Advances in quantum computing and machine learning are likely to change the face of quantitative portfolio construction and risk management as we know it today, and the focal point will be optimization processes. While financial optimization theory is highly sophisticated and complex, the current state of practice leaves much to be desired and may best be described as a patchwork quilt held together by band-aides and duct tape. On the horizon, however, are potential improvements in the analytical techniques underpinning how optimization methods are used, including the promise of exhaustive searches using quantum computers and advances in pattern recognition available through structured machine learning. To understand the importance and promise of the new developments in technology for financial optimization, it is imperative to appreciate the state of current practice. Critical challenges exist in the internal consistency of volatility and correlation estimates given the mixed methods used in many quantitative practices. With the heightened occurrence of event risk coming from politics, policy, and disruptive innovation, common assumptions concerning the stability of volatility regimes and correlation estimates are in question. Moreover, event risk can create short periods when bimodal expected return distributions dominate, often resulting in underestimation of the potential for pricing gaps and volatility regime shifts. Future progress with exhaustive search optimization using quantum computers and structured machine learning offers the possibility of a much deeper assessment of the probabilities surrounding event risk, improved analysis of the potential presence of bimodal and other non-normal return distributions, and the construction of more robust portfolios to handle the extreme (or fat-tailed) risks that seem to be happening more and more often than traditional approaches tend to predict.