(Left to Right) Kuldip Mohanty, CIO, State of North Dakota; Molly Herrington, Chief People Officer, State of North Dakota; Tammy Miller, Lieutenant Governor, State of North Dakota; Michael Mattmiller, Senior Director State Government Affairs, Microsoft; Taya Spelhaug, Microsoft TechSpark – North Dakota Lead; Jenny Olson, Director of Fargo Campus, Microsoft
The evolution of generative artificial intelligence (AI) in recent months has been outstanding and we are only in the technology’s infancy. However, individuals, companies, and governments need to act fast to educate themselves if they want to be prepared for the shift that is about to happen and to be able to utilize the full potential of the technology. Thankfully, the State of North Dakota began recently taking those proactive measures through collaboration with gener8tor and Microsoft.
The partnership started with a call from Governor Doug Burgum, who has a notable background in the tech industry as the former president of Great Plains Software, to Microsoft with a request to bring out a subject matter expert to educate cabinet members on artificial intelligence in June 2023. The meeting with the cabinet lasted 90 minutes and the cabinet broke out into groups for an ideation session where they asked 4 questions:
- How do you see the state using AI?
- What are some barriers?
- How are we breaking down silos within different agencies around North Dakota?
- How could we make our systems better using AI?
After the meeting with the cabinet members, Chief People Officer for the State of North Dakota, Molly Herrington reached out to Taya Spelhaug, Microsoft TechSpark – North Dakota Lead, and told her that they wanted to do a similar session with the state’s extended leadership team. So, Senior Director of State Government Affairs Michael Mattmiller flew out and gave a 90-minute presentation to that very group.
“We talked about the evolution of AI, the trends that they are seeing or anticipating in government, and responsible use of AI,” Herrington said. “Then, we talked about some of the basic resources that are available. We talked about Copilot and some of the resources that aren’t quite available to us yet in government because of the different security checkpoints we have to go through. We talked about what is coming—the resources that will be available.”
The state enjoyed the programming so much that they wanted to move forward. That’s where gener8tor, one of the nation’s premier startup accelerators, which has a location in Fargo, came into play.
“The state of North Dakota really wanted to lean into generative AI,” Spelhaug said. “While some states are pulling away from it, we see the opportunity to make our state better using this technology. We got a ton of good feedback from people at each session.”
Molly Herrington
Chief People Officer for the State of North Dakota
Photo submitted by the State of North Dakota
Taya Spelhaug
Microsoft TechSpark – North Dakota Lead
Photo submitted by Microsoft
About Career Essentials in Generative AI by Microsoft and Linkedin
Career Essentials in Generative AI by Microsoft and LinkedIn is a six-module course with an estimated completion time of 4 hours and 30 minutes. In the course, you will learn about the history of AI, receive demos on Microsoft Copilot, and learn about different ethical considerations in the realm of generative AI.
“This tool is accessible, free, and doable,” Spelhaug said. “This isn’t something that takes 20 hours worth of studying. We’ve had over 2 million people take the course since it launched.”
Joe Kirgues
Co-Founder, gener8tor
Photo provided by gener8tor
Ben Stanley
Principal, gener8tor
Photo provided by gener8tor
Danie Remmick
Strategic Account Manager, gener8tor
Photo provided by gener8tor
Forming the Cohort
Though the cohort was just recently completed by state employees, the seeds for the program were planted during the 2020 COVID-19 pandemic.
During that time, Microsoft launched opportunity. linkedin.com, a set of 10 free learning opportunities to help people upskill and make job changes while they had extra time during quarantine. Though this was the original purpose of the opportunity.linkedin.com, Microsoft has continued to provide online users access to free upskilling information through this domain long after the height of the pandemic. Along the way, they have continued to update their curriculums and offerings and most recently launched their generative AI skilling known as “Career Essentials in Generative AI by Microsoft and LinkedIn.”
In an additional effort to help people make the most of a difficult situation in 2020, Microsoft Philanthropies via the TechSpark program teamed up with gener8tor to help people go from unemployed to employed in the top 10 most in-demand jobs in the country.
About opportunity.linkedin.com
Microsoft launched opportunity.linkedin.com in 2020 and opened 10 job paths during the pandemic to help people upskill. However, the program has continued to develop over time and the generative AI course is now part of it.
