Agencies' digital transformation efforts in areas like artificial intelligence must also consider workforce needs, according to a panel of government technology experts. Speaking at an ATARC event on Thursday, the panelists asserted that it does not matter how good the data or AI is, if people do not know how to use it correctly or understand it. As a result, the panelists emphasized the need for data literacy, education and training. "I can build the best AI model, but if I put it in the hands of my investigator, and if he has a ton of questions, then we just lost them," Ben Joseph, chief data officer for the United States Postal Service Office of Inspector General, said. "Earlier this year, we actually punched out a small program in terms of data literacy…so we educate my workforce, investigators, auditors and everybody else, like'how do you interpret data?'" "It's almost like you have to right-size the AI education for the position or the role that the individual is playing in the lifecycle," William Streilein, chief technology officer at the Department of Defense's Office of the Chief Digital and Artificial Intelligence Officer, said.

Amazon, Microsoft, Google, Uber, Facebook -- all spend millions of dollars on AI solutions. Although each of these systems were designed and built to accomplish a task in good faith, they each failed because of unintended predictions. How could this possibly happen? Each of these models made poor decisions not based on the data they had, but based on the data they didn't have. As an AI PM, I created the Data Sprint to shortcut the endless debate and compress months of work into a single hour.

Because AI applications require thousands of images to be tagged, humans can tag poorly or introduce errors - especially because current tools are simple picture editing tools, like Microsoft Paint, which weren't built for this purpose. Even small imprecisions, compounded over thousands of images, can have a large impact on the accuracy of a computer vision model. And if you think about a production-grade product or solution, every percentage point increase in accuracy can have a big impact on the organization.