Now, I’m a firm believer that absolutely, any 2-d artist should be able to work from life, and should whenever they can (these skills are essential in simply being able to paint well. Full stop.) BUT, there are always going to be those situations where it just makes more sense to work from a photograph. Now, some artists (especially the more “old school” ones) take issue with working from photographs as opposed to direct observation. But I believe that one can work from a photograph while still making a great painting that isn’t a mere copy of the reference image.
For example, I did a painting of a grenade using a reference photo. I found it simpler to do this because of the placement (flat on a table) and lighting I wanted would be easier to reference from a photo rather than having to do a setup where I would somehow have to position my face directly parallel to a table while trying to paint.
The photo gave me the key information I needed- things like proportion, angles, major color/value changes. But as you can see, the finished painting isn’t an exact copy of my reference:
I think a good rule of thumb is to not spend too much time looking at the reference itself. Especially once you’ve got the key information down, you have to look at the painting that you’re making, and make decisions based on how to make your painting a successful one- not a mere copy of the reference photo. Basically, as I go, I look less and less at the reference (it helps that all my reference photos quickly become obscured with paint smudges anyway.) The camera can be the enemy of a great painter, but it can also be a great asset. Like most tools, it’s all in how you use it.