AI is an umbrella term that brings together multiple branches, including machine learning, deep learning, and generative AI.
![ai]()
Artificial intelligence (AI) is one of the most exciting and impactful technologies of our time. At its core, AI is about creating systems that can perform tasks that normally require human intelligence, things like problem solving, decision making, recognizing faces, understanding speech, or even creating new content.
Machine Learning (ML)
Machine learning allows computers to learn from data and improve their performance without being programmed.
![ML]()
Instead of following a strict set of rules, ML models recognize patterns and make predictions or decisions based on the information they are trained on.
Example: Netflix recommending your next show based on your watch history.
Remember: machine learning models are only as good as the data you feed them.
If you train a model on movie reviews but only from people who love horror films, don’t be shocked when it calls Finding Nemo ‘not scary enough.’
We'll cover ML in great detail in next article..
Deep Learning (DL)
We already understand how machines work and how the human brain works, then some brilliant minds thought, what if we combine the two? And that’s how Deep Learning was born.
![brain-inspired-ai]()
Deep learning is a subfield of ML inspired by how the human brain works. It uses artificial neural networks to recognize patterns in complex data like images, sound, or video.
In fact, we're using DL every day to unlock your phone; yes, face recognition is an application of DL.
Instead of learning simple patterns, deep learning models have many layers (hence deep) that automatically extract features from raw data. This makes them extremely powerful for images, audio, natural language, and complex tasks where traditional ML struggles.
Azure has Cognitive Services– Computer Vision: extracts information from images and videos, such as identifying objects or reading text.
AWS has Amazon Rekognition - Which analyzes images and videos for objects, people, or activities.
Google has Cloud Vision API - which extract insights from images, documents, and videos
ML vs. DL
Machine Learning: Works well with smaller datasets, requires feature engineering, simpler algorithms like decision trees, SVM, logistic regression.
Deep Learning: Needs huge amounts of data and computing power, uses neural networks with many layers, and learns features automatically.
Generative AI
Fast forward to today… We stand at the dawn of a new era, the age of Generative AI.
Generative AI takes deep learning a step further; instead of just analyzing data, it can create new data (like text, images, or audio) by learning patterns from existing examples.
![genai]()
What we've seen so far,
LLMs: Generate text, chat, summarize, write code. Your GPTs and Claude.
Image AI: Create art, designs, and edit photos. DALL·E, MidJourney, Stable Diffusion.
Video AI: Turn text into movies or animations. Runway Gen-2, Pika Labs
Audio AI: Make music, clone voices, or narrate. AIVA, Suno, Jukebox
3D AI: Build models for games, design, and AR/VR. NVIDIA GET3D, OpenAI’s Point·E.
Azure Example: Azure OpenAI Service, gives access to OpenAI’s models (like GPT and DALL·E) directly through Azure.
AWS Example: Amazon Bedrock, allows developers to build and scale generative AI applications without managing infrastructure.
Google Cloud Example: Vertex AI, innovate faster with enterprise-ready AI, enhanced by Gemini models.
Final Notes
Artificial intelligence is all around us, shows we stream to the way our photos are tagged automatically. In this article, we explored AI basics, machine learning, deep learning, and generative AI, along with real examples from Azure, AWS, and Google Cloud.
In the next article, we’ll dive into Artificial General Intelligence (AGI) and foundation models, the big-picture concepts shaping the future of AI.