Academy · Learn · Free programs

Study AI Like a Top-University Student, Free

Three self-study programs, at Bachelor's, Master's and PhD level, built only from free courses that universities in the QS World University Rankings 2027 top 100 publish for anyone. Each is laid out like a university program: foundations, a core, a specialization and a capstone.

No fees, no enrolling, and no degree. The learning is real; the credential isn't.

Programs
3
Courses
26
Universities
12
Countries
7

Program 1 · Bachelor’s level

A Free Bachelor’s-Level Program in AI

For anyone starting out. It assumes high school algebra and trigonometry, and no programming. Four stages: programming, the math, AI, then deep learning, with a capstone at the end.

  1. 1Programming and computer science
  2. 2The math under machine learning
  3. 3Artificial intelligence
  4. 4Deep learning
  5. ✓Capstone

Stage 1Programming and computer science

Take both, in order

Harvard University

QS #5

CS50x: CS50’s Introduction to Computer Science

Free: the whole course as OpenCourseWare: lectures, problem sets and a final project; a certificate costs extra on edX

Self-paced; no programming experience needed

Open the course

MIT

QS #1

6.006 Introduction to Algorithms

Free: lecture videos, notes, and problem sets and exams with solutions

Spring 2020

Open the course

Stage 2The math under machine learning

Take all three, in order

MIT

QS #1

18.01SC Single Variable Calculus

Free: lecture and recitation videos, notes, and problem sets and exams with solutions, built for independent study

OCW Scholar, Fall 2010; assumes high school algebra and trigonometry

Open the course

MIT

QS #1

18.06SC Linear Algebra (Gilbert Strang)

Free: every lecture on video, summary notes and problem-solving videos

OCW Scholar, Fall 2011

Open the course

Harvard University

QS #5

Stat 110 Probability (Joe Blitzstein)

Free: lecture videos on YouTube, a free online edition of the textbook, and a free edX course

Needs single-variable calculus and some matrices

Open the course

Stage 3Artificial intelligence

Take CS50 AI and CS 188

Harvard University

QS #5

CS50’s Introduction to Artificial Intelligence with Python

Free: audit it on edX at no cost; a certificate costs extra

Self-paced, about 7 weeks at 10–30 hours a week

Open the course

UC Berkeley

QS =20

CS 188 Introduction to Artificial Intelligence

Free: slides, lecture videos, the Pac-Man projects and past exams, open for self-study

Videos from 2012–2014; projects still used today

Open the course

MIT

QS #1

6.034 Artificial Intelligence (OpenCourseWare)

Optional

Free: video lectures and course materials under a Creative Commons license (CC BY-NC-SA)

Fall 2010 recording, a classic

Open the course

Stage 4Deep learning

Take 6.S191

MIT

QS #1

6.S191 Introduction to Deep Learning

Free: lecture videos, slides and software labs

2026 edition

Open the course

National Taiwan University

QS #54

Introduction to Generative AI and Machine Learning (Hung-yi Lee)

OptionalTaught in Mandarin

Free: lectures on YouTube and the course page

Fall 2025

Open the course

Capstone

Build something, then govern it.

Your CS50x final project is the build. Then run it through the Foundation’s Practice Level 100 worksheet: one hour on a real case, a rule in your own words and a signed receipt.

Program 2 · Master’s level

A Free Master’s-Level Program in AI

For anyone who has finished the Bachelor’s-level program, or has the same programming, calculus, linear algebra and probability. A shared core first, then one track in depth.

  1. 1Core: machine learning
  2. 2Core: deep learning
  3. 3Specialize
  4. ✓Capstone

Stage 1Core: machine learning

Take CS229

Stanford University

QS =2

CS229 Machine Learning

Free: the full 2018 lecture series on YouTube and the lecture notes; some current class files need a Stanford login

Autumn 2018 lectures

Open the course

Stage 2Core: deep learning

Choose one of the four

Carnegie Mellon University

QS #55

11-785 Introduction to Deep Learning

Free: lecture videos on YouTube and the course site

Every semester

Open the course

New York University

QS #58

DS-GA 1008 Deep Learning (Yann LeCun, Alfredo Canziani)

