Americans are increasingly concerned about the rapid advancement of artificial intelligence (AI), yet they remain divided over the nation’s approach to AI development, particularly in relation to China. A recent POLITICO poll highlights this division, revealing deep-seated anxieties about the existential risks posed by AI while also showing a strong desire to maintain technological supremacy on the global stage.

The poll indicates that while 63% of Americans acknowledge the potential for AI to endanger humanity, opinions are split on the risks associated with slowing down its development to prioritize safety. Approximately 43% of respondents believe it is acceptable for the U.S. to lag behind other countries if it means ensuring more secure and responsible AI development. In contrast, 37% argue that the U.S. should continue its rapid development pace to avoid falling behind international competitors, notably China.

This polarization is not just among the public but also evident in Congress, where AI safety has become a central topic of discussion. Many Republican lawmakers argue that pausing AI development could cede ground to China, potentially compromising U.S. leadership in the tech sector. Conversely, Democrats emphasize the dangers of unfettered competition, advocating for stringent regulations to mitigate risks.

"The AI race is not an arms race that you win," stated Rep. Greg Casar (D-Texas) at the POLITICO Decoded Summit. Casar, along with Sen. Bernie Sanders (I-Vt.), has called for a ban on superintelligence AI to prevent possible existential threats—a concern underscored by alarming warnings from AI researchers about the potential for AI to cause catastrophic outcomes.

Senior Republicans, however, view these proposals with skepticism. Rep. Brett Guthrie (R-Ky.), chair of the House committee overseeing tech, emphasized at the summit that halting AI development would be akin to surrendering to Chinese dominance. This sentiment resonates with former President Donald Trump, who has dismissed AI safety concerns as a "HOAX" and criticized federal regulations as a gateway for China to surpass the U.S.

Amid these tensions, some lawmakers strive for a balanced approach. Rep. John Moolenaar (R-Mich.) acknowledges the need for public safety while also recognizing the importance of staying ahead in the AI race for economic and national security reasons. Similarly, Rep. Josh Gottheimer (D-N.J.) pointed out that while the U.S. cannot afford to exit the race, it must proceed cautiously and responsibly.

The debate extends to international diplomacy, with sharp differences on whether Trump should discuss AI safety cooperation with Chinese President Xi Jinping. While some Republicans trust Trump's negotiation skills, Democrats like Rep. Ted Lieu (D-Calif.) suggest using the talks to set basic safety standards, such as commitments against AI-controlled nuclear launches.

As these discussions unfold, the challenge for the U.S. remains clear: balancing the imperative to lead in AI innovation with the necessity to ensure it does not compromise global safety.