A new poll found that about six in 10 Americans view artificial intelligence (AI) as a threat to human civilization.

The Reuters/Ipsos poll published Wednesday found that 61 percent of Americans believed that AI is a threat to humanity’s future, while about 22 percent reported that they disagreed and 17 percent said they were not sure.

Those who voted for former President Trump in 2020 were more likely to be concerned about the threat of AI, with about 70 percent of his voters agreeing that AI could threaten humanity compared to 60 percent of President Biden’s voters agreeing.

Evangelical Christians were also more likely than non-Evangelical Christians to report higher concerns about AI, with 32 percent of Evangelical Christians saying that they “strongly agree” with humankind being threatened by AI. Non-Evangelical Christians who reported that they “strongly agree” stood at 24 percent.

The rise of AI programs, like ChatGPT, has prompted concerns among lawmakers and experts about the potential negative effects of the emerging technology. Tesla CEO Elon Musk and a group of technology experts called for a six-month pause in advanced AI development earlier this year, warning that the technology could pose “risks to society.”

Musk also told CNBC’s David Faber in a new interview Tuesday that there was a chance that AI “goes wrong and destroys humanity.” He called AI a “double-edged sword,” noting that it is difficult to predict where the technology could go next.

The Reuters/Ipsos poll included 4,415 U.S. adults and was conducted between May 9-15. It has a credibility interval of plus or minus two percentage points.