Where AI Is Concerned, Long-Game Policy Matters

It’s no longer enough to solve for the challenge in front of you when it comes to artificial intelligence-related policy. That was the consensus reached by a panel of public- and private-sector experts at the Los Angeles Digital Government Summit last week.

Since AI and generative AI burst onto the scene, organizations have been rushing to implement it to solve the problems that seemed out of reach with their human-only workforces. The policies that came with those implementations were often too narrow, like looking through a keyhole.  

“A lot of the policies that we’re seeing early are built around the way a lot of the technologies are looking and working now from a consumer AI approach, and not as much as the enterprise AI approach,” said Jayson Dunn with AWS. “Everyone is concerned about the data and how the data is going to be used, how the bias is going to be identified, as the data can be secured. And so a lot of policies are not specific enough for being written around use cases.”

CIO Ted Ross said AI represented a “seismic shift” for the city that began before the explosion of ChatGPT. The city released its AI Roadmap in June, which divides the technology into categories — embedded solutions, artificial intelligence as a service, and vendor custom-built solutions.

“It used to be that machines did the work and humans did the thinking. AI provides very powerful tools in which now machines can participate in the thinking,” Ross said. “What I just said is extremely exciting and extremely terrifying at the same time, and that’s where we see AI. It’s really important to break it up into its pieces.”

Ross added that the tools are just not going to fall out of fashion and will only increase in prevalence, adding even more incentive to make sure that staff are prepared to interact from a position of knowledge.

“If I want a workforce that is modern, I’m not just deploying a tool to them. I need them to start to really think through and be a part of this new partnership and how to engage,” Ross said.

“I think we see this as a golden opportunity to not just implement tools, but to really raise the digital IQ of our organization,” he added.

Kris Kademian, division chief of the emergent technology group within the Los Angeles County Auditor-Controller, said that AI has become a useful partner in the work the office does — especially where daily reporting processes are concerned — but he emphasized the importance of being able to verify and monitor the information the tools produce.

“I think that’s going to take a little bit of time to get to a place where we can trust the information that it delivers,” Kademian said.

The Los Angeles County Department of Human Resources has also been relying on the technology to help manage its workforce — from recruitment and onboarding to performance management. Department CIO Roozan Zarifian said that one essential part of implementing any new technology is educating employees and approaching implementation with the strategic goals of the broader organization in mind.

The department is currently in the process of building a generative AI tool that will help its investigative team generate reports across documents, transcripts, and other information channels.