What is generative AI? Most of the public sector workforce doesn’t know
Connecting state and local government leaders
A recent survey found that only about a third understands the technology, and that even fewer use it daily. But a few basic approaches could change that, experts say.
Uses for generative artificial intelligence are rapidly increasing as more governments embrace the technology. It has already been applied to streamlining procurement, finding data to see how cities compare to their peers, drafting communications and summarizing meetings, among other approaches.
Despite the optimism and the initial use cases, though, most employees don’t have a clue what generative AI is. What is the difference, after all, between that and just plain, old artificial intelligence?
This knowledge gap throws up barriers to getting employees on board. A recent survey by global data and AI company SAS underscores the challenge facing government leaders. It found last week that just 37% of the public sector workforce understands generative AI well or completely. (In other words, more than a third understand that generative AI focuses on creating new content, while traditional AI focuses on analyzing and interpreting data.)
The survey also found that just 13% of public sector organizations said they use the technology daily to at least some extent. Fourteen percent of public sector agencies said they do not use generative AI at all, which SAS said is the highest non-usage rate among the sectors surveyed.
Jennifer Robinson, a global government strategist at SAS who is also mayor pro tem of Cary, North Carolina, said those responses show there is still a lot of “confusion” around generative AI and its capabilities among workers. They also show the caution governments are taking in adopting the technology.
“One thing that governments don't want to do is they don't want to get into trouble for stepping out,” she said. “I often say that governments, at least local governments, always want to be first to be second; they want to see somebody else do it. Then they want to know there were no pitfalls [before] they’ll go ahead and jump into it.”
One obstacle to adoption could be the lack of policies governing its use. Only 52% of governmental organizations surveyed have a generative AI policy, the lowest of any sector. That may come as a surprise to many given the seemingly daily drumbeat of news announcing cities and states’ new AI policies. It appears those frameworks are the outliers, and that most governments are still in the early stages of creating guidelines for the new technology's use.
Having no clear policy could prevent employees from feeling comfortable experimenting with generative AI, although Robinson suggested that it is initial experimentation that can help drive policy in the first place.
“Once you get into something, you understand where those vulnerabilities are, and it makes you better able to say, ‘This is what we need to impose upon ourselves, and this is what we can anticipate being imposed upon us,’” she said. “Whereas if you haven't messed around at all with generative AI, everything is a risk. You don't know what is possibly going to be asked of you.”
Data privacy and security remain top concerns for public sector employees, SAS found, something that reflects the need for workers to be trained in how to best apply the data AI utilizes. Of less concern to respondents were worries about bias, something Robinson suggested could be because many people are still not truly up to speed on AI.
“Sometimes I think people think, ‘Well, a machine is less biased than the human making a decision,’” she said. “But that's only true if the data that you're feeding into your models isn't riddled with bias. Maybe the average person working in government doesn't recognize yet how much risk there is with AI and bias.”
Robinson noted that understanding of generative AI is growing, and future surveys of the public sector workforce should reflect that growing knowledge.
And while governments may be excited about the technology’s potential cost and efficiency savings, Kim Majerus, Amazon Web Services’ leader of U.S. education, state and local government business, said in a recent interview that it will be crucial to manage individual employees’ expectations of what generative AI is capable of.
“Treat it like a junior analyst,” she said. “When you're looking at how AI is supporting some of these initial projects, treat it like a junior analyst, review what the analyst’s provided. And you're earning trust through the models and you're tuning the models to get to the right answers.”
NEXT STORY: Skills-based hiring smashes through ‘paper ceiling’