A senior technology official from the U.S. Patent and Trademark Office (USPTO) offered three lessons today from the agency’s use of AI technologies to optimize its operations.
“We really needed a way to better share information with our examiners and get through that process to help them” with faster and higher-quality decision making, said Scott Beliveau, branch chief of advanced data analytics at the USPTO, at a NextGov event today where he explained the agency’s drive to adopt the technology.
One of the important lessons learned during the agency’s AI adoption is making a decisive plan for how to use the technology. “You need that vision, that plan of where you want to be,” said Beliveau.
He said the agency started out the process by aiming for “small wins,” which in turn allowed the value of AI tech deployment to be better understood at the operational level at USPTO.
“That was foundational,” he said, to “build that credibility organizationally, to really help us get to that next level.” He added that to get to “higher value or more complicated use cases, you have to build that foundation.”
The second lesson, he said, has been to cultivate support within the agency for innovation.
“You really need to get folks to move along with that risk along with your innovation, your change journey, and your people process in conjunction with the technology,” said Beliveau. “You’ll never be able to scale if you’re the only one. You’re always going to need friends on that particular activity to help you scale your organization activity to the larger enterprise.”
The final lesson he talked about was positioning AI adoption for its benefits to users.
“You want to do good whenever you deliver any sort of AI,” he said. “AI technologies can be viewed in different ways, and you need to view that process” from the end-user’s perspective “because they need to be able to trust the decision,” Beliveau said.