Columbia Engineering Launches a New AI Master's Program (MSAI): Inaugural Admissions Now Open

Columbia Engineering has officially launched a Master of Science in Artificial Intelligence (MSAI) this year, with the inaugural class enrolling in fall 2026. The online part-time version is currently open for applications. Here are several points worth noting.

Program Positioning

According to the official description, the MSAI is designed to fill the gap "between purely technical degrees (computer science, data science) and purely domain-focused degrees (healthcare, finance)." Technical depth combined with cross-disciplinary application is at the heart of the design. Columbia's overall investment in AI is substantial: the engineering school has more than 70 full-time AI faculty, more than 10 major AI research centers, and a cross-disciplinary structure linking multiple schools, all of which add up to a strong resource base.

Program Structure

30 credits, 5 consecutive semesters, and roughly 2 years to complete on a part-time basis. The curriculum consists of 4 core courses (Machine Learning, Deep Learning, Natural Language Processing, and Computer Vision), 4 specialization courses, and two semesters of a capstone project, or 2 elective courses in its place.

Differences Between the Online and On-Campus Versions

The two versions confer the same degree and share the same graduation requirements. The main difference lies in the range of specialization options available.

The on-campus version is offered jointly with multiple Columbia schools, covering more than 10 directions, including medicine, public health, arts (officially called "Art, Creativity, and Media"), statistics, architecture, robotics, and policy, with a journalism direction potentially being added in the future.

The online version currently offers 3 specializations: AI for Finance and Operations, AI Infrastructure, and AI and Advanced Computing, focused on engineering and financial applications.

Who It's For and Prerequisites

The online part-time version is geared toward working professionals with a background in computer science or a rigorous quantitative field. All entering students need: programming proficiency in Python, knowledge of data structures and algorithms, linear algebra, and foundations in calculus and probability. Columbia provides bridge courses for applicants from completely different fields, but the online version maintains relatively strict quantitative prerequisites.

Key Dates (Online)

Priority application deadline: June 19

Regular application deadline: August 15

Application fee waived for a limited time

Han Education's Overall Read on the Inaugural Class

As a brand-new program, there is no graduate employment data yet to draw on, which is something to factor in when considering an early entry. That said, the program's academic structure and resource alignment are already clear. The co-directors are Vishal Misra (Chair Professor of Computer Science and Vice Dean for Computing and AI) and Garud Iyengar (Director of the Data Science Institute and Professor of Industrial Engineering and Operations Research), both well-established scholars. On the teaching infrastructure side, Columbia's CVN (Columbia Video Network) platform has years of experience running mature online engineering programs. Admissions standards for an inaugural class are also generally a bit more flexible, which is an opportunity for those who can catch the application window.

Han Education's Preliminary Take

Worth applying. The program's academic positioning, faculty lineup, and the university's deep AI resources all hold up, making it well-suited for working professionals with a quantitative background who want a structured way into AI.

If you have any questions towards college application, feel free to reach out to us, our consultants are more than happy to provide more insights to you!