by Spencer Pauley
New polling data among registered and likely voters reveals widespread agreement that artificial intelligence could pose a threat to humans as it develops.
Artificial intelligence, or AI, is a technology widely used to perform tasks, learn information, and enable computers to perform tasks that typically require human intelligence. Recently, AI technology has become more sophisticated and increasingly widely used.
However, the latest polling data from Noble Predictive Insights shows that the majority of US citizens do not believe that AI is advanced enough to have an impact on their work or life, but this is likely to change in the future.
Noble Predictive Insights conducted the poll July 8-11 and surveyed registered voters across the United States. The sample included 2,535 respondents, including 1,073 Republicans, 1,221 Democrats and 241 independents. Of the total sample of registered voters, 2,295 were classified as “likely voters.” The margin of error is 2.0%. Center Square Voter Vote Poll is one of six nationally tracking polls in the United States.
When survey participants were asked whether they agreed with the statement “I am afraid that artificial intelligence will take my job,” 55% of registered voters and 54% of likely voters responded differently.
The results for the question, “When I interact with a chatbot like ChatGPT, I can trust that the information I receive is accurate,” indicate that U.S. voters do not trust that AI is fully developed. According to the poll, 51% of likely voters disagree and 34% agree, while 49% of registered voters disagree and 35% agree.
Noble Predictive Insights research director David Byler said the results show people are aware of the short-term risks of hallucinations when it comes to AI, but the current state of AI will not turn the economy upside down.
Despite the common wisdom that voters do not currently feel threatened by AI, the results show there are concerns about the future.
The majority of participants (64–65%) believe that if AI is used carelessly or inappropriately, it could pose a threat to the “survival of the human race.”
The data indicates that around 55% of all survey participants believe that AI could have some impact on their lives.
Byler notes that the question is general, so not all survey respondents believe that the continued development of AI will have a specific negative impact on their lives.
“There’s a portion of about 36% that says, ‘I’m afraid this is going to take my job,’ and there’s a larger portion — more than half — that says, ‘yes, I think this is going to impact my life,’” Byler said in a video interview with The Center Square. “I think the implication is that among the people who think this is going to impact their lives, some can really imagine the negatives and some can really imagine the positives.”
Finally, 55% of voters and 54% of likely voters do not trust AI companies to not share their personal data. In 2022, US President Joe Biden ordered the White House Office of Science and Technology Policy to create a “Draft Bill of Rights for Artificial Intelligence”, a set of five principles to guide the design, operate and implementation of automated systems.
The project aims to protect citizens’ data in the face of the increasingly widespread operate of artificial intelligence.
The five principles identified are: sheltered and effective systems, algorithmic protection against discrimination, data privacy, notice and explanation, and human alternatives.
– – –
Spencer Pauley covers Seattle and King County, Washington. He was previously a freelance filmmaker and worked on “The Clinton Affair,” a documentary series examining the impeachment proceedings of former President Bill Clinton. Pauley currently works as a reporter for Central Square.
Photo by “ChatGPT” Matthew Bertelli.
