"Let's break this company apart"! Trump reveals: Considered splitting NVIDIA

Former U.S. President Donald Trump revealed at the AI Summit in Washington on Wednesday that he had considered splitting NVIDIA to increase competition in the artificial intelligence chip market, but later found that "it's not easy to do."

According to reports, Trump stated that his aides told him that splitting NVIDIA was "very difficult" because the company has a significant lead in the AI chip field, and it would take years for other competitors to catch up. Trump then praised NVIDIA CEO Jensen Huang, commending his achievements.

On that day, Trump signed three executive orders and released the "AI Action Plan," emphasizing that the U.S. will "spare no effort" to lead globally in the field of artificial intelligence. The core goal of this "AI Action Plan" is to ensure that the U.S. maintains its dominance in artificial intelligence and provides an environment for American companies to grow and expand rapidly.

Trump stressed that the U.S. must observe the systems of all other countries regarding artificial intelligence. At the same time, the AI competition will require "patriotism," including Silicon Valley, and American tech companies need to put America first (America First).

Several tech industry leaders attended the event co-hosted by the "All-In Podcast" and the "Hill & Valley Forum" to witness the release of this plan. Attendees included NVIDIA CEO Jensen Huang, AMD CEO Lisa Su, and other tech giants.

Considered Splitting NVIDIA

Trump revealed in his speech that he had considered splitting NVIDIA to increase market competition in the AI chip field, but later found that "it's not easy to do."

"I said at the time, 'Let's split this company up,' but then I learned about the realities of the industry."

He stated that his aides told him that doing so was "very difficult" because NVIDIA has a significant lead in the field, and it would take years for other competitors to catch up.

"I thought we could step in, split them up, create some competition, but it turns out this industry doesn't work that way."

Trump then praised NVIDIA CEO Jensen Huang, who was also present at the event. "You are doing a fantastic job," Trump said. Throughout the speech, Trump repeatedly mentioned and praised Huang and other tech industry leaders for their investments in the U.S.

Earlier that day, Huang also praised Trump's stance on artificial intelligence during his speaking segment:

"The U.S. has a unique advantage that no other country can possess, and that is President Trump."

Trump Signs Three Executive Orders

On that day, Trump signed three executive orders related to the "AI Action Plan" to promote its implementation. The directives include: utilizing the U.S. International Development Finance Corporation (DFC) and the Export-Import Bank to support the deployment of U.S. technology globally; another order will require all large language models procured by the government to remain neutral and unbiased.

David Sacks, the White House's head of AI affairs, previewed that there may be further advancements in replacing federal AI policies at the state level in the future. He pointed out:

"We have not formally established relevant policies in this action plan, but I believe this is an issue that must be seriously examined in the next year or two."

Michael Kratsios, Director of the White House Office of Science and Technology Policy, emphasized the role of Congress in this matter. He stated:

"Discussions about federal priorities are largely related to Congress's authority. Therefore, we will not forcefully push in this area, but rather focus on matters we can realistically implement."

Legal Questions Surrounding "Neutrality Measures"

However, constitutional law scholars in the United States have expressed doubts about the legal validity of the "AI neutrality" restrictions proposed by Trump.

Rory Little, a professor at the University of California, San Francisco School of Law, told the media:

"If you punish liberal software but do not punish conservative software, then this executive order is based on content discrimination."

"I don't even know how to determine whether a software is liberal or conservative," Little added, pointing out that the First Amendment of the U.S. Constitution views intellectual property as a form of speech, and the government cannot target it for suppression.

However, the constitutionality of this order may not significantly impact companies like Amazon, Anthropic, Google, OpenAI, Microsoft, and Perplexity, which are competing to provide AI systems to the government in the short term.

Even if the order faces legal challenges, AI development companies may not be able to wait for court rulings.

Little stated:

"Many companies are currently trying to strike deals with the Trump administration, so they do not view these executive orders as laws, but rather as starting points for negotiation."

"If you are an AI company, like Google, you may try your best to negotiate with the government to find a way to continue advancing your business, regardless of the external political atmosphere, as long as your software can make money."

White House Releases "AI Action Plan"

Earlier that day, the Trump administration released the "AI Action Plan," aimed at accelerating the development of artificial intelligence in the United States by easing regulations and expanding energy supply for data centers. The newly released guidelines suggest suspending funding support for states that impose excessive regulations on emerging technologies.

The "AI Action Plan" recommends reforming the licensing approval process and simplifying environmental standards to expedite infrastructure projects related to AI. The blueprint also aims to position U.S. technology as the foundation for global AI.

This 23-page plan was ordered by Trump shortly after he took office in January this year, marking the most significant policy directive from this U.S. administration in the field of artificial intelligence, a technology that could reshape the global economy.

The blueprint reflects Trump's campaign promise to position the U.S. as a global leader in AI while abolishing the Biden administration's overly strict regulatory framework on AI:

The Biden administration issued an order in 2023 that required extensive safety testing and mandated major AI developers to publish transparency reports. Trump, on the other hand, called for a new policy path for AI and set a six-month deadline for David Sacks, the White House AI affairs head, to complete it The new plan suggests that the federal government solicit opinions from businesses and the public to understand the current regulatory policies that hinder AI applications, and use this as a basis to promote the withdrawal of regulations. The White House Office of Management and Budget will also collaborate with federal agencies responsible for AI-related funding, and if a state's regulatory system may affect the effectiveness of grants, the eligibility for funding for that state will be considered for restriction.

The guidelines also require the federal government to contract only with companies that develop AI models "without top-down ideological bias," while also requiring the removal of content related to misinformation, diversity, equity, and climate change from the risk management framework