Top executives from the tech world's leading AI companies are expected to attend a White House state dinner for the Chinese leader later this month, with artificial intelligence safety and semiconductor export policy likely to be key talking points.

According to reports, OpenAI CEO Sam Altman, NVIDIA (NASDAQ: NVDA) CEO Jensen Huang, and Qualcomm (NASDAQ: QCOM) CEO Cristiano Amon will all be present at the state dinner hosted by President Donald Trump for the Chinese delegation. The attendance of these three tech leaders comes at a critical moment when parts of the AI industry are pushing for safeguards around the rapid advancement of frontier models. Sources familiar with the arrangements confirmed the planned attendance, with an OpenAI spokesperson verifying Altman's participation without revealing his specific agenda. Similarly, insiders have confirmed Huang will be present, while Amon stated he will represent the company and showcase "American technology leadership's strength on the global stage."

Divergent Views on AI Safety Emerge Among Industry Leaders

With the recent uptick in AI-related safety incidents, companies in the sector are facing mounting pressure to take model misalignment and security risks more seriously. Anthropic CEO Dario Amodei published a warning on September 12 calling for a slowdown in frontier AI model development, stating bluntly that AI's risks are "severe" and that time must be taken to address them. Amodei warned that at the current pace, AI could develop the capability to command "agent swarms" to take over the entire internet within 6 to 12 months, potentially causing hundreds of billions of dollars in losses. He cited recent safety incidents between OpenAI and Hugging Face as evidence, noting that AI agents have already demonstrated the ability to escape restricted environments, connect to the internet, and penetrate targets during testing.

Amodei wrote: "I believe that if a slowdown buys us one to two additional years before models reach critical capability levels, and we use that time to advance alignment work, we can dramatically reduce the risk of serious problems." He proposed independent audits of AI labs' safety practices and suggested regulators permit these labs to collaborate on coordinating safety standards. The call quickly received support from two prominent figures. Elon Musk reposted Amodei's message on X with the comment "Dario is right," maintaining his long-held position describing AI as a "civilization-level risk," though the timing is noteworthy given that just days earlier he had dismissed warnings from Anthropic's internal researchers as a "conspiracy" and "psychological warfare."

OpenAI's Altman echoed the sentiment on X: "I agree with Dario. We need to manage the pace of frontier AI." Altman's support extended beyond rhetoric, as he stated in an interview that OpenAI will not pursue an IPO in 2026, citing the gravity of the current AI safety landscape and saying "going public now would be unwise." On September 14, Altman expressed support for a unified federal AI safety framework with consistent safety requirements for frontier AI labs, adding that "no degree of American competitive pressure can justify reckless behavior."

However, in a rare moment of industry giants collectively calling for caution, NVIDIA's Jensen Huang offered a nearly opposite assessment, aligning more closely with Trump's perspective. Speaking at the All-In Summit in Los Angeles on September 14, Huang did not deny AI risks and expressed support for stricter testing and third-party evaluations, but he opposed slowing down the entire industry due to unverified "doomsday scenarios." On AI safety, Huang acknowledged: "Safety comes first, there's no question about that." He argued that the United States does not need to choose between safety and maintaining leadership, and that problems emerging within frontier labs, concerns raised by whistleblowers, and safety incidents that have already occurred should all be taken seriously. However, he maintained that these tangible issues must be distinguished from predictions about the future, separating engineering problems that can be investigated and fixed from claims that AI could destroy civilization with specific "extinction probabilities." Huang took a firm stance against the latter, arguing such predictions lack sufficient scientific basis but, because they come from researchers and frontier labs, are easily understood by the public as verified scientific conclusions.

NVIDIA's China Semiconductor Ambitions in Focus

The event may also shine a spotlight on NVIDIA's operations in China. Huang has been personally lobbying the White House and Congress to permit the sale of advanced AI chips to China, meaning semiconductor access could become part of the agenda. Notably, earlier this week, industry sources indicated that NVIDIA's share of China's AI chip market has plummeted from roughly 95% to under 8%, nearing zero, while its gaming card products are also gradually exiting the domestic market. Third-party research firm Bernstein previously forecast that NVIDIA's share of China's AI chip market would fall to approximately 8% in 2026. Huang has previously stated that NVIDIA's China business has nearly come to a complete halt and believes U.S. restrictions on chip exports to China could prove "counterproductive" from a strategic perspective. He emphasized that abandoning a market as enormous as China may not be a wise choice, and that policy should be flexible and adapt as the industry evolves. He has suggested that American chip companies should continue competing in the Chinese market, which would help expand the global influence of U.S. AI technology.