Opening Statement (Engaging Start)

(Presenter 1 - Opening Statement)

Artificial Intelligence has rapidly changed the way we work, learn, and even create. From
generating text and images to writing complex code, AI is now becoming a common tool in
various fields—including programming. In recent years, AI-powered code generation tools like
GitHub Copilot and Google Gemini Code Assist have gained popularity, offering real-time
assistance to programmers. These tools provide code suggestions, automate repetitive tasks,
and even help with debugging, making coding more efficient.

But while AI can enhance programming, there is also growing concern about its impact on
learning. Are students truly improving their coding skills, or are they just relying on AI to do the
work for them? Does AI help in problem-solving, or does it weaken a student’s ability to think
critically? These are the questions that have sparked debates among educators, researchers,
and programmers.

With AI becoming more involved in education, especially in programming, we saw the need to
understand its real effects on students—not just in terms of benefits, but also the possible
challenges that come with it.

And that is why we conducted this study. Entitled

“Challenges and Benefits of AI code generation among senior high school programming
students at Zamboanga Del Sur National High”

To further explain our research, we have Chrizelle Belocura to begin with the Background of the
Study.

THEORETICAL FRAMEWORK

To better understand how AI affects learning, we used two learning theories that explain how
students process and build knowledge:

Constructivist Learning Theory

This theory was developed by Jean Piaget and emphasizes that learning is an active process
where students build their knowledge through experience, problem-solving, and interaction with
the environment
From the word "constructivist," this theory emphasizes that students construct their own
knowledge through experiences. Instead of just memorizing information, they actively engage in
problem-solving, experimentation, and critical thinking.

How does this relate to your study?

AI-powered code generation tools, such as GitHub Copilot and Google Gemini Code Assist, act
as learning scaffolds—they assist students in coding by providing real-time suggestions and
explanations.

According to constructivist theory, learning happens when students engage in active

problem-solving and experimentation. AI tools can help students explore different coding
approaches and learn through trial and error.

However, over-reliance on AI can be a problem. If students rely too much on AI-generated code
without understanding it, they might skip essential cognitive processes needed for
problem-solving.

This aligns with Piaget’s theory, which states that students should be active participants in
learning rather than passively receiving information.

Example in your study:

Your research examines whether students truly learn from AI-generated code or just copy it
without understanding the logic. If AI tools promote exploration and learning, they align with
constructivism. But if they reduce active engagement, they contradict it.

---

2. Sweller’s Cognitive Load Theory

Developed by John Sweller, this theory explains how the brain processes and retains
information. It identifies three types of cognitive load:

1. Intrinsic Load – The natural difficulty of a subject (e.g., programming concepts).

2. Extraneous Load – Unnecessary distractions that make learning harder (e.g., AI-generated
code that is too complex).

3. Germane Load – The effort put into understanding and processing information (e.g.,
debugging AI-generated code to improve learning).
How does this relate to your study?

AI-powered tools reduce intrinsic load by simplifying coding tasks, allowing students to focus on
understanding logic instead of memorizing syntax.

However, they can increase extraneous load if students struggle to interpret AI-generated code
or don’t know how to debug it.

The goal is to balance cognitive load so that AI helps learning rather than overwhelming
students with complex, unexplained solutions.

Example in your study:

Your research investigates whether AI tools make coding easier (reducing cognitive load) or
create confusion (increasing extraneous load). If students can’t understand or debug
AI-generated code, it means the tool is adding more cognitive strain instead of helping.

RESEARCH PARTICIPANTS

To ensure that our research will collect relevant and reliable data, we will carefully select 10
Grade 12 programming students from Zamboanga Del Sur National High School,
specifically from the Technical-Vocational-Livelihood (TVL) – Information and
Communication Technology (ICT) track.

Why will we only select 10 participants?

Since we are conducting qualitative research, our goal is to deeply understand experiences,
not to gather numerical data. We will focus on quality over quantity.

With fewer but carefully selected participants, we will be able to conduct in-depth
interviews and gain detailed insights, rather than just surface-level responses.

What is our criteria for selecting participants?

To ensure that our findings will be relevant and meaningful, we will only select students who
meet the following conditions:

1.​ Currently enrolled in the TVL-ICT Programming track​

 ○​ Since our study is about AI-powered code generation, we will focus on
programming students because they actively engage in coding tasks.
○​ Students from other ICT specializations (such as animation or hardware
servicing) will not be included because they do not focus on programming.
2.​ Have prior experience using AI-powered code generation tools​

○​ Participants must have used AI coding tools such as GitHub Copilot or

Google Gemini Code Assist in their programming tasks.
○​ Students who have never used these tools will not be included because they
would not be able to provide insights on AI’s impact.
3.​ Willingness to participate and share experiences​

○​ We will only include students who agree to participate voluntarily and are
willing to discuss their experiences in detail.
○​ This will ensure that our data comes from students who genuinely understand
and use AI in their learning process.

Research Participants

Why will we choose these students?

Not all students in the ICT strand have used AI-powered tools. And since our study focuses on
the challenges and benefits of AI-powered code generation, we need participants who can
provide real insights based on actual experience.

Because of this, we will use purposive sampling, a non-probability sampling method that
allows us to select participants who meet our specific criteria. By using purposive
sampling, we will be able to gather insights from students who have actual experience
with AI-powered code generation tools,

DATA GATHERING PROCEDURE

END

Thank you for your attention throughout our presentation. We've seen how artificial
intelligence (AI) can enhance education by offering personalized learning experiences
and automating routine tasks, allowing teachers to focus more on student engagement.
By thoughtfully integrating AI tools, we can enrich the educational experience while
maintaining the essential human elements of teaching and learning.
This concludes our presentation.