Back in May of this year, I read an article in Wired about India’s seeming embrace of using artificial intelligence in their political and election processes.
AI-generated deepfakes of politicians were actually sanctioned by national parties, even though officials and the technologists creating the images and audio admitted a large majority of constituents probably didn’t know they were interacting with fake posts.
The good: Gathering info and sparking interest
As it relates to US elections, there does seem to be a general consensus to allow artificial intelligence to do what AI does best, which is analyzing large amounts of data.
Numerous articles, including those from the Brookings Institute and the Ash Center for Democratic Governance and Innovation at Harvard University, extol the benefits of AI in performing functions like identifying anomalies in voter lists, efficiently scanning paper ballots, and interpreting and enhancing poll results.
Asked by Love Now Media if he thought AI had any place in government, Isaiah from Cheltenham agreed, saying, “I think that maybe we can use AI as a tool when it comes to information gathering.”
Megan from Center City also believes there could be a place for artificial intelligence in elections. “I think there’s a place for AI…it could be used to bring up information that has been forgotten or help process some data.”
Darian from Southwest Philadelphia was a big proponent of using AI in elections. He believes one way to make use of artificial intelligence is in ensuring everyone’s voice is heard.
“AI could be used in surveys. You could have surveys and use the AI to get the answers to [reach] more people. I feel like people’s word isn’t being heard nowadays, and with AI it could be better.” He even went so far as to say that “maybe if AI is introduced [in elections], it would be a cool addition, and everyone would be more interested [in voting].”
The use of artificial intelligence in elections has also been touted as technology that could level the political playing field by lowering the cost of running campaigns.
By analyzing voter demographics data, campaigns can better target potential supporters to maximize advertising spending or monitor social media and other platforms to get real-time feedback on campaign performance. And even with all the potential for its misuse, AI can still be used to provide some level of election security by analyzing patterns to detect irregularities in voter registrations and electronic voting machines.
Treasure, an interviewee from Northeast Philadelphia, doesn’t think AI has much of a place in government, but still believes artificial intelligence could “maybe correct data on the counts about who voted.”
The bad: Spreading misinformation and suppressing votes
While even some skeptics can see a use case for how AI could streamline our electoral process, there are still lots of other things to consider.
There are no shortage of ways in which artificial intelligence could be used to deceive voters by spreading misinformation and disinformation, with examples already popping up in this year’s election cycle. Misleading AI-generated robocalls targeting then-candidate Joe Biden in the New Hampshire primary made national headlines, while the Trump campaign came under fire when Donald Trump reposted fake images implying he had Taylor Swift’s endorsement for president.
Voters are hearing rumors through various platforms and sharing the information, like Massara from Southwest Philadelphia who shared that she has “seen some pictures…of past presidents…meddling with kids and stuff like that. It wasn’t real, but it fed into the media like it was real.”
The ability of misinformation to spread quickly among voters of all demographics is a legitimate concern that this country doesn’t yet seem to have an answer for.
Using AI as a tool in voter suppression is one of the biggest threats this country will face in the upcoming presidential race.
In addition to the New Hampshire robocalls that attempted to trick voters into not voting in the primary election, groups like the Conservative Partnership Institute have been deploying AI systems to perform what have been called ‘faulty and error-prone’ analysis on demographic data in an attempt to file mass voter challenges, potentially removing thousands of people from registration rolls. Tactics like these typically disproportionately affect voters of color, voters from lower-income communities, and voters with disabilities as well.
There are also examples of how artificial intelligence negatively affects individual candidates and drives election outcomes in local races. Consider the case of Adrian Perkins, who in 2022 was running for reelection as mayor of Shreveport, Louisiana. A TV commercial funded by a rival political action committee used AI to superimpose Perkins’ face onto an actor’s body. He believes the effect of that negative ad had a significant impact on his losing the election. And earlier this year in Philadelphia, Sheriff Rochelle Bilal’s campaign was exposed for fake news stories posted to her campaign website that had been generated by ChatGPT.
As to the question of whether AI has a place in government, Abby from South Philadelphia was clear. “No,” Abby said. “It’s a people’s government.”
Here are some tips from MIT’s Media Lab to help you spot deepfake photos and videos, and you can take a quiz to test your AI-detection skills here.
As it relates to misinformation in written articles, here are a few more tips.
- Consider the source — Is the information coming from a reputable source
- Check the date — Just because this may be the first time you’re seeing an article doesn’t necessarily mean it’s new or current.
- Check other sources — Always confirm information with multiple legitimate sources online.
- Check the location — Is the photo you’re looking at a picture that accurately reflects what you’re reading?
- Check your emotions — Misinformation is usually deliberately inflammatory and relies on elevated feelings of anger or outrage to spread. When it comes to anything you see on the internet, a good rule of thumb is “Verify, then share.”
And the unexpected: Support for AI candidates?
In a surprising twist, when Love Now Media asked interviewees if they would trust an AI-generated elected official, 2 of 7 respondents actually said yes.
“Seeing as though AI is computer technology,” Faheemah from Germantown told the Love Now Media interviewers. “I feel like they are smarter than most people so maybe so.”
Darian, who had suggested using surveys to gather voter sentiment, was also supportive. “I think pretty much, most of the time I would [trust an AI elected official],” they said. “It depends on how good it is and how it’s used.”
That reality may be coming sooner than you might have thought. In the UK, an AI chatbot recently ran for parliament.
This summer, AI Steve was listed on the ballot under the independent SmarterUK party. Entrepreneur Steve Endacott created the chatbot as a listening agent, with the plan for AI Steve to conduct conversations with voters to understand the policies they care about.
The human Steve said he planned only to only serve as proxy, someone who would physically cast votes, answering based on the uncovered during AI Steve’s interactions with constituents. The advantage of an AI politician, Endacott said, is “its ability to increase efficiency and transparency in politics by having conversations with voters 24/7, then analyzing and summarizing these conversations so the party can form policies voters actually care about.”
Spoiler: AI Steve didn’t win, only garnering 179 votes overall.
But do we really need artificial intelligence and all the technology that goes with it to effectively represent ourselves in government?
The final question in the Love Now Media interviews asked respondents what they thought elected officials could do to spread more love in the community. We’ll let the voices of the people be the final word.
“Just being authentic and trying to understand people, reach out to people and being more in touch than they are currently.” — Isaiah
“Come down from the offices. Connect more with people of color, people in general. I feel like in today’s society we lack trust within people in politics. They aren’t even listening to people’s voices anymore it’s just like ‘vote for me, vote for me’… but they aren’t actually taking the time to consider how people are trying to change the city and change where they’re living…they need to come out and engage with actual living people…” — Massara
“I think they [politicians] can show people what they’re passionate about. Not just tell people what they want to hear. But our officials should genuinely care about their citizens. And, you know, just make people feel comfortable and accepted for who they truly are.” — Megan
This is the third in a series of essays produced by Love Now Media and Technical.ly exploring the impact of artificial intelligence on various aspects of life in Philly. Read the first essay on AI and personal safety.
The series is produced with the support of the Philadelphia Journalism Collaborative, a coalition of more than 25 local newsrooms doing solutions reporting on things that affect daily life where the problem and symptoms are obvious, but what’s driving them isn’t. Follow at @PHLJournoCollab.
