AI in government relies on supportive leadership and a supporting ‘glass breaker’

More technologically savvy stakeholders and committed leadership within the federal government are critical to the successful implementation of scalable artificial intelligence technology in public offices.

Speaking during a virtual roundtable, officials including Jack Shanahan, the first director of the Defense Department’s Joint Center for Artificial Intelligence, discussed the need for modernization advocates in federal agencies, having specifically need a “disruptor” leading the implementation of new software.

“If this is really the first big AI project designed to go fast and at scale, you need an almost classic type of icebreaker that’s just going to jump through all these bureaucratic hurdles,” he said. -he explains.

He added that agency leaders and deputies must be fully on board with new technology deployments to help ensure that adequate oversight and accountability are present in AI acquisitions. Shanahan also noted that more employees with technical aptitude are essential.

“You need to bring in someone who knows how it’s done in the commercial software world,” he said. Shanahan added that clear funding and investment is also a key part of the agencies’ artificial intelligence programs, namely Project Maven, an initiative to apply artificial intelligence to data collection through surveillance.

Suzette Kent, the former federal chief information officer, agreed that the internal culture must respond to the implementation of more advanced technologies.

“The more powerful the abilities, the more skills…that need to be at the table,” she said.

Artificial intelligence technologies that manage large amounts of data also pose challenges to effectively scale these technologies.

Deputy Director of Science and Technology at the Central Intelligence Agency, Dawn Meyerriecks, described her office’s experience with organizing data that is supposed to work with new machine learning technology.

“People really need to understand their data in order to make it available in a way that makes sense,” she said. Meyerriecks said she and her colleagues at the CIA ran several pilot projects to see which software worked best with different types of data collected by the agency.

Shanahan added that the JAIC has taken similar steps in implementing machine learning technology, saying his organization has focused on the stages of the AI ​​stack, including forming a larger pipeline. solid between the initial collection of data and the construction of a sustainable digital platform.

His team also worked to develop key infrastructure to support the initiative, including cloud, testing professionals, data protection strategy, centralized management, and credential security.

He added that bringing these technologies into government can help close the gaps in public agencies introduced by legacy technologies.

“There are all kinds of opportunities to use AI and ML,” Shanahan said. “It’s not going to happen overnight, but you need to have some kind of technology emergency. It’s time to move beyond small-scale pilot projects and move towards fusing innovation into the It’s difficult, but we don’t have a choice at the moment.

About Perry Perrie

Check Also

Technology-Assisted Review Improves eDiscovery

Over the past decade, technology-assisted examination (TAR) has become an increasingly important tool in eDiscovery, …