I did not do my Master's in Artificial Intelligence straight out of my undergraduate degree, and I did not do it to get into the job market. I had already been working in data and analytics for several years and I was progressing in that field. What I wanted was a small pivot into something more specialised, some deeper knowledge, and a few extra doors opened. I believe it did open them.
That is the honest frame for everything that follows. If you are asking whether a Master's in AI is worth pursuing, my answer comes from someone who did it part-time, alongside a job, in a field he was already working in.
The rubber stamp
The first thing the Master's gave me was credibility. A lot of companies and a lot of industries still value formal accreditations like degrees over certifications and career-based training. That is not always the case, and in AI it is becoming less and less the case because of how quickly the field moves. It is worth noting that even where organisations say they have moved to skills over qualifications, the follow-through is patchy: research from Harvard Business School and the Burning Glass Institute found that fewer than 1 in 700 new hires actually benefited from companies dropping degree requirements.
In practice, the credibility showed up in two ways for me. It contributed to a slight pivot in my career, where I was hired as an Analytics Domain Architect, and then on through consulting work to my current role as AI Orchestration Architect. And in consulting conversations, having the Master's worked like a rubber stamp from a university saying I know what I am talking about. Someone who is completely self-taught potentially does not have that.
I want to put a caveat on that straight away. That credibility decays very fast. It is not something I believe you need, and it is not something I believe is a requirement in my field any more, or will be as I go forward. The things I have built in my current role and my previous one, the tools, the tech stack, I think those all matter more.
In my own situation the decay is hard to measure. I did the course part-time over four years, including a year out, and a number of doors opened just from starting it and having some results to show. When it came to getting my current role, a lot of the focus was on demonstrating what I can deliver rather than on what degree I hold. I believe the degree gave me skills that let me deliver more, at better quality and faster, but it was not an explicit requirement. That might not be the case for someone doing the Master's straight after their undergraduate degree.
Where it has kept its value for me is in the speaking space, which I have really enjoyed. The Master's lets me go into genuinely technical detail, and my commercial experience from working gives me the business side of it. The two together are what make those conversations work.
The lag
One of the criticisms I would make of my Master's is academic lag. Academic lag is common in most industries. Computer science is probably the most extreme example, along with any other emerging field, and AI even more so. This is a recognised problem rather than a complaint about one course: an AAAI paper on generative AI education describes generative AI competency as increasingly valued in industry but not in higher education, with students experimenting without formal guidance.
What my Master's covered was a lot of the fundamentals. Optimisation, probability, evaluation, and the understanding of what an LLM is actually doing under the hood. That has genuine value.
But almost none of what I do day to day came from the Master's. I am doing prompting, multi-agent orchestration, building agents and writing guardrails in my role at the moment, and none of those are skills the course taught me. All of that I have learned through work, and those are the things I have valued most from work.
One thing that gives me some hope here is that my course was not static. Because I did it part-time alongside my job, I could see them making changes for the cohorts coming after us while I was still on it. I believe that is how they are trying to keep up. There will still be a lag, but hopefully they keep that going.
Technical or general
There were two paths on my Master's, a technical one and a more general one. I opted for the technical path, coming from a computer science background anyway, and I thought that would be the most useful for me.
What I hope gets added in future is more of what businesses are actually doing now, which is building agents. Models and prompt engineering feel to me like skills that need to be in the course. I can build a neural network from scratch and code it purely that way. I can write natural language processing from scratch. Those skills have helped me, they are part of my skill set and part of how I build things, and that understanding supports what I do. It is just not what I do any more.
So would I recommend it?
If you want to go on to a PhD and do research-based work in AI, then yes, and strongly. That is a very valid field and something I would encourage a lot of people to do.
If what you are after is the rubber stamp, or you are trying to get into the field, then I do not think it is a choice between one thing and the other. I think formal accreditation and career-based credentials are complementary skills. I have both, and I would encourage anyone to get both.
Whilst I did come from a computer science and analytics background, I still believe this advice applies to those trying to break into AI from other fields, or with no experience at all. There were a number of people on my course who did not come from a computer science, analytics or AI background, and they were using it to make a full career change rather than the small pivot I made. They were successful in doing that.
The cost is real and worth being straight about. Mine came to around the £12,000 mark, done part-time alongside work. I would not have been able to do it full-time. I could not have afforded it and I could not have taken that gap in my responsibilities.
Part of the reason is how the funding works. When I did my course the loan was lower than it is now, and it was not enough to cover the cost of the course, so the difference had to come out of savings. Unlike an undergraduate degree there is no separate tuition loan and maintenance loan. It is a single loan that is not means tested and that you split between fees and living costs as you decide, and it is meant to cover all of it. For anyone in England looking at similar numbers now, the postgraduate Master's loan is £13,206 for courses starting on or after 1 August 2026, and it covers part-time study. It is also repaid alongside your undergraduate loan rather than added to it, which is probably a topic for another day.
Feel free to reach out if you'd like to know more or have any questions.