About Gener8tor
gener8tor is a national startup accelerator that offers critical resources like funding, mentorship, and networking opportunities.
We’ve had a long-standing relationship with TechSpark where we worked with them to pilot this program,” said gener8tor Co-Founder Joe Kirgues. “We started in northeast Wisconsin and we have been expanding it across the country and have been focused on helping people who are predominantly underemployed gain new skills through these programs. Throughout this process, we’ve worked with these people much like we work with a startup. But, instead of working with them on things like a business plan, or a pitch deck, or an executive summary, we would do mock interviews and resume prep on top of the hard skills.”
That partnership developed further when Spelhaug approached Kirgues at gener8tor’s conference around innovation in education and workforce and said she had been working with the Governor’s and Lieutenant Governor’s office on how to skill on generative AI and that she wanted to know if gener8tor would be open to a partnership where they would adapt their upskilling program to serve state employees with generative AI.
The gener8tor team accepted the challenge and came up with a curriculum, the first of its kind for a state government, in just 6-8 weeks using “Career Essentials in Generative AI by Microsoft and LinkedIn” as the cohort’s textbook, if you will. However, there was plenty more to the cohort than just the LinkedIn Learning modules, which can be found for free online.
“At gener8tor, In the work that we do with startups, we often help the leaders of these companies come up with creative ways to solve problems using new technologies,” said Ben Stanley, Principal, gener8tor 1889. “We thought it was a very fascinating and fun challenge to take that same approach toward helping leaders in state government find ways to solve problems using artificial intelligence.”
The Cohort
In the program, which originally was supposed to have 40 or so participants from agencies around the state but instead had 67 because of significant interest, participants met weekly with program directors like Danie Remmick, an educator at heart, to discuss the previous week’s content, ask questions, and prepare for the content in the week ahead. They also had the opportunity to take in weekly Lunch and Learns from subject matter experts. And, perhaps most importantly, each person, towards the beginning of the cohort, had to bring and pitch an idea on how the state could improve its workflows to get more done or on how the state could leverage AI to provide better experiences for citizens of the state.
“Participants were not merely voluntold; they willingly engaged,” Remmick said. “The decision to increase the number of available spots aimed to accommodate diverse interests across various state agencies. Each organization sent a representative, fostering collaboration among individuals who typically operate independently. This crossagency interaction revealed insights into departmental workflows, intersections, and potential synergies. Beyond learning about generative AI, the program sparked discussions on operational collaboration and practical implementation to benefit other agencies. Truly, a beautiful outcome emerged from this cohort.”
AI Already in Action
“We want to protect our citizens’ data. We’ve got a huge commitment to do that. It’s a moral obligation that we have to do that. And thanks to the adoption of AI that we’re doing through our NDIT team, we’ve dramatically increased our Cybersecurity Operations Center. It’s one of the most sophisticated in the country. They’re responding, this last year in 2023, to 4 billion attacks… automatically. We can’t take care of all of those [automatically]—50,000 of those had to be handled with some element of manual interaction. But AI helps us speed through all of those, again, to be great stewards of the citizens’ data, and to make sure that our team members and everybody we’re working with, that we can make sure that we secure the privacy and the security and integrity of that data.”
-North Dakota Governor Doug Burgum
By the Number
- 67 cohort participants
- 938 hours invested in learning
- 335 LinkedIn Learning courses completed through the cohort
After the original ideas were pitched, poked, and prodded by the rest of the cohort’s participants, 12 different teams were formed to work together and explore ideas for potential use cases further. Towards the end of the cohort, their capstone was to present their finalized ideas, and “10 to 11” unique ideas came forward.
“Honestly, a lot of the ideas that were coming forward were related to simplifying the user experience,” Herrington said. “Believe it or not, if you want to come to the State of North Dakota and get a service, you’re going to have to go to the website of that unique agency and create an account there and then you would have to go to the next agency that you need to service from and create an account there. We’re currently working on how we can create one portal for that… All the ideas brought forward are now sitting with Kuldip Mohanty, the State of North Dakota’s Chief Information Officer, and Kim Weis, the State of North Dakota’s Chief Data Officer. They are trying to figure out what we can act on now. They’re also looking at what the bigger items are that need to be flushed out and carried forward on their own as different initiatives.”
