Why trusted data is vital to AI success

Investment in trusted data will be vital to the progress of AI in the public sector. “At the center of artificial intelligence is going to be, what data does the algorithmic analysis use to deliver the results?” says Rob Carey, president of Cloudera Government Solutions and former CIO of the U.S. Navy in a new podcast for FedScoop.

According to Carey, many federal agencies face significant hurdles in this area, such as incomplete or inaccurate data from legacy systems. These challenges — including data quality, location and compliance — require attention before deploying AI solutions. “If you’re assuming the data is perfect or near perfect, it’s not,” says Carey. “You need to go back and take a look and fix it first because it will affect the accuracy of the analysis.”

Carey also stressed the need for a solutions-driven mindset. “AI should support specific decisions rather than being pursued for its own sake,” he says. Agencies must define clear objectives, understand data requirements and tackle projects incrementally. “Start small and move toward more complex applications,” says Carey, cautioning that AI is not a “silver bullet” and requires careful planning.

The complexities of hybrid cloud environments add another layer of difficulty, with data often spread across on-premises systems and multiple cloud platforms. According to Carey, using tools that enable data analysis without relocation is key. “You want to get to the answer without worrying about where the data resides,” he says, underscoring the importance of robust access controls and governance frameworks to ensure security and accessibility.

In addition, Carey championed modern data architectures, such as data lakehouses, to break down silos and enable seamless access to structured and unstructured data. “A data lakehouse simplifies management and allows agencies to analyze diverse data types effectively,” he says.

Carey’s advice for agencies embarking on their AI journey is clear: “Get your data house in order. Build proficiency in AI first, and then expand.”

Listen to the entire podcast conversation here. And learn more about how Cloudera can help government agencies accelerate AI with trusted data.