Getting Started with Azure AI - Core Concepts

Note: Applies to certification exams AI-900 / AI-102)

Azure AI services provide a comprehensive suite of tools and services to help you build, deploy, and manage AI solutions. In this guide we are going to give an overview of Azure AI and some of the core concepts of Aritificial Intelligence. This will give you the foundation as you proceed upon your AI studies for the AI-900 certification (Microsoft Azure AI Fundamentals exam) or decide to take things further by learning to implement Azure AI solutions with the AI-102 certification (Designing and Implement a Microsoft Azure AI Solution).

1. What is Azure AI?

**Azure AI** is a set of AI services and tools that enable you to integrate AI capabilities into your business applications. These services include pre-built AI models, customizable AI models, and tools for building and deploying your own AI models.
Before we get too deep into the Azure AI services, and I get it, the buzz is everywhere, but let’s clarify what we mean when we say Artificial Intelligence…

2. What is Artificial Intelligence?

Artificial Intelligence (AI) refers to the development of computer systems capable of performing tasks that typically require human intelligence. These tasks encompass a range of capabilities, including:

Visual Perception: The ability to process and interpret visual information from the world, enabling applications such as image and video analysis.
Text Analysis and Conversation: Utilizing natural language processing (NLP) to comprehend, generate, and respond to human language, facilitating interactions like chatbots and language translation.
Speech Recognition and Synthesis: Converting spoken language into text and vice versa, allowing for voice-activated systems and conversational agents.
Decision Making: Leveraging past experiences and learned patterns to make informed decisions, such as identifying anomalies in data and initiating appropriate actions.
Robotics: AI is increasingly utilized to develop sophisticated robots that can perform tasks autonomously or semi-autonomously. This expansive field includes a wide range of applications, encompassing everything from manufacturing robots designed for assembly lines to drones used for various purposes, as well as advanced robotic surgery systems that assist medical professionals in complex procedures.
Generative AI: This means making new things, like images, music, or text, from existing information. Examples are AI-created art and deepfake technology. This is a fast-growing area of AI that you likely see often in everyday products like ChatGPT, Grok, Claude, and Gemini.

As you no doubt have seen, AI capabilities are increasingly integrated into everyday applications, enhancing user experiences and automating complex tasks. Taking this a step further, it is important to understand some key AI-Related terms for the AI-900 and AI-102 exams.

AI-Related Terms: Understanding Key Concepts

Artificial Intelligence (AI): The simulation of human intelligence processes by machines, especially computer systems. These processes include learning (the acquisition of information and rules for using the information), reasoning (using rules to reach approximate or definite conclusions), and self-correction.
Machine Learning (ML): A subset of AI that involves the use of algorithms and statistical models to enable computers to improve their performance on a task through experience. ML algorithms build a model based on sample data, known as training data, to make predictions or decisions without being explicitly programmed to perform the task.
Deep Learning: A subset of machine learning that uses neural networks with many layers (deep neural networks) to analyze various factors of data. Deep learning models can achieve state-of-the-art accuracy in tasks such as image recognition, speech recognition, and natural language processing.
Natural Language Processing (NLP): A field of AI that focuses on the interaction between computers and humans through natural language. The ultimate goal of NLP is to enable computers to understand, interpret, and generate human language in a way that is both meaningful and useful.
Computer Vision: A field of AI that enables computers to interpret and make decisions based on visual data from the world. This includes tasks such as image recognition, object detection, and video analysis.
Neural Networks: A series of algorithms that attempt to recognize underlying relationships in a set of data through a process that mimics the way the human brain operates. Neural networks are the foundation of deep learning algorithms.
Reinforcement Learning: A type of machine learning where an agent learns to make decisions by taking actions in an environment to maximize some notion of cumulative reward. It is inspired by behavioral psychology and is used in various applications, including robotics, game playing, and autonomous vehicles.

Considerations for AI Engineers

When developing AI solutions, engineers must take into account several important considerations to ensure the effectiveness, fairness, and ethical use of AI technologies. Here are some key considerations for AI engineers:

Data Quality and Quantity: The success of AI models heavily depends on the quality and quantity of the data used for training. Engineers must ensure that the data is accurate, relevant, and representative of the problem domain. Additionally, sufficient data is required to train robust models that generalize well to new, unseen data. [Source]
Bias and Fairness: AI models can inadvertently learn and perpetuate biases present in the training data. Engineers must actively work to identify and mitigate biases to ensure that AI systems are fair and do not discriminate against any group. Techniques such as bias detection, fairness-aware algorithms, and diverse training datasets can help address these issues.
Ethical and Legal Considerations: AI engineers must be aware of the ethical and legal implications of their work. This includes ensuring compliance with data protection regulations, respecting user privacy, and considering the societal impact of AI technologies. Engineers should adhere to ethical guidelines and best practices to build trustworthy AI systems.
Model Interpretability and Transparency: It is crucial for AI models to be interpretable and transparent, especially in high-stakes applications such as healthcare and finance. Engineers should strive to develop models that provide clear explanations for their predictions and decisions. This helps build trust with users and allows for better debugging and improvement of the models.
Scalability and Performance: AI solutions must be designed to scale efficiently and perform well under various conditions. Engineers should consider the computational requirements, latency, and resource constraints of their models. Optimizing algorithms, leveraging cloud infrastructure, and using efficient data processing techniques can help achieve scalable and high-performance AI systems.

To apply this further, it is important to understand the term inferencing. This essentially means taking the model you have created in order to make predictions. In essence, you are taking the trained machine learning model to make predictions and possibly decisions based on brand new data. The model that you spent all the time training and acquiring knowledge in the training phase is now being applied to real-world scenarios. This is why it is vitally important to take into consideration the ethical and legal considerations above.

***AI can do a lot of good, but it also has the power to do a lot of harm.***

As such, it is imperative to have a guiding set of principles you live by as you rollout your AI initiatives. Here are some core industry wide principles you can adopt to ensure you are practicing AI responsibly.

Considerations for Responsible AI

Fairness and Bias Mitigation: AI systems should be designed to treat all individuals fairly and avoid perpetuating biases. This involves using diverse and representative datasets, implementing fairness-aware algorithms, and continuously monitoring and addressing any biases that may arise.
Transparency and Explainability: AI models should be transparent and provide clear explanations for their decisions. This helps build trust with users and allows for better understanding and debugging of the models. Techniques such as interpretable machine learning and model-agnostic explainability methods can be employed to achieve this
Privacy and Security: Protecting user privacy and ensuring the security of AI systems is paramount. This includes implementing robust data protection measures, adhering to privacy regulations, and securing AI models against potential attacks or misuse.
Accountability and Governance: Establishing clear accountability and governance frameworks is essential for responsible AI. This involves defining roles and responsibilities, setting up oversight mechanisms, and ensuring that AI systems are used ethically and in compliance with legal and regulatory requirements.
Human-Centric Design: AI solutions should be designed with a focus on human needs and values. This includes involving diverse stakeholders in the design process, considering the societal impact of AI technologies, and ensuring that AI systems enhance human capabilities rather than replace them.

On to Azure AI Services…

Now we have a good baseline on Artificial Intelligence, let’s move back to the part I’m sure you are excited to try…Azure AI Services!

Exploring Azure AI: Key Concepts for AI-900 / AI-102 Exams