Ethics First
“When ChatGPT really came into public awareness, one of the first things that was recognized was a need for some policy around that and some guardrails so that our team members knew how to appropriately, responsibly, and ethically use some of this. They needed to know what they could use and where there were security risks posed. So, we worked with a cross-functional group to develop AI policy and supporting guidelines for it—specifically for the executive branch agencies.
-Kim Weis, State of North Dakota’s Chief Data Officer
You can learn more about the state’s guidelines by scanning the QR code.
Kuldip Mohanty
Chief Information Officer for the State of North Dakota
Photo provided by the State of North Dakota
Kim Weis
Chief Data Officer for the State of North Dakota
Photo provided by the State of North Dakota
The Weekly Lunch and Learns
One of the powerful aspects of the generative AI cohort for the state of North Dakota was the opportunity to learn from subject matter experts working in the field.
- Tuesday, October 17 – Generative AI Ethics and Government Use – Mark Muro, Senior Fellow at Brookings Metro and Jason Haislmaier, Partner at Bryan Cave Leighton Paisner Law Firm »
- Tuesday, October 24 – Generative AI and North Dakota Startups/Research – Jimmy Slagle, CEO and Founder, GenerAI; Jon Cabak, CEO, Detect Auto; Karthik Balaji, CEO, Aethero; and Scott Meyer, CEO, Chipp.AI
- Tuesday, October 31 – Demonstrating ND Leadership in AI – Michael Mattmiller, Senior Director of Government Affairs, Microsoft
The Ideas
“There were four or five shuttle-ready use cases which are very appealing to the entire review committee,” Mohanty said. “The one that we’re very keenly focused on exploring is making it possible to go to the state of North Dakota website and have an interface that allows you to comb through the treasure trove of information across the state’s multiple websites through a simple request or prompt. Our goal is to create a frictionless experience for our citizens to interact with the government.”
An example of this would be going to the Governor’s Office website and asking the large language model, “How do I start a business?” Currently, you might have to go through five or six clicks until you get to the right information. Or, there are instances when trying to comply with a tax requirement that you could find yourself going through 50 clicks to get the information that you want.
“We are exploring how to do that at the moment and Kim and her team are leading that effort,” Mohanty said. “At the moment, we’re doing it at a very small scale and limiting it to a certain cross-section of agency users that can exhibit the behavioral pattern. Then, the next step is to understand the cost basis that is needed because everything with chatbots is all based on token pricing. Currently, we don’t have a pool of money sitting around for this. So, we have to get that appropriation so we can start to build on that.”
“Another one of the use cases that came out of the cohort centered around searching our statutes,” Weis said. “Our citizens, at times, have struggles looking up laws and finding answers for where they can find certain statutes. This solution is very similar to the first in that it’s essentially a beefed-up search engine where people can use natural language to query the large language model and find the statue—they can ask questions that will point them to where they want to go.”
“The third use case that is shuttle-ready revolves around the legislative session and the plethora of information that comes out of the legislative session and how challenging it is to stay informed on all of it because there’s so much data that comes out,” Weis said. “So, whether it’s bills that are written, whether it’s testimony and videos, or agendas, it’s about figuring out how state employees as well as citizens can search for those things during a legislative session. It’s really challenging to do that. This third potential use case is a larger project because it also includes the component of the transcription of the videos from testimony, committee meetings, and floor hearings.”
Those ideas are just the beginning.
The State of North Dakota is leading the way as one of the first states to embrace artificial intelligence and is the first to complete the gener8tor cohort—because it was designed for them. What the future holds, remains to be seen.
In the meantime, gener8tor would love for more states to adopt the forward-thinking approach championed by Gov. Burgum & Lt. Gov Miller.
“One of our guest speakers during the program, Mark Muro of the Brookings Institute, studies the interplay between technology and people in different regions,” Kirgues said. “He spoke about how in past technological revolutions, heartland communities in the United States have historically been left behind. With the emergence of AI, that could change. Communities in the heartland have an opportunity to be early adopters and embrace this powerful new tool to solve local and regional problems. We hope more states in the middle of the country begin adopting the same mindset as North Dakota.”