A majority of Americans are worried about AI’s progress as tech industry leaders and lawmakers call for its development to slow, according to a poll released Thursday.

Three quarters of respondents to CNN’s new survey reported some level of “fear and concern” when it came to “growth and development of AI technology.” The remaining 25 percent of those polled instead said they felt “hope and excitement” surrounding its development.

Out of poll participants who evaluated AI’s growth positively, 7 percent felt “strongly” on the topic, while 18 percent were “somewhat” enthusiastic. In contrast, 43 percent of those fearful of the emerging technology had strong opinions on it.

The poll comes amid broader political discussion around the limits of AI regulation, as concerns mount surrounding its environmental and data privacy implications. While Sen. Bernie Sanders (I-Vt.) and Rep. Greg Casar (D-Texas) float legislation seeking to ban artificial “superintelligence,” President Trump and his allies have embraced the technology.

Over the weekend, Trump announced plans for the creation of an “AI Force” and the appointment of a “czar” to oversee the technology’s development, though he has rejected calls for federal regulations.

Trump, in a Truth Social post, referred to concerns about a lack of federal AI guardrails as a “hoax,” saying he would not make space for the technology’s “decimation” as its U.S.-based industry leaders compete with other countries.

“AI is the next Industrial Revolution, or Internet, but will be even larger and more impactful, possibly as much as 25% of our Country’s GDP [gross domestic product]. We are leading China, and the rest of the World, and I intend to keep it that way!” he wrote on Saturday.

In the CNN poll, 71 percent said the federal government wasn’t putting enough effort into AI regulation, while 10 percent said lawmakers were doing too much to address the topic. Eighteen percent described their AI regulation efforts as sufficient.

The CNN poll took place from Sept. 16-17, featuring 1,206 respondents and a margin of sampling error of plus or minus 3.5 percent.