ARCH 2360 — Artificial Intelligence in the Built Environment

Lecture Record: Recap, Bridging, and Orchestrator’s Deliberation

Session 3 & 4 Combined

1. Opening and Housekeeping

Ladies and gentlemen, peace be upon all.

Let us begin in the name of God, the Most Gracious, the Most Merciful.

Before we continue, please make sure your names are written clearly in the shared class list. We will use this list later as the basis for group work.

For those who missed last week’s class, I have uploaded the materials for Session 1 and Session 2 on the course website. The voice recording had too much background noise, so I am sharing the cleaned-up text and deliberation notes instead. Please review them so that you understand the foundation of this course.

Today, before we move into Session 3 and Session 4, I need to recap the essential framework from last week.

2. Why This Course Is Not Just About AI Tools

This course is not designed merely to teach you every AI tool available today.

There are hundreds of AI tools now, and new ones appear almost every few months. If we try to chase all of them, we will always be late.

Instead, this class is built around three layers:

Philosophy.
Academic understanding.
Practice.

The philosophy gives you grounding.
The academic part gives you structure.
The practice shows you how to apply AI responsibly.

Without philosophy, you may know the tools, but you may not know how to judge their outputs. Without academic understanding, you may use AI blindly. Without practice, everything remains only theory.

So the purpose of this course is not only to use AI.

The purpose is to learn how to see AI, how to engage with AI, and how to remain responsible while using it.

3. Session 1 Recap: Niat, Laku, Hasil

In Session 1, we discussed the idea of:

Niat → Laku → Hasil
Intent → Action → Outcome

This applies to architecture, life, and AI.

In architecture:

Client brief → design process → final building

In AI:

Input → process → output

If your intention is weak, the outcome will also be weak. AI does not fix bad thinking. It amplifies it.

That is why the human architect must remain at the centre.

AI may assist the process, but the architect carries the judgment.

4. Session 2 Recap: AI Systems, Models, and Knowledge

In Session 2, we discussed the current reality of AI.

At present, most AI systems are still under Narrow AI. They are task-based, pattern-based, and do not truly think independently.

ChatGPT, Gemini, Grok, DeepSeek, and similar conversational AI tools are still Narrow AI. However, they feel different because they can converse, generate ideas, respond across many domains, and appear almost human-like.

This is why people often confuse generative AI with AGI.

But AGI, or Artificial General Intelligence, is still conceptual. We are not there yet. Super AI is even further away and remains speculative.

So, do not be overwhelmed by science-fiction fears. AI still depends on physical infrastructure: data centres, servers, electricity, water cooling, and human-controlled systems. If the system is shut down, AI stops.

But even Narrow AI already has risks. It can disrupt industries. It can create dependency. It can create emotional confusion. Some people around the world have already developed unhealthy relationships with AI.

That is why this class must include ethics, philosophy, and judgment.

5. Knowledge Is Cheap, Wisdom Is Rare

Today, knowledge is everywhere.

Last time, we searched using Google, Yahoo, Baidu, or books. Now we ask ChatGPT, Gemini, Grok, Claude, DeepSeek, or Perplexity.

Knowledge has become cheap and abundant.

But wisdom is different.

Wisdom comes from judgment, experience, failure, reflection, responsibility, and humility.

AI can give you information.
AI can organise knowledge.
AI can generate options.

But AI cannot carry wisdom for you.

That is why the key message remains:

AI provides answers.
Architects provide judgment.

6. Decision-Making: Beyond 50/50

In life, if you remain at 50/50, you cannot make decisions.

At minimum, you need 51/49. Better, 60/40. As you mature, you may reach 80/20, where your judgment becomes clearer.

This applies to life, architecture, and AI.

AI can advise you. Your lecturers can advise you. Your friends can advise you. But at the end of the day, you are responsible for your own decision.

You cannot say:

“AI told me.”
“My lecturer told me.”
“My friend told me.”

You decide. You carry the consequence.

7. The 80/20 Rule in Learning and Becoming

I often share the 80/20 rule with students.

You may study 100% of the syllabus in university, but in real life, perhaps only 20% will be directly used.

