Artificial Intelligence (AI) is one of the most transformative technologies of the 21st century. Its applications range from self-driving cars to personal assistants, and its impact on industries like healthcare, finance, and entertainment is immeasurable.

It is also one of the most controversial and muddied topics on the Internet. Just ask one question on Twitter about the state of AI and you’ll get back 10 different answers, ranging from “AI will fix everything” to “Hide the children, the Roombas are coming!”.

Oh… you didn’t click on this article to hear me rant about the amount of AI proselytizing and AI doom on the Internet? (which I can control, by like, not looking at my feed)? Well then, I’ll skip right to the point: AI is an extremely important topic to be informed about.

Whether you’re the average person leveraging AI for school or work, a hobbyist writing the 1000th article on how AI multiplied their productivity / creativity ten-fold, or an aspiring programmer-researcher hoping to become the next Yann LeCun someday, I promise you that there’ll be at least be one course in this guide that’ll interest you. And here’s the best part: all of the courses are free or free-to-audit, and some of them even have free certification!

Some of the courses listed here will help you understand what AI is and its consequences for the human race. Others go a step further, offering hands-on experience in coding your own AI algorithms. Then, for those who are eager to dive into the complex and intellectually stimulating realm of mathematics and computer science, there are select courses that immerse you knee-deep in the intricate algorithms and theories that form the backbone of AI.

Thus in this Best Courses Guide, I’ve selected these 10 free courses to help you navigate the fascinating and sometimes downright absurd world of Artificial Intelligence in 2023. Click on one to skip to the course details:

Course Highlight	Workload
Best Math/Programming-Free Introduction for Non-Experts with Free Certificate (University of Helsinki)	16–24 hours
Best Practical Introduction for Programmers of All Levels (University of Helsinki)	50 hours
Best Technical Introduction for Python Programmers with Free Certificate (Harvard University)	70–210 hours
Best Business-Focused Introduction for Non-Experts (DeepLearning.AI)	12 hours
Best AI Math Course for Beginners (KAIST)	8 hours
Best Hands-On Introduction for Non-Experts (IBM)	9 hours
Best University-level Introduction with Common Lisp for Programmers (University of Oslo)	18 hours
Best Introduction to Generative AI for Non-Experts (Google)	10 days
Best Technical Introduction for Python Programmers (Microsoft)	12 weeks
Best University-level AI Math Course for the Confident (MIT)	24–25 hours

What is Artificial Intelligence?

Artificial Intelligence refers to the simulation of human intelligence in machines that are programmed to learn from experiences, adapt to new observations, and make decisions just like a human would. AI can understand drawn images, perceive its environment, solve complex problems, and even exhibit creativity in some instances.

Though of course, let’s not forget that AI doesn’t only apply to very intelligent models like ChatGPT or StableDiffusion. The term Artificial Intelligence is very broad, and can also refer to classifying things based on their characteristics (is this animal a cat or a dog?), or predicting the unknown based on known data (how much can I expect if I sell this house now in this neighborhood?). Thus, AI is an incredibly progress-driven field with a wide area of applications ranging from medical to marketing to virtual reality and much more.

What’s the difference between Artificial Intelligence, Machine Learning, and Deep Learning?

AI is an extremely broad field that includes several approaches and techniques, and two of the most essential subfields of AI are Machine Learning (ML) and Deep Learning (DL). To put it simply: ML is a subset of AI that involves the use of algorithms to parse data, learn from it, and make predictions or decisions, while DL is a further subset of ML and involves artificial neural networks with three or more layers to simulate a human brain. If you’re interested in exploring either of these topics, I’ve already linked my BCGs on them (yet to be updated), so feel free to redirect your webpage.

Why Learn Artificial Intelligence?

As businesses and governments invest in AI technologies, the demand for AI professionals is increasing exponentially. According to LinkedIn Talent Solutions, demand is high for tech professionals with artificial intelligence experience. And if you look at the World Economic Forum’s Future of Jobs Report, it is peppered with references to Artificial Intelligence, denoting its importance to the jobs of tomorrow. And of course, if you’re easily swayed by financial incentives (and who isn’t?), consider this: the average salary of a Machine Learning Engineer in the United States is a substantial USD 120K, according to Glassdoor.

And even if you aren’t planning to make AI a profession, there’s still a lot of reasons why you should try to understand it anyways, on a non-technical level of course. You’ve probably heard the debate and controversy over AI-generated content such as AI art copyright laws or ChatGPT plagiarism in schools. 2023 sure is a monumental event in the history of AI, never has AI been so accessible to the average person. Which is probably why even if you are that said average person, you should also try to inform yourself more on AI to the extent that you are able to articulate your opinions on AI without stumbling over your own feet.

