This paper reviews recent work proposing paradigm changes for the currently popular approach to constitutive and failure modeling, focusing on the use of non‐associated flow rules to enable greater flexibility to capture the anisotropic yield and flow behavior of metals using less complex functions than those needed under associated flow to achieve that same level of fidelity to experiment, and on the use of stress‐based metrics to more reliably predict necking limits under complex conditions of non‐linear forming. The paper discusses motivating factors and benefits in favor of both associated and non‐associated flow models for metal forming, including experimental, theoretical, and practical aspects. This review is followed by a discussion of the topic of the forming limits, the limitations of strain analysis, the evidence in favor of stress analysis, the effects of curvature, bending/unbending cycles, triaxial stress conditions, and the motivation for the development of a new type of forming limit diagram based on the effective plastic strain or equivalent plastic work in combination with a directional parameter that accounts for the current stress condition.