Yes - and for most managers it is the right choice. The confusion comes from the word "certification" covering two completely different assessments.
Two kinds of AI certification
Practitioner certifications assess whether you can build: Python, notebooks, model training, sometimes a timed lab. Without coding you will fail these, and no amount of study weekends fixes that in a month.
Leadership certifications assess whether you can decide: use-case selection, risk and governance, business cases, adoption. The assessment is written or project-based. No code, no maths exam.
If a syllabus mentions environments, libraries or datasets you must run, it is the first kind.
What non-technical candidates are assessed on
Typically a short applied piece of work - an AI opportunity assessment for your own organisation, a risk register, or a business case. That plays to your advantage, because you have the organisational context that technical candidates usually lack. More on the split in AI skills vs AI leadership skills.
What you do need
Not code, but three things: comfort with numbers at a business level, willingness to learn new vocabulary, and the patience to sit with concepts that stay slightly fuzzy. You will never fully understand how a model works internally. Neither do most people who deploy them.
Where to start
Understanding AI: What Leaders Need to Know is written for exactly this reader, and Leading AI Innovation & Organisational Transformation takes it into practice. Both are leadership-track, no coding.
The bottom line
A non-technical background is a barrier only to practitioner certifications. For leadership ones, it is the intended starting point. Related: can I learn AI strategy without a technical background and should I learn Python before AI.