Conversational Technologies
5 min read

Conversational AI’s Role in the 2020 U.S. Presidential Election and Beyond

Aaron Bours VP Marketing, Hyro
Conversational AI’s Role in the 2020 U.S. Presidential Election and Beyond

In 48 hours what is without a doubt one of the most pivotal moments in a drastically pivotal year will take place; the 2020 U.S. Presidential Election.

 

In full disclosure, politics isn’t really our thing. We’re less stir-the-pot, and more simplify-access-to-information. Still, these elections serve as another prime example of the ways in which conversational AI plays a key role in some of the more crucial junctures of our lives; where there are vast amounts of (often confusing) information to be delivered to millions of people, there are conversational AI solutions.

 

Earlier this year, Google and Amazon launched new features for their respective virtual assistants (Google Assistant and Alexa) to help voters locate the nearest polling stations, answer FAQs, and deliver real-time official, vetted updates.

Source: Apple

In May, as COVID-19 forced businesses and public services to close, Idaho held its first mail-in only election. Expecting a high volume of calls to their support center from puzzled voters, the Secretary of State’s department deployed a chatbot on their website. The chatbot fielded common questions about voting-related issues, including upcoming deadlines and typical election procedures, such as how to request a ballot.

 

Crowning the solution a success, Chad Houck, Chief Deputy Secretary of State in Idaho, estimated that the bot took about 3,000 calls away from the department’s call center over the course of 30 days, saving time and effort for the department’s staff members and enabling them to focus on more complex tasks.

 

Scratching the Surface

 

Voter support is just the tip of the iceberg when it comes to conversational AI for government agencies. As critical as this use case is in the age of fake news and misinformation, elections are isolated events. In contrast, the many other use cases conversational AI fulfills for government agencies and their citizens are intrinsic to our daily routines.

 

Of course, using the term routine has a somewhat ironic ring to it, as the past eleven months have been anything but routine. The overwhelming spread of COVID-19 across the nation compelled government agencies of all sizes to respond creatively to a crisis that has affected every one of their constituents.

 

In NASCIO’s 2020 State CIO Survey, 76% of respondents (state CIOs) said that chatbots (virtual agents) for citizen service inquiries were the first automation solution and emerging technology they introduced in response to COVID-19.

In fact, the implementation of conversational AI across state governments has been so widespread that at the time of writing, 75% of all U.S. states have deployed AI-powered virtual assistants to alleviate the stress on their support centers and to better serve their residents. 

 

In an August poll by the State Policy Network, one in five Americans said they are confused about COVID-19, 58% are confused by messages from the U.S. government, and 46% feel that their sense of social order has worsened. To ease their qualms about the situation, Americans have turned in troves to their trusted government agencies’ call centers. Throughout the pandemic, call-volume to the CDC’s support center increased 100-fold. 

 

By placing conversational AI on their digital front doors, government agencies have found incredible success in funneling the uptick in volume to their support centers to qualified AI-powered agents that are providing relevant, timely information to concerned citizens. 

 

Operating around the clock and sourcing frequently updated information, conversational AI solutions are not only a highly credible source of truth but a place people can turn to complete a wide variety of tasks. Take Fairfax County, Virginia, for example. This urban county on the outskirts of Washington, D.C., with a sizable population of 1.2 million, features a chatbot on its home page that enables residents to pay taxes, register to vote, apply for various licenses, and even peruse the county’s yearly budget. 

Residents Are Standing In Line To Vote In Fairfax County, Virginia, September 2020. Source: Wusa9

Many other counties are following Fairfax’s lead. In Government Technology’s 2020 Digital Counties Survey, about 25% of respondents (county officials) said they were already using chatbots, and another 39% said they had plans to start using them in the next 18 months. The remaining 36% said they had no plans to start using chatbots.

A Self-Imposed Roadblock

 

So what is slowing down these 36% of county officials who state they have no plans of implementing conversational AI? There are several explanations for this phenomenon, but it’s safe to say they’re all rooted in preconceived notions about conversational AI’s actual capabilities, cost, and deployment. 

 

There’s a school of thought among CIOs and digital officers that chatbots are inevitably limited in their scopes, demand hundreds of staff-hours, and ultimately offer meager Return On Investment (ROI). This fear is entirely justified, given the complexity inherent in designing an adequate and useful conversational flow using DIY chatbot builders and accurate in that these types of flows require ongoing maintenance.

 

However, some conversational AI companies have made giant strides in advancing the quality of these types of solutions over the past few years. Breaking away from the If “x,” reply “y” formula that most chatbots still rely upon, there are solutions out there that harness their conversational abilities through more advanced techniques. The fusion of conversational AI with Machine Learning (ML), Natural Language Understanding (NLU), and computational linguistics has given birth to virtual assistants that provide robust “knowledge-based” conversations as opposed to limited “intent-based” flows. Better yet, some of these solutions automatically scrape their users’ data and web pages daily, eliminating the need for manual updates. 

Hyro's Knowledge Graph Approach Vs. Traditional Chatbot Solutions

Conversational Government

 

Full-blown adoption of conversational AI across all government agencies is not a question of “if” as much as it is a question of “when.” According to NASCIO, all 75% of states who have implemented conversational AI on their digital channels intend to continue using it even after the pandemic subsides. 

 

As this technology continues to trickle down from the state level to the county level, it’s highly likely that citizens will shortly handle most of their interactions with their government agencies, from voting to filing taxes to applying for social security benefits, through a conversational AI virtual assistant. 

About the author
Aaron Bours VP Marketing, Hyro

Aaron is Hyro’s VP Marketing and a conversational AI expert with almost a decade of experience under his belt working on next-gen natural language-enabled technology, including Google Duplex. Aaron is a former New Yorker who now spends his days casually knowing where all the coolest spots in Tel Aviv are without ever really trying.