Artificial intelligence in conversation has improved and scaled rapidly since chatbots first entered social media in 2016. The first iterations of chatbots on Facebook Messenger were simple, allowing for restaurant reservations, flower deliveries, and other structured calls to action. Now, roughly a term later as US President, conversational experiences are becoming increasingly intuitive. The AI technologies behind it can cope with additional individual complexity, contextualize language more easily and simulate human reality better – even when it comes to politics.
Amplify.ai, an enterprise-level conversational AI platform, helped senatorial campaigns drive engagement with local constituents in 2020. It uses natural language processing (NLP) and machine learning to link existing social media pages, analyze public sentiment and intent, and answer individual questions through human-like interactions with AI chatbots.
Several Democratic candidate campaigns have incorporated the platform into their existing digital strategy, although Mahi de Silva, CEO of Amplify.ai, told VentureBeat that contractual obligations prevented him from sharing a full list of clients. However, elected Senators Mark Kelly (D-AZ) and John Hickenlooper (D-CO) have publicly partnered with the company. Jaime Harrison (D-SC), vice chairman of the Democratic National Committee and youngest challenger to incumbent Senator Lindsey Graham (R-SC), has done the same.
Mark Kelly was not a typical candidate: before running for office, he served as captain in the United States Navy, completed missions as a NASA astronaut, and formed a political action committee with his wife, former Senator Gabrielle Giffords. The choice also turned out to be atypical. After John McCain passed away in 2018, his Senate seat was occupied by two different Republican Republicans from Arizona – Jon Kyl, then Martha McSally – within two years. Kelly challenged incumbent McSally for McCain’s remaining tenure. Kelly’s campaign also had to consider factors like COVID-19 restrictions and Arizona’s history as a swing state.
In a statement to VentureBeat, Justin Jenkins, the digital director for Kelly’s campaign, commented on his team’s adoption of conversational AI. “When the pandemic broke out, the campaign quickly began exploring new and creative ways to repeat the face-to-face conversations we traditionally had at the doors. We decided to test Amplify’s conversational AI as it can scale and customize the user experience based on user history with the campaign. “
Kelly’s campaign couldn’t risk spreading COVID-19 through visits to constituents. In some ways, however, the accessibility and ease of personalization of chatbots are akin to the door-to-door personal acquisition that pre-pandemic campaigns relied on for voter education and engagement.
Older chatbots were mainly based on strict entrances and exits. For example, if a user typed “what is the capital of Arizona” into a bot on Facebook Messenger four years ago, the bot might have replied “phoenix”. The conversational AI used in the last election goes even further, working to interpret the user’s intent or the various expressions people can use to ask about a topic. Then an attempt is made to put together a helpful, friendly, and relevant answer – and to reflect the back and forth exchange of a human conversation.
As with a home visit, individual messages from a chatbot can more closely connect users to a candidate’s platform and enable a campaign to recruit them as donors, volunteers, and voters. The Mark Kelly campaign reported that it interacted with over 180,000 voters on Facebook Messenger in the first month of its conversational AI program.
De Silva distinguishes these chatbot conversations from those with virtual voice assistants such as Siri or Alexa, where the user receives precise information that the product has amassed from a database. In an interview with VentureBeat, de Silva said that AI messaging “creates a conversation between consumers or citizens and brand organizations … so it’s very dynamic and doesn’t try to put all resources in one system.” This dynamic is amplified by machine learning that tracks user behavior.
For example, the platform could decode a comment based on the model of “Mark Kelly Rocks” on the campaign’s public Facebook page and autonomously access the poster in Messenger. The chatbot thanks the user for participating in Kelly’s campaign and leans towards the positive vibe by asking if he is open to conversation. When the platform analyzes a less positive comment, it may show interest in understanding different points of view.
Amplify.ai also tabulates comments and reactions that are transferred to a campaign’s Facebook account. The platform then performs sentiment and intent analysis for each data point and visualizes it on a dashboard so that team members can closely follow audience interactions over the course of a campaign. “You can imagine that a smiley face is pretty easy to find and associate positive feelings with,” said de Silva. “But when you get a comment, we actually have to process it in the context of what the post was trying to achieve.”
In addition to the insights campaign reps gained from analyzing Amplify.ai, the conversational AI can interact with people at a speed and size that human teams cannot. For example, if a campaign received over 100,000 written commitments in a month, it would result in over 3,000 unique messages or comments per day that would require at least six full-time volunteers or staff to respond to an average of a message in a minute. The right AI could theoretically reduce and manage this task while engaging constituents and inspiring them to volunteer, vote, and donate. According to de Silva, Amplify.ai has signed over 10 billion customer contracts with over 500 million consumers since its launch.
Conversational AI is likely to increase in intelligence, credibility, adoption and speed of delivery in the coming years. Startups like Hyro, Pypestream, and Orbita are also working to provide companies with conversational AI solutions for customer loyalty. Hyro’s customers include government agencies, and Marketing Director Aaron Bours told VentureBeat, “If AI is smart, fast, and human enough, you can actually discuss important issues [such as] Tax programs or foreign policy. “