Course Ranking Statistics

Here are some aggregate stats about the ranking:

• The most popular course in this ranking has around 1M enrollments.
• All of the courses in this BCG are free, free-to-audit, or have free certification.
• Three of the courses here require no knowledge of programming, while the rest require programming experience.
• About 320K people are following Artificial Intelligence Courses on Class Central.

Without further ado, let’s go through the top picks.

Best Math/Programming-Free Introduction for Non-Experts with Free Certificate (University of Helsinki)

“What is artificial intelligence? How does it affect your work and life? How will artificial intelligence develop?” Do you want these questions answered without the garbling of AI evangelists / doomers and without mathematical mumbo jumbo?

Helsinki’s free Introduction to AI course aims to make AI more understandable for the AI uninitiated. Thus, virtually anyone can start their AI-learning journey here. That’s not to say that what you’ll learn in this course will be trivial, though. You’ll not only be able to define what AI is and how they can think through and solve problems, but also have a firm mathematical intuition behind machine learning and neural networks, as well as be informed enough to participate in conversations around the philosophy and ethics of artificial intelligence.

And how will the course accomplish such a feat? Through illustrative examples, beautiful diagrams, interactive engaging exercises, MCQ questions, and accessible but informative prose, of course! I mean look at how beautiful the course chapters page looks.

Even if you do know some programming and aren’t afraid of math, I’d still recommend that you take this course regardless. Why? Because this course has a free certificate of completion, and also because this introduction leads up to the next more practical and hands-on course, Building AI (which you can read about in the next listing).

Institution	University of Helsinki
Level	Beginner
Workload	16–24 hours
Exercises	Yes
Enrollments	1M
Cost	Free
Exercises	Available
Certificate	Free

Best Practical Introduction for Programmers of All Levels (University of Helsinki)

The second and last part of the Elements of AI series, Building AI extends on what you’ve learned by teaching you the algorithms involved in creating AI methods from scratch. That means no TensorFlow, no Keras, no PyTorch. Just pure Python!

What makes this course so unique from the others, though, is that it has three difficulties: beginner if you have no coding experience, intermediate if you know Python, and advanced if you’re a Pythonista 🐍.

All levels will view the same course content, the only differences are the exercises. For example, Exercise 1: Listing pineapple routes is a multiple-choice question for beginners, but for intermediate and advanced learners it is a coding exercise.

And from that example it should be clear that a beginner won’t finish the course with the same technical knowledge as an intermediate or advanced learner. Thus, I highly recommend that you learn some Python before enrolling.

Assuming that you’ve followed my advice, after completing the course, you’ll be able to:

• describe different types of AI such as optimization, reasoning, and learning
• choose a suitable AI approach to solve simple tasks such as route planning, probabilistic inference, and pattern recognition
• implement a straightforward brute-force optimization algorithm
• implement simple probabilistic inference based on statistical data using the Bayes rule
• build linear regression models from data, and use the models to predict variables of interest, such as apartment prices
• use the nearest neighbor method to predict variables of interest
• use cross-validation to avoid under- and overfitting
• build and apply logistic regression and simple neural network models for prediction

Oh, and the final (optional) task of the course will ask you to propose your own AI idea — what kind of problem you’d like AI to solve, and how?

Institution	University of Helsinki
Level	All Levels
Workload	50 hours
Enrollments	1M
Cost	Free
Exercises	Yes
Certificate	Paid

Best Introduction for Python Programmers with Free Certificate (Harvard University)

If you’re confident with Python and are willing to charge head-first into AI, you’ll definitely love this CS50 course! If you take it on Harvard OCW, you can earn an unverified certificate for free. Or pay to get a verified certificate from edX.

This 7 weeks-long rigorous course produced by Harvard University will take some time and effort to complete, as expected of a Harvard course, but it’s definitely worth it. By the end, you’ll be able to work with libraries for machine learning (scikit-learn, pillow, and TensorFlow) as well as design your own intelligent systems based on the AI principles.

Unlike some other free-to-audit courses, Harvard’s CS50 courses on Harvard OCW usually have no time limit and you’ll get full access to the course material, including 7 quizzes and 12 coding projects. So if you prefer well-produced university-style video lectures over text articles, you’d like this course more than the previous Helsinki course.

As for what you’ll learn, I’ll state them below:

1. Search: Finding a solution to a problem, like a navigator app that finds the best route from your origin to the destination, or like playing a game and figuring out the next move.
2. Knowledge: Representing information and drawing inferences from it.
3. Uncertainty: Dealing with uncertain events using probability.
4. Optimization: Finding not only a correct way to solve a problem, but a better—or the best—way to solve it.
5. Learning: Improving performance based on access to data and experience. For example, your email is able to distinguish spam from non-spam mail based on past experience.
6. Neural Networks: A program structure inspired by the human brain that is able to perform tasks effectively.
7. Language: Processing natural language, which is produced and understood by humans.

