Analytics Requires Trusted and Ethical Use of Data and AI/ML
Analytics are critical to the efficient and effective operation of most modern organizations; to that end, trusted analytics require trusted data. Inaccurate and untrustworthy data means organizations cannot rely on the analyses produced using that data. Trustworthiness requires accurate, high-quality data, but also appropriate use of the data that complies with an organization’s access and risk management policies. High-quality data enables high-quality decisions, but with the ever-growing volume of data flowing within organizations, conducting quality control can be extremely labor-intensive. Nearly two-thirds (64%) of organizations in our research report that ensuring the quality of their data is one of the tasks where they spend the most time, second only to preparing data for analysis.
Not only does analytics need trusted data, so does artificial intelligence (AI). AI models must be based on accurate data to produce accurate results. And with many AI models being used in automated processes it is important invest in ensuring the trust and quality levels in the data that feeds AI. Ethical use of data and AI has also become an important concern as organizations increase their investments in AI and machine learning (AI/ML). Our research shows nearly three-quarters (72%) of organizations are planning to increase their use of AI/ML. With the rise in usage, concerns over ethical use of AI/ML have risen as well. This is because data sets may contain biases that manifest themselves in the models developed using those data sets. Consequently, organizations are working to detect and prevent biases before they are deployed into production. And legislators in Europe and North America have introduced legislation that would regulate the use of AI/ML.