How Our Beliefs Shape AI’s Future?
Have you ever wondered why some people are so excited about artificial intelligence (AI), while others are worried? A lot of it comes down to how we think about intelligence itself. Let’s dive into this big question:
How do our ideas about what intelligence is change the way we create and use AI?
Photo by Google DeepMind on Unsplash
What is intelligence, anyway?
This might seem like an easy question, but it’s actually pretty tricky! There are two main ways people tend to think about it:
- The Computer Brain: Some scientists say our brains are like really fancy computers. They think all our thoughts, feelings, and even our sense of self just come from the way our brain cells talk to each other. This view is called the “computational theory of mind.”
- The Soul and the Mind: In Islamic thought (and some other traditions too), intelligence isn’t just about brain cells. Muslim scholars like Al-Ghazali and Ibn Sina talked about the soul (nafs) as something separate from the body. They believed true intelligence comes from this spiritual part of us.
The Quran even mentions a divine spirit (ruh) that God breathes into humans:
“And when I have proportioned him and breathed into him of My [created] soul,1 then fall down to him in prostration.” [15:29]
But it also says the soul is a big mystery:
“And they ask you, [O Muḥammad], about the soul. Say, “The soul is of the affair [i.e., concern] of my Lord. And you [i.e., mankind] have not been given of knowledge except a little.” [17:85]
Why does this matter for AI?
If intelligence is just fancy math in our brains, then maybe we can copy it in computers. But if it needs a soul… well, that’s a whole different story! Can we really make a “thinking machine” if thinking requires something spiritual?
Let’s look at some cool AI stuff happening right now:
- ChatGPT: This AI can write like a human and answer all sorts of questions.
- Stable Diffusion: This one makes amazing artwork from just a few words.
- AlphaFold: This AI figured out how proteins fold, which is super important for medicine.
- A cool documentary on how an AI (company) beat arguably the best (human) Go player at the time.
These AIs do incredible things, but do they really understand like we do? Or are they just really good at faking it?
There’s a famous thought experiment called the “Chinese Room” that makes us think about this. Imagine a person who doesn’t speak Chinese in a room full of Chinese books. If they follow really detailed instructions, they might be able to answer questions in Chinese without actually understanding the language. Some people think AI is like this — it can give the right answers without truly understanding.
What about knowledge?
Intelligence isn’t just about being smart — it’s about knowing things. But what does it really mean to know something?
A Muslim thinker, Syed Muhammad Naquib al-Attas, had an interesting idea. He said knowledge is when the meaning of something “arrives in the soul” or when the soul “arrives at the meaning.” It’s not just about having information, but really understanding how things fit together in the big picture of creation.
This makes us ask: Can a computer program ever really know things the way humans do? Or will AI always be limited to just processing information without true understanding?
But then, are we using the right words?
Maybe we need to rethink how we talk about AI. If real intelligence needs a soul, can we call what computers do “artificial intelligence”? Should we use different words like “advanced computation” or “machine reasoning” instead?
The way we talk about things can change how we think about them. If we stop calling it “artificial intelligence,” it might help us remember the limits of what computers can do.
Making AI that fits with our values
Some scientists, like Stuart Russell, talk about making “human-compatible AI.” He says AI should:
- Want to help humans, not hurt us
- Know when it’s not sure about things
- Try to figure out what humans want
These are good ideas, but from an Islamic view, we might want to add more. What about “tawhid-compatible AI”? This means making AI that fits with the idea that everything in creation is connected and points to God’s oneness. It’s not about making religious AI, but about thinking bigger about how AI fits into the world.
What does this mean for Muslims who work with AI?
If you’re a Muslim who’s excited about AI, you might feel torn. On one hand, AI can do amazing things that could help a lot of people. On the other hand, the Islamic view of intelligence is very different from how most AI researchers think about it.
Here’s an idea: What if instead of trying to make AI that copies human intelligence, we focus on making AI that helps human intelligence? ==We could create tools that make us better at thinking and understanding==, while remembering that there’s something special about human consciousness that computers might never have.
Big questions to think about:
- If intelligence needs a soul, can we ever make truly intelligent machines?
- How can we use AI in ways that respect Islamic ideas about knowledge and understanding?
- Should we change the way we talk about AI to better reflect what it can and can’t do?
- How can we make sure AI development considers different cultural and religious views on intelligence?
These are huge questions without easy answers. But by thinking deeply about them, we can hopefully create AI that’s not just powerful and useful, but also wise and good for everyone.