Institution	Harvard University
Provider	edX
Instructors	Brian Yu and David J. Malan
Level	Beginner
Workload	70–210 hours
Enrollments	730K
Cost	Free-to-Audit
Exercises	Yes
Certificate	Free (unverified), Paid (verified)

Best Business-Focused Introduction for Non-Experts (DeepLearning.AI)

DeepLearning.AI’s AI For Everyone course is similar to Helsinki’s Introduction to AI, in which it is accessible to everyone: no prerequisites required. How it differs is that it focuses more on the business aspects of AI: What can AI do for your team? How do you utilize it just like how Google and Baidu did? And more importantly, what are its ethical and societal implications?

Through multiple case studies, video lectures, and quizzes, by the end of this course you’ll have a better understanding of AI technology than most CEOs of large companies (your smart toaster will finally respect you), and additionally have the skills to help yourself and your company navigate the rise of AI.

Institution	DeepLearning.AI
Provider	Coursera
Instructor	Andrew Ng
Level	Beginner
Workload	12 hours
Enrollments	1M
Rating	4.8 / 5.0 (39K)
Cost	Free-to-Audit
Exercises	Yes
Certificate	Paid

Best AI Math Course for Beginners (KAIST) 

Talking about AI at a party is all fun and thought-provoking, until someone walks up to your group and asks how does linear regression work — then it’s just a game of “Who can pretend to understand linear algebra the longest?”

Well, if that’s a game you’re tired of losing, why not take Math for AI Beginner: Part 1 Linear Algebra? By spending only a little bit of your time, (3 weeks with 2 hours a week) you can get an intuitive understanding behind some of the most famous machine learning algorithms used in AI.

This comprehensive free-to-audit course, brought to you by the Korea Advanced Institute of Science and Technology (KAIST), breaks down the intimidating world of linear algebra into digestible, easy-to-understand components. Of course, this course also gives you an overview of what AI and machine learning is, just in case you’ve just started exploring the world of AI.

Institution	Korea Advanced Institute of Science and Technology
Provider	Coursera
Instructor	Yoon Yong Jin
Level	Beginner
Workload	8 hours
Enrollments	4.3K
Rating	4.8 / 5.0 (20)
Cost	Free-to-Audit
Exercises	Available
Certificate	Paid

Best Hands-On Introduction for Non-Experts (IBM)

Offered by IBM on Coursera, Introduction to Artificial Intelligence (AI) is designed for anyone who wants to learn about AI with hands-on lab experience — no technical or programming background required! Whether you are an executive, a developer, a student, this is a great course to begin learning AI.

The free-to-audit course teaches you what Artificial Intelligence is, explores use cases and applications of AI, discusses AI concepts and terms like machine learning, deep learning and neural networks, and exposes to various ethical issues surrounding AI. Additionally, you’ll get the chance to hear advice from experts about learning and starting a career in AI.

Institution	IBM
Provider	Coursera
Instructor	Rav Ahuja
Level	Beginner
Workload	9 hours
Enrollments	232K
Rating	4.7/5.0 (11K)
Certificate	Paid

Best University-level Introduction with Common Lisp for Programmers (University of Oslo)

INF4820 is probably the most interesting and unique pick in this listing, because of the nature of the programming language it teaches in the course: Common Lisp.

Learning and using Lisp for artificial intelligence may sound incredibly unusual — but Lisp used to be the ‘mother tongue’ of AI researchers, mind you. So if you aren’t afraid of learning a completely new language with a brand new way of thinking (assuming you don’t already know Lisp, of course), then hop aboard the Lisp train!

In this course, you’ll explore foundational theory in AI and NLP, and learn how to implement these concepts using Common Lisp. For example, what is vector space semantics, and how do they tokenize and stem words in corpus? You’ll also learn about Hidden Markov Models that allow you to predict the next word in a sentence, as well as other models.

To take this course, you should have knowledge of computational linguistics (theory of formal languages and grammars).

Institution	University of Oslo
Provider	YouTube
Instructors	Stephan Oepen and Murhaf Fares
Level	Intermediate
Workload	18 hours
Cost	Free
Exercises	Available
Certificate	None

Best Introduction to Generative AI for Non-Experts (Google)

Have you seen the hype behind content-generating large language models (LLMs) like ChatGPT and StableDiffusion? How do they work, and what is a large language model anyways? All these questions and more will be answered in this learning path on Generative AI by Google, no prerequisites whatsoever!

