Is AI really Intelligent?

The Movie, "Wizard of OZ" ends with Dorothy, the Lion, Scarecrow, Tin Man, and, of course, the dog Todo discovering that the Wizard was not a Wizard but a man behind a machine, that was able to convince the whole of Munchkinland that he could perform magic and was the most powerful wizard.                              

 AI, unlike the wizard, can actually perform magic (pun intended). But similar to the story, AI is a technology behind the scene that processes our request and delivers an intelligent output. So, back to the question: Is AI really intelligent?                                                                  

The Chinese Room Experiment     

Imagine a person who doesn’t understand Chinese is locked in a room with a set of rules and Chinese characters. This person receives Chinese characters through a slot in the door and uses the rules to produce appropriate responses, which are then passed back out through another slot. To an outside observer, it appears as though the person in the room understands Chinese, but in reality, they are just following instructions passed through the slot.                                    

That’s how Generative AI works. The computer is trained with enough information to respond to your questions (prompts) intelligently. But the device itself is not intelligent, as it is incapable of being curious, yet it can act and respond intelligently.                                                                  

Not Artificial but Real Intelligence     

Now, another interesting twist about the name AI is that the computer is not trained with artificially generated information but genuine information from real people: Educators, Healthcare Professionals, Engineers, Economists, Financial Analysts, and Creatives. Every casual word shared on the internet or a highly rated article written by a college Professor forms the foundational knowledge base for AI. 

How AI Works     

Artificial intelligence (AI) combines large amounts of data with sophisticated algorithms that enable machines to learn, reason, and interact with humans. Here is a simplified list of the AI Process   

Step 1: Data Collection                                   

Gathering relevant data from various sources.                                   

Step 2: Data Preprocessing                                   

Cleaning, transforming, and preparing data for use.                                    

Step 3: Model Training                                    

Using Machine Learning /Deep Learning algorithms to train a model on the prepared data.                     

Step 4: Model Deployment                                    

Integrating the trained model into a larger system or application.                                    

Step 5: Continuous Learning                                    

Updating the model with new data to improve its performance and adapt to changing environments. 

I’ll end this article with a quote from an AI researcher from Canada, Sasha Luccioni, who said in a talk,     

 ‘AI is not a done deal. We’re building the road as we walk it, and we can collectively decide what direction we want to go in, together.’                           

References:     

Elements of AI - https://www.elementsofai.com/                                    

Wizard of Oz - https://www.warnerbros.com/movies/wizard-oz                                    

Meta AI - https://www.meta.ai/ (Writing Support)                                    

NapkinAI - https://www.napkin.ai/ (Image Creation) 

If you enjoyed reading this article, please like, share or repost,