b'The Tools: Success Criteria for Artificial Intelligence It is said that artificial intelligence is neither artificial nor intelligent. Nevertheless, the phrase captures the sheer power of a suite of tools which, over the past decade, have begun to transform industry after industry.In the old way, leaders managed with glorified spreadsheets, and individuals making human-powered predictions. They looked at data, entered it into spreadsheets, created simple trend lines, and judged how to respond. If the future did not pan out according to the prediction, the human at the helm would revise and try to do better next time. Keep in mind that the complexity and speed of data in the past were a fraction of what we see today, and trends were stable.With the new methods, all of that is now differentand indeed, obsolete. The new tools are integrated to create what we now refer to as artificial intelligence: Enterprise data warehouse: accumulating and Computing power & simulation engines: enabling storing data. AI cannot work without access toperformance and scale. The industries that led the enormous amounts of data. In U.S. health care, manychange for AI invested millions of dollars into their own larger hospitals and systems have had access togigantic servers. The purpose of all that computing such data for decades. Yet, they have been unablepower is to run hundreds of thousandseven millionsto properly leverage that data to build anythingof simulations. The more complex the environment other than rudimentary dashboards or glorifiedbeing simulated; the more power is needed. Until spreadsheets. recently, this amount of computing power has beenSophisticated algorithms: ensuring efficientinaccessible or cost-prohibitive for all but the largest, processes and continued learning. The new fieldmost well-resourced organizations. In the last 3-5 years, that has changed. Cloud-based computing power of data science has been instrumental in helpingis much more accessible, making these simulations organizations begin to put this data to effective use.available to many more sectors.The era of big data would not be an era without the expertise of data scientists to make sense of it. In this Machine Learning: leveraging data, algorithms, and sense, artificial is not artificial at all: AI still relies onmodels. Machine learning, a phrase often intended to be the judgments of those who build the algorithms.synonymous with artificial intelligence, is the outcome Yet those judgments must be informed by dataof the above three components. Computer systems that science, statistical analysis, and strategic leadershipcan learn and adapt, by data, algorithms, and models, decisions. can draw inferences and create predictions useful for decision making. Enterprise dataComputing power & warehouse simulation enginesSophisticatedMachine LearningalgorithmsTogether, we healTogether, we heal 2Together, we heal SCP HEALTHIINVESTING IN AI: PROVEN STRATEGIES TO FUEL TRANSFORMATION IN HOSPITAL OPERATIONS'