In this free course, you’ll discover some of the techniques and mechanisms embedded in LLMs like BERT, such as Encoder-Decoder architecture and attention. By the end of this course, you’ll know how to create your own image captioning model in a lab walkthrough by using deep learning, and also prototype and customize generative AI with Generative AI Studio, a product on Vertex AI.

Provider	Google
Level	Beginner
Workload	10 days
Cost	Free
Exercises	Available
Certificate	None

Best Technical Introduction for Python Programmers (Microsoft)

Microsoft’s AI for Beginners 12-week course gives you a technical overview of the concepts in AI. Do you want to understand what Neural Networks and Deep Learning are? Do you want to explore other alternative approaches to AI? And do you also want to know how to code them with a popular machine learning library like PyTorch or TensorFlow? Then this course is for you!

How does this course compare with CS50’s AI course? This course is text-based, meaning that if you don’t have long stretches of time where you can sit down and listen to hours-long lectures, you’ll find this course easier to pace your learning. And the course doesn’t skimp out on examples and labs either — there are many Jupyter Notebooks to be explored. And don’t worry about the math, this course doesn’t dive deep into it. It’d rather focus on the intuition and the coding behind the algorithm instead.

As for how you should take this course, it is mentioned in the course. There are pre-lecture quizzes and post-lecture quizzes to warm up and solidify your understanding.

The following are just some of the questions you’d be able to answer in each of the course chapters:

• Introduction to AI: What is AI? What are the categories of AI? What is its history, and how much have we progressed till now?

• Symbolic AI: How can AI possess knowledge? Can AI reason? What are the approaches to this?

• Introduction to Neural Networks: How do we emulate the neurons in a biological brain in an artificial machine? What are the strategies behind layering these neural networks? How does this lead to machine learning?

• Computer Vision: How can computers ‘see’ images as binary data? What are convolutional neural networks and generative adversarial networks? How do we detect ‘objects’ in images?

• Natural Language Processing: What are Recurrent Neural Networks and Generative Recurrent Networks? How do computers differentiate the ‘bark’ in trees and ‘bark’ of dogs? And what in the world is a transformer?

• Other AI Techniques: What are the other ways to teach AI? By mimicking evolution through selection and mutation? From rewarding success and punishing failures in a game? Through achieving emergent behavior of many simpler agents?

• AI Ethics: Fairness. Transparency. Accountability. Privacy. Interpretability. How do these help ensure responsible AI?

• Extras: Is it possible to combine vision and natural language capabilities into a single model?

Provider	Microsoft
Level	Beginner
Workload	12 weeks
Stars	17K
Forks	2.8K
Cost	Free
Exercises	Available
Certificate	None

Best University-level AI Math Course for the Confident (MIT)

If you’re committed to learning the deep mathematical foundations behind many of the algorithms in AI, then the only way to ease your suffering is with a comprehensive, rigorous curriculum and a clear, enthusiastic professor. And to whom could this praise be better directed than to MIT professor Patrick Winston and his course, 6.034 Artificial Intelligence?

This free course introduces students to the basic knowledge representation, problem solving, and learning methods of artificial intelligence. Students will gain a ‘big picture’ understanding of AI, and by the end of the course, able to develop intelligent systems by assembling solutions to concrete computational problems, understand the role of knowledge representation, problem solving, and learning in intelligent-system engineering; and appreciate the role of problem solving, vision, and language in understanding human intelligence from a computational perspective.

Though 6.034 Artificial Intelligence is not centered on programming, the quizzes and problem sets require an understanding of programming. Thus at least some programming experience is required to take this course. The prerequisites for this course are college-level calculus and linear algebra.

Institution	Massachusetts Institute of Technology
Provider	MIT OpenCouseWare
Instructor	Patrick Henry Winston
Level	Beginner
Workload	24–25 hours
Cost	Free
Exercises	Available
Certificate	None

Best Courses Guides Methodology

I built this ranking following the now tried-and-tested methodology used in previous Best Courses Guides (you can find them all here). It involves a three-step process:

1. Research: I started by leveraging Class Central’s database with 150K+ online courses and 200K+ reviews. Then, I made a preliminary selection of 2200+ AI courses by rating, reviews, and bookmarks.
2. Evaluate: I read through reviews on Class Central, Reddit, and course providers to understand what other learners thought about each course and combined it with my own experience as a learner.
3. Select: Well-made courses were picked if they presented valuable and engaging content and they have to fit in a set of criteria and be ranked accordingly: comprehensive curriculum, release date, ratings and enrollments.