-generative world

In my previous article, I emphasised the importance of creating effective inputs. However, that’s merely one side of the coin. The other equally vital aspect is the ability to discern the outputs generated by AI.

Why discerning high-quality output is crucial

The most advanced AI systems can produce outputs that might not be entirely reliable or suitable, even when they're well-managed. The nature of current large language models (LLMs) is such that each token/word’s generation is grounded in probability, which can sometimes lead to ambiguous or unintended results.

Blindly accepting any AI-generated information can lead to misinformation or worse, reinforce existing biases. Consider the instance highlighted by BBC News where a lawyer from New York faced a court hearing due to his over-reliance on AI, specifically ChatGPT, for legal research. The lawyer's filing referenced non-existent legal cases, with its author claiming to be unaware of the AI tool's potential to generate false information. This case further highlights the importance of discerning high-quality output from low-quality ones when relying on AI-generated information. Given that our children are growing up in a world increasingly mediated by these technologies, equipping them with the skills to critically evaluate AI outputs is more crucial than ever.

Criteria for assessing output

Teaching our children to evaluate these outputs will strengthen their interactions with technology and provide them with the necessary tools to sift through vast amounts of information. Here are three areas you can guide your children on:

• Relevance: An output's value is first determined by its relevance to the initial query. It's essential to ensure the response directly addresses the original question. For instance, if your child asks an AI tool about a frog’s life cycle, receiving an answer about other types of amphibians isn't directly relevant. Another example can be seen in a Maths paper, where if your child requires understanding the concept of ratios, a detailed explanation about fractions, while related, doesn't serve their immediate need. Teaching children to recognise this information mismatch ensures they can refocus their inquiry or seek a more relevant source to prevent wasted time. I have also included some good and bad examples at the end of the article so that you can clearly differentiate high-quality outputs from low-quality ones.

• Accuracy: Not all information provided by AI tools is necessarily accurate. Parents can prepare their children by emphasising the importance of cross-referencing multiple reputable sources, avoiding exclusive reliance on a single AI tool and even discussing findings with teachers, parents, or peers - all of which can help validate their understanding and offer different perspectives. This approach modernises the age-old wisdom of not believing everything you read, adapting it for a world where necessary scrutiny extends to AI-generated information as well.

  Geniebook has examination preparations, academic resources, and blog articles for parents and children to refer to.

• Completeness: Ensuring that the output one receives is not only correct but also thorough is important, especially in academic contexts. Take the topic of digestion from the Grade 3/4 Science syllabus as an example; if a child asks about the digestion process, a well-rounded answer wouldn't simply explain the breaking down of food in the stomach. It should minimally encompass the following points:

  • The roles of different organs, as digestion starts from the mouth, then onto the oesophagus, stomach, and the small (absorption of nutrients and continuation of food breakdown) and large intestines.
  • The end-to-end journey of food, from ingestion to excretion, highlighting the various stages of digestion and absorption.

• Just as a puzzle requires every piece to complete the picture, understanding a topic like digestion requires a comprehensive overview of each stage and component. Parents should motivate their children to spot gaps in information and question inquisitively, ensuring they grasp the subject fully, rather than settling for piecemeal knowledge.

Conclusion

Equipping our children with the ability to discern high-quality outputs in our AI-driven world is essential. The goal is to ensure that our children can skillfully navigate digital landscapes by rigorously assessing the direct relevance of a given response to the original query, validating its accuracy through cross-referencing and demanding comprehensive well-rounded answers. As we move forward, it will be increasingly important to delve into how these essential skills can be integrated more seamlessly for the dual objectives of efficiency and digital safety.

Appendix

Examples of good and bad outputs

Maths (Fractions):

Question: How do you add 2 fractions with different denominators?

Bad answer: Fractions represent parts of a whole. They have numerators and denominators.

Good answer: To add 2 fractions with different denominators, you first find a common denominator, convert both fractions to equivalent fractions using the common denominator, and then add the numerators. For example, to add 1/4 and 2/3, the common denominator is 12. Convert the fractions to 3/12 and 8/12. Now, add the numerators: 3 + 8 = 11. The answer is 11/12.

Maths (Area and perimeter):

Question: How do you calculate the area of a rectangle?

Bad answer: A rectangle has 4 sides, and opposite sides are equal in length.

Good answer: To calculate the area of a rectangle, you multiply its length by its width. If the length is 10 cm and the width is 5 cm, the area would be 10 cm x 5 cm = 50 cm².

Science (Life cycle of a frog):

Question: What are the stages of a frog's life cycle?

Bad answer: Frogs are amphibians and they live both in water and on land.

Good answer: A frog's life cycle consists of 3 main stages: egg, tadpole, and adult frog. It starts when the mother frog lays eggs in water, which hatch into tadpoles. The tadpoles then develop into adult frogs.

Science (Photosynthesis process):

Question: How do plants make their food?

Bad answer: Plants are green because they have chlorophyll in their cells.

Good answer: Plants make their food through a process called photosynthesis. They use sunlight, carbon dioxide from the air, and water to produce glucose (food) and oxygen. Chlorophyll in the plant cells captures sunlight, which is essential for this process.