In the realm of science, anything that is quantitative can be used. Math is a popular subject for people engineering, statistics, engineering, and, occasionally, Physics.

In particular, it is dependent on the specific aspect of AI/ML you're performing. I'll go through some examples and then go over them separately for you.

Research: To conduct the appropriate research in this area, it is often necessary to have a foundation of technical information.

From a computer science perspective In the context of computer science, unless you intend to study the development of new kinds of AI It is important to ensure to be math-literate rather than math-theoretic. A large portion of your job will involve finding ways to use artificial intelligence or creating new models for ML. In the event that you follow the typical way of thinking.

From an applied math or statistics viewpoint, the majority the work you do will involve ML. Most likely because mathematicians view issues in AI as issues that fall within one of the domains and the deficiency of rigor that is associated with artificial intelligence solutions.

In engineering, unless you're an engineer in software it is not common to do research until after you've completed a masters research, and even then you'll lack an inordinate amount of knowledge. It is possible to blame your common core science for this. Similar to physics.

Each of these fields with a master's degree in data science holds the potential to bring their ideas to the forefront.

Industry:
If you are proficient in basic statistics and calculus then you are able to begin working in any of these positions. But the degree expertise you have will determine the job you will be doing. In the past, lots of engineers are analysts, and this is still the case. Physicists too, computer scientists, mathematicians, etc. Since data science is an intersection with AI as well as data science, the majority will do it if they have the opportunity.

If you've got a solid CS or STAT background, the best choice is usually engineering based on ML, as it will pay more and is closer to the top levels of management currently. You're highly visible as you are involved in crucial issues and can provide directly applied applications for wealth within the organization. It can be complicated, however.

Conclusion:
It's a huge reduction, this should provide you with a base to begin. I suggest picking between analytics using statistics and data-oriented information and computer science to build your engineering, technical knowledge. The amount of knowledge is less important than the effort you invest in gaining domain knowledge and practical knowledge.