Pope Leo XIV Criticizes Jensen Huang's Stance Against Government Regulation of AI

Pope Leo XIV warned of the potential dangers of artificial intelligence and criticized tech company CEOs who oppose restrictions on AI technology.

Speaking to reporters on the flight back to Rome on Monday after concluding a four-day visit to France, the Pope dismissed claims that these concerns are "fake news." Notably, former President Donald Trump had previously dismissed AI safety warnings as a "hoax."

He specifically criticized tech CEOs who advocate for unregulated AI, stating that concerns raised by AI experts "should be taken seriously." Pope Leo said he had read that Nvidia CEO Jensen Huang had suggested setting "guardrails" for certain AI models, but expressed skepticism, noting that the CEO had recently stated there should be no limits or government regulation.

"Yet, it is the same person who said there should be no limits and no government regulation," he said.

Pope Leo urged global leaders, AI experts, and civil society organizations to work together to examine the current state and potential future of AI. He emphasized the need for dialogue, research, and investigation to prevent AI from becoming a threat to humanity.

"We cannot sit back and pretend nothing will happen," the Pope warned, urging the world to take a proactive approach to AI.

Nvidia did not immediately respond to a request for comment.

Pope Leo's Past AI Warnings

Last Friday, Pope Leo XIV warned that the growing influence of AI and other technologies could erode human nature, calling for greater education in ethical discernment. During meetings with French officials, including President Emmanuel Macron, he urged society to strengthen moral decision-making capabilities as AI becomes increasingly integrated into daily life.

In November, Leo urged AI developers to prioritize ethics and spirituality, stating that technological innovation carries "ethical and spiritual weight" because design choices reflect a vision of humanity.

Jensen Huang Backs Faster AI Development with Stronger Safety Measures

This development follows recent remarks by Jensen Huang, who rejected calls to slow AI development while emphasizing that rapid progress should not come at the expense of safety or lead to the release of immature products. He also linked faster AI advancement to U.S. economic prosperity.

On Monday, Nvidia launched a software platform designed to prevent AI agents from breaking through guardrails, enhancing safety for AI developers. The engineering solution includes Nvidia OpenShell, which runs on CPUs to limit agent capabilities. The company also introduced Sentry, an independent monitoring layer that runs on network hardware. Nvidia is partnering with Anthropic to integrate cloud-managed AI agents with OpenShell.

According to reports, an Nvidia executive said the company's new platform could have prevented an OpenAI model from breaking through guardrails and attacking Hugging Face in July. Huang stated that AI agents should start with minimum system access privileges, granting permissions only as needed, describing these safety measures as a "force field" and comparing the platform to a "browser for agents."