Out of that 20%, only another 20% may truly help you perform well in your daily work. That becomes 4%.

Out of that 4%, only a smaller portion may take you toward mastery.

So the journey looks like this:

100% → Study
20% → Use
4% → Apply well
<1% → Mastery

This is not meant to discourage you.

It is meant to guide you.

If you can truly use even 20% of what you learned, that is already good. If you can reach the 4% level, you may begin to stand out. If you reach mastery, that is where you become different in the industry.

AI may help you move faster, but it cannot replace the journey of becoming.

8. Find Your Anchor

Because there are too many AI tools, you must not try to subscribe to everything or chase every new platform.

Instead, find one anchor.

Your anchor should be one conversational AI that helps you think, reason, reflect, and explore.

It can be ChatGPT.
It can be Gemini.
It can be Grok.
It can be something else.

My personal anchor is ChatGPT because it is broad, flexible, and useful across learning, writing, work, life reflection, and decision-making. Gemini is strong for research and workspace integration. Grok is useful for contrarian or disruptive thinking.

Choose what works for you.

If possible, subscribe to at least one paid conversational AI. If you can afford more, use two or three so you can compare their responses.

That is how deeper thinking begins.

9. Build Your AI Ecosystem

Once you have an anchor, you can expand.

Your anchor AI can help you understand other tools:

MidJourney for image generation.
DALL·E for visuals.
Sora or Runway for video.
Perplexity for research.
Coding tools for scripts.
Other architectural AI tools for design, BIM, performance, or visualisation.

Conversational AI becomes your gateway to the entire AI ecosystem.

But remember: do not simply accept whatever it gives you. Compare, test, reflect, and decide.

That is orchestration.

You are not merely using AI.

You are orchestrating AI.

10. Switching Modes

You must also understand that AI engagement has different modes.

There is Technical Mode: fast, direct, task-focused.

There is Work Mode: structured, professional, related to reports, presentations, workflow, and decision-making.

There is Personal Mode: reflective, exploratory, deeper, and more personal.

Maturity is knowing when to switch.

You may use AI technically for a task. You may use it professionally for work. You may use it reflectively to think. But you must know which mode you are in.

The danger is not AI.

The danger is not knowing when you have lost control.

11. Living With AI or Blending With AI

You have a choice.

You can live without AI.
Or you can blend your life with AI.

There is no right or wrong, but there are consequences.

If you refuse AI completely, others may work faster than you. In practice, someone using AI properly may complete in one hour what takes you two days manually.

But if you go too deep into AI and detach from reality, that is also dangerous.

So design your ratio.

Maybe 80% real world, 20% AI.
Maybe 60/40.
Maybe 50/50 temporarily.

But do not remain too long in uncertainty. And do not allow AI to replace your real life, family, friends, responsibilities, and grounding.

AI can sharpen your thinking, but it must not replace your living.

12. Humility in the Age of AI

No matter how advanced you become with AI, remain humble.

AI may help you produce better writing, stronger visuals, sharper analysis, and faster presentations. But do not become arrogant.

If you think nobody can stop you, God can stop you.

As architects, we are creators, but we are only small creators. The Ultimate Creator is God.

So use AI. Engage AI. Embrace AI if you choose to. But stay humble.

13. The Core Message

If you remember only one thing from this recap, remember this:

AI provides answers.
Architects provide judgment.

AI can generate.
AI can predict.
AI can simulate.
AI can suggest.

But humans decide.

The architect gives meaning, responsibility, and authorship.

Only human hands can raise the soul of architecture.

14. Transition to Session 3 and Session 4

After this recap, we will move into Session 3 and Session 4.

Session 3 focuses on Generative AI: how it creates text, images, ideas, and design possibilities.

Session 4 focuses on AI in Building Performance: how AI helps us evaluate energy, comfort, structure, function, simulation, and reality.

But before we begin, remember:

You are not here to learn every AI tool.

You are here to learn how to think, how to judge, and how to become an architect who can survive and lead in the AI era.

So now, we move from foundation into application.

Now, we begin.