RGB is the colour model in which every pixel is described by three numbers: how much red, how much green, how much blue. Mix those three coloured lights at different intensities and you can produce any colour that human vision can normally perceive. This is the natural way for any light-emitting device — TV screens, phone displays, projectors — to work, because their pixels physically are tiny red, green and blue light sources.
Cameras also capture in RGB (or in a close cousin), and image editors and design tools work in RGB. Where RGB isn't used is the middle step — actual video compression. Inside H.264, HEVC and AV1, video is converted to a different colour format called ycbcr that separates brightness from colour, which compresses far better and matches how the eye works. So the typical journey is: camera captures RGB → encoder converts to YCbCr → file/stream stays in YCbCr → player converts back to RGB to send to the screen.
For most product and business decisions, the only thing to remember is that RGB and YCbCr both describe the same colours and convert between each other reliably; you don't have to think about it unless you're doing colour-critical work (mastering, broadcast graphics, VFX). When you do, the conversion is governed by a chosen color-space like BT.709 or BT.2020 — the recipe that says exactly what "red", "green" and "blue" mean.
