DIGITAL LEARNING
Prompt engineering
Prompt engineering is the practice of forming GenAI prompts in a manner that is more likely to yield a desired outcome .
From UNESCO studies , it is found that prompts are most effective when focused on a specific problem with relevant context structured coherently and rationally . UNESCO recommends the SECRE framework for prompting , which are five guiding principles mentioned below .
• Use simple , clear and straightforward language that can be easily understood , avoiding complex or ambiguous wording .
• Include examples to illustrate the desired response or format of generated completions .
• Include context , which is crucial for generating relevant and meaningful completions .
• Refine and iterate as necessary , experimenting with different variations .
• Be ethical , avoiding prompts that may generate inappropriate , biased or harmful content .
For example :
“ Using the IGCSE CIE history marking scheme , mark the following essay . Provide feedback based on each of the criteria and suggest improvements for each paragraph .
Essay is as follows : ( copied essay )” And a follow up prompt :
“ Based on the essay above on why the impact of the first world war was a key cause for the Roaring Twenties , provide me with numerical statistics and facts that would be relevant to my three main points : deferred spending , growth in consumer goods , wartime debts . Priorities economic statistics and present them in context by providing a reference value . E . g . for GDP growth of a country , provide global average GDP growth for comparison ”
AI usage framework
When utilizing AI , it is important to have a framework to get outcomes without compromising learning outcomes . I use a straightforward framework derived from the UNESCO co-design framework that is suitable for student usage .
The purpose of the framework is to educate students on the limitations of AI and have them build around that so they can benefit from AI . Risks can range from AI errors , such as AI providing the wrong information , to humanbased-errors , such as having AI fully write an essay and losing out from cognitive growth . However , the benefits can be sizable as AI can provide customised and focused support when leveraged correctly .
Purpose |
Expected outcome |
Risk |
How can the user |
|
|
|
reduce risk |
1:1 Spanish language skills coach |
Learning conjugations and having conversation practice . |
AI may provide wrong conjugations due to “ AI hallucinations ”, causing the student to learn incorrect conjugations
AI may utilize contextually inaccurate or incentive language .
|
Have a foundational level in the language to identify basic issues .
Have reliable sources to cross-reference when AI utilizes provides inaccurate information .
When obtaining new language phrases from AI , check with teachers .
|
Conclusion :
AI is highly impactful to students and educators , and it continues to advance at a rapid rate . The three points I have shared in this short article relate to mainstream points on AI in education . As AI is continually advancing there is more day-by-day to learn . . It is best for students to be educated about AI , so they can utilise AI to its fullest extent without impairing their cognition .
13