Free: captioned lecture videos, slides, PyTorch notebooks and written notes

Spring 2021

Open the course

Technical University of Munich

QS #25

I2DL Introduction to Deep Learning

Free: lectures and exercises, open to learners outside TUM

Current semester

Open the course

University of Toronto

QS #32

CSC413/2516 Neural Networks and Deep Learning

Free: slides, Roger Grosse’s course notes and notebooks

Winter 2023 site

Open the course

University of Amsterdam

QS #60

Deep Learning 1 and 2 tutorials (MSc AI)

Optional

Free: Jupyter notebook tutorials

Pairs with any of the four

Open the course

Stage 3Specialize

Pick one track

Track A · Language and large language models (take CS224N, then 11-711)

Stanford University

QS =2

CS224N Natural Language Processing with Deep Learning

Free: lecture videos (latest Spring 2024), slides and assignments

Spring 2024 lectures

Open the course

Carnegie Mellon University

QS #55

11-711 Advanced Natural Language Processing

Free: Fall 2024 lectures on YouTube; course site and assignments

Spring 2025 site

Open the course

The University of Tokyo

QS #39

Large Language Model course (Matsuo-Iwasawa Lab)

OptionalTaught in Japanese

Free: the LLM course content, released free of charge; most of the lab’s courses are free, a few are paid

Released September 2024

Open the course

Track B · Reinforcement learning (then Berkeley CS 285 at PhD level)

UCL

QS =8

Reinforcement Learning (David Silver)

Free: ten lecture videos and slides

2015, still a standard start

Open the course

Capstone

Ship a governed system.

Build a project in your track, then take it through the Foundation’s Practice Levels 200 to 400: retrofit a rule onto it, hold it, and prove it.

Program 3 · PhD level

A Free PhD-Level Program in AI

For anyone who has finished the Master’s-level program. A PhD is research under an advisor, and no page can give you that. This program takes you to the research frontier: depth first, then the seminars where the field is argued.

  1. 1Qualifying depth
  2. 2Research seminars
  3. ✓Capstone

Stage 1Qualifying depth

Choose two

Stanford University

QS =2

CS336 Language Modeling from Scratch

Free: lectures on YouTube; assignments on GitHub

Spring 2025

Open the course

UC Berkeley

QS =20

CS 185/285 Deep Reinforcement Learning

Free: the Fall 2023 lecture recordings on YouTube and the course site

Fall 2023 recordings

Open the course

MIT

QS #1

6.5940 TinyML and Efficient AI Computing (Song Han)

Free: slides and lecture videos

Fall 2026

Open the course

Stage 2Research seminars

Follow all three

Stanford University

QS =2

CS25 Transformers United (seminar)

Free: anyone can audit the livestream, no sign-up; recordings go up on YouTube

V6, Spring 2026

Open the course

UC Berkeley

QS =20

Advanced Large Language Model Agents (Dawn Song)

Free: lecture recordings, slides and the reading list for each lecture

Spring 2025 open course

Open the course

Harvard University

QS #5

CS 2881R AI Safety (Boaz Barak)

Free: lecture recordings on YouTube and the course site

Fall 2025

Open the course

Capstone

Ask one question and answer it in public.

Pick one open question from the seminars. Reproduce one paper on it, write up what held and what didn’t, and publish it. Then put your result through Practice Level 400: prove it.

How we built these programs

Only top-100 universities

Every course comes from a university in the QS World University Rankings 2027, published 18 June 2026. Ranks shown are QS 2027; "=" means a tie.

Free without enrolling

A few ask for a free account, and some current class files are for enrolled students only. Each card says exactly what’s free.

Our sequence, their courses

The order, the levels and the capstones are the Foundation’s suggestion. No university designed or endorses these programs, and none of them gives credit or a certificate through us.

Checked 15 November 2026

Every link goes to the university’s own course page, checked on 15 November 2026. Offerings change; if one has moved, please tell us.

Your next step

Learning AI so you can govern it?

Try your first gate on one AI you already know. Ten minutes, no account.

Run the Gate Check
Created using AI tools

These are the universities' courses, not the Foundation's, and these programs are not degrees. Listing a course isn't an endorsement by the university of the Foundation, or the other way round, and we receive nothing for it. Suggest a course: email Bob and Ken. Rankings: QS World University Rankings 2027 results.