Over the last couple of months, I’ve asked several government affairs leaders how they’re incorporating artificial intelligence into their work and I mostly received blank stares. A few answered that their teams were just starting to get smart and think about it.
I want to offer a three-phase framework to think about how artificial intelligence will impact your teams and the government affairs industry.
Phase 1: Content Generation
We are currently in phase one, where it is easier than ever to create content using ChatGPT and other consumer-accessible large language models. One of the leading public affairs firms has recommended that each one of their associates start client projects using ChatGPT. I heard a story from another professional about how they used ChatGPT to write an introduction script for an NGO event they were sponsoring that they found quite good.
The most memorable use-case came from a round table with heads of communications where one Fortune 100 leader shared she had an underperforming team member who one day sent around a press release that was significantly better than prior work she had done. When asked what she did differently, the junior staffer responded, “Well, I started with ChatGPT.”
The AI era is here and your teams have likely explored or are actively using tools like ChatGPT. At Quorum, I found our managers using it to help them write their performance reviews. As a leader, you should be thinking about what training and guidance you need to give your teams about how and when they can use artificial intelligence and large language models. And then, you should be providing and encouraging education about the need to check the information for accuracy, or you are going to end up like the lawyer who wrote a factually incorrect brief.
When it comes to policymaking, there are two large impacts of AI-generated content.
First is the possibility of generating unique grassroots advocacy messages every time an employee, advocate, or association member reaches out to Capitol Hill. The days of generic letters will soon be over. Interestingly, a Brookings study found that members of Congress can’t tell the difference between AI-generated and constituent-generated communications. At Quorum, we have released technology that rotates each message based on a series of template options that you share with advocates — but we make sure that every message is sent by a real human,
Second, it is now possible to create unique position papers and impact reports for every legislator. Imagine being able to customize an impact report 435 different times and highlight the things most relevant to a legislator. This previously required a large amount of effort. Today, it is going to become much easier and much faster.
Phase 2: Smarter Software
The second phase of the artificial intelligence revolution is smarter software where you are going to start to see artificial intelligence incorporated into the software you use on a daily basis; whether that be Outlook, Google Docs, or Quorum. The goal is to save you and your team significant time with their work and make the process of using AI as easy as possible.
Here are a couple of solutions we have launched:
Faster Searches. Instead of clicking through 3-5 filters to find all statements by House Republicans mentioning your organization last year, you’ll be able to simply type that into a search bar and find the answer. Then you can also take it to the next level by having it run a comparison between this year and last year.
More Actionable Insights. Finally, AI is going to lead to a greater ability to create meaningful reports and dashboards. Rather than having to customize everything, you’ll be able to tell AI what you are looking for and have it pop up, leading to faster insights.
But we’re not done. Another feature we’re working on is seamless interaction logging. We will get to the point where you don’t have to fill out a form in a mobile or desktop app to log an interaction with a member of Congress, but instead, you’ll be able to write a text summary of who you met with and when and AI will put that into your CRM system.
Now what is the impact on your teams? Well, two things. First, it will be easier than ever to bring metrics and reporting into your work. And second, your team is going to have more time to spend doing what matters — talking with legislators.
Phase 3: Government Relations Large Language Models
The third phase will be government relations large language models that function like ChatGPT but are powered with legislative information (bills, press releases, tweets, floor statements, committee testimony, etc.). The benefit is that the large language models will have more information than ChatGPT currently has and it will be more accurate as it will be pulling from official data sources rather than the broader internet. The result will be that every government affairs team will feel like they have an in-house legislative researcher on their team.
Using a chat interface, you’ll be able to ask questions like, “How have members of the Blue Dog Democrats changed their dialogue on a given issue over the last year.” Or, “How has Senator Barrasso changed his position on access to public lands for hunting over the years?” It will make staying on top of the tens of thousands of documents that come off Capitol Hill and state capitals every day much easier.
Furthermore, you’ll be able to ask these government affairs large language models for summaries of legislation or how an issue is being considered on Capitol Hill or in a state legislature. As the model gets smarter, it will be able to help with legislative tracking by looking at all the bills you tracked in previous years and suggesting other bills that are similar in what they accomplish for you to review the same way a human would.
Finally, the large language model will get to know all of the statements, bills, and votes from each legislator and be able to provide recommendations on the best way to position an issue in approaching them.
Will AI replace government affairs?
I was standing in the lunch line at the Public Affairs Council Digital Media and Advocacy Summit and one nervous staffer shared with me her fears that AI is going to take her job. This is something that organizations need to address with their teams as it is going to be on people’s minds and may result in a slower adoption of the technology.
At its core, the government affairs industry is about human-to-human relationships and that is something that AI is never going to be able to replace. As this audience well knows, it is the relationships that change lawmakers’ minds on a policy issue, get text added to a bill, or collect valuable intelligence.
You are going to need to encourage your teams to focus even more on relationships as the office work of government relations will become faster and more accessible.