Morgan State transforms its cloud computing degree into an AI degree

A university degree can be renamed in a single line; educating someone who can build and evaluate artificial-intelligence systems requires much more. On 17 July 2026, Morgan State University announced that it would launch a Bachelor of Science in Artificial Intelligence that autumn. The news did not describe a school created from scratch, but the transformation of its former cloud computing degree. That distinction turns the case into a practical lesson: to understand what an AI programme really teaches, look beneath its title.

First check: what was actually approved

Morgan State’s official announcement says the degree was approved by the university’s Board of Regents and the Maryland Higher Education Commission. The state agency’s public directory now lists Artificial Intelligence as a Morgan State bachelor’s degree. That confirms the programme’s administrative existence; it does not yet establish its quality, employment outcomes or the readiness of every course. Authorisation answers a different question from outcomes assessment.

The board documents also show that the wording evolved. In the book for the May 2026 meeting, the item appears as a title and academic-code change for the B.S. in Cloud Computing. Its explanation initially proposed “Artificial Intelligence and Cloud Computing”, while the presentation and the July announcement use B.S. in Artificial Intelligence. That does not invalidate the programme, but it is a useful warning: during a transition, a press release, a departmental page and a catalogue may not be synchronised.

The May documentation did not sell the change as a clean break, either. Morgan argued that cloud computing had moved from an emerging specialty to common infrastructure for software, data, cybersecurity and AI. It sought to change the CIP code—the US classification for fields of study—from Information Technology to Artificial Intelligence. The document itself said the change did not abandon the earlier programme’s rigour or technical content, but aimed to reflect its existing emphasis more accurately. The centre of gravity moves; the foundations remain.

Second check: count credits, not buzzwords

The official B.S. in Artificial Intelligence page divides a minimum of 120 credits into 44 for general education and university requirements, 14 for supporting courses and 62 for the AI major. That distribution matters. The degree name describes only part of the journey: almost half of the credits cover general and supporting education, as is common in four-year programmes. A useful comparison calculates what share belongs to the field, which courses are compulsory and how much room is left for electives.

The visible course list combines introductory data science, cloud computing, high-performance computing, programming-language organisation, artificial intelligence and machine learning. It also includes cloud and AI applications, cloud data analytics, AI in cybersecurity, data science for social good and machine learning in the cloud. The pattern is more informative than any isolated module: programming and systems support implementation; mathematics and data support reasoning about models; cloud and infrastructure support training, deployment and observation; cybersecurity and social context force students to consider consequences and risks.

The starting point can also be compared. The official cloud computing proposal approved in 2019 already required 120 credits and included computer science, mathematics, artificial intelligence and a senior project. AI did not suddenly appear in July 2026. The right question is not “did they add AI?”, but “how did its weight, sequence and assessment change from the earlier plan?”. Without a complete, definitive table for the new pathway, the movement of every block cannot be calculated precisely.

Third check: read the fine print on the institution’s own site

The departmental page contains signs that its documentation was still being updated when the launch was announced. The old cloud-computing URL redirects to the AI degree, but several objectives still discuss understanding the principles and importance of cloud computing. That may be consistent with Morgan’s academic case—cloud as an underlying layer for AI systems—but it also shows why a promotional page should not be treated as a final catalogue. Some learning-outcome text even contains typographical errors. A new name does not automatically update every descriptor, prerequisite or assessment mechanism.

This does not justify concluding that the change is merely cosmetic. The board reported 18 AI courses across the department’s undergraduate and graduate offerings and listed topics including reinforcement learning, deep learning in the cloud, explainability and fairness, high-performance computing and quantum machine learning. But “18 courses offered” does not mean “18 compulsory courses in this bachelor’s degree”. An audit must distinguish the department’s total inventory from the modules every student must pass.

The same caution applies to project-based learning. Morgan says every AI and machine-learning course includes real-world projects developed by students and faculty. That is a promising pedagogical choice because it requires students to turn a definition into data, code, testing and documentation. Yet “projects are included” does not reveal their difficulty or marking criteria. Better questions are concrete: is there a held-out test set the student has not seen? Must the system be compared with a baseline? Are failures and costs documented? Is individual work assessed within a team?

Fourth check: distinguish goals from results

The announcement says the programme will prepare students for paths in AI, machine learning, cybersecurity, robotics, cloud computing and data science. These are possible destinations, not guarantees. Because the programme was due to launch in autumn 2026, it could not yet report graduates under the new title, completion rates or programme-specific employment outcomes. Stories about jobs or internships held by departmental students are not a substitute for measurements from the new degree. Verb tense is a clue: “will prepare” expresses an aim; “its graduates achieve” would require observed data.

Ethics deserves the same treatment. Morgan names fairness, bias, privacy, transparency, security and trustworthy systems among its priorities. Including those concepts is better than omitting them, but a list does not prove that they are assessed capabilities. A reader can ask where they appear: in a dedicated module, in technical projects, in pass criteria, or in data and model reviews? The durable distinction is between mentioning a value and requiring evidence that a student can apply it.

It is also useful to distinguish this degree from general computer science. Morgan’s B.S. in Computer Science already identifies AI and cloud computing as learning areas and offers an AI and cybersecurity track. The specialised bachelor’s degree is not valuable simply because it contains AI, but if it organises greater depth, continuity and practice around the field. Comparing the two pathways reveals what specialisation adds and which general foundations may receive less space.

A protocol that works for any AI degree

Before choosing a programme by its name, an applicant can apply five filters. First, verify its status in the regulator’s directory or official catalogue. Second, count credits and separate general requirements, foundations, specialisation and electives. Third, map the sequence: mathematics and programming should precede advanced models, rather than forming an attractive but disconnected course collection. Fourth, request examples of assessment—projects, exams, baselines, documentation and risk review. Fifth, demand outcomes that belong to the programme and its cohort, not testimonials inherited from a former title.

Applied to Morgan State, the protocol produces a more precise reading than “cloud becomes AI”. There is approval and a state listing, continuity in computing foundations, a 62-credit major block and an offering that joins models to systems. Questions also remain that the documents available on 22 July did not fully answer: the definitive semester-by-semester pathway, which courses are mandatory within the broader offering and how the first cohort’s outcomes will be measured.

The transferable capability is this: judge an artificial-intelligence degree by its authorisation, credit map, prerequisites, assessments and verifiable outcomes—not its label. The title states where a university wants to orient a programme; the curriculum and its tests show what a student will actually have to learn.