In 2016, tariff man couldn’t care less about tech. Newly elected US President Donald J. Trump knew that the people who created and ran America’s tech giants were richer and smarter than him. Moreover, they had different values. They embraced openness, accountability and the rule of law in international affairs. Moreover, they generally advocated for increased immigration of skilled workers, open markets and relatively unfettered cross-border data flows. So, not surprisingly, his first term was characterized by disputes with many tech giants, including Facebook and Amazon.
But tech leaders and Trump administration officials had a common concern — and “enemy” — China’s innovative tech firms. They deemed Chinese companies’ growing prowess in data-driven technologies as a threat to America’s economic progress and national security. As Chinese companies (most visibly TikTok) gained market share, the Trump administration came to understand that protecting the competitiveness of data-driven sectors was imperative if America were to become “great again.” And in February 2019, Trump signed Executive Order (EO) 13859 on Maintaining American Leadership in Artificial Intelligence.
The EO created the American AI Initiative to “focus the resources of the Federal government to develop AI in order to increase our Nation’s prosperity, enhance our national and economic security, and improve quality of life for the American people.” Moreover, his administration set the initial rules for US government use and governance of AI. Finally, the first Trump administration thought the US government could best counter China’s growing competitiveness in key technologies such as AI by limiting access to the essential components of the AI supply chain. Many of these policies were quietly lauded by America’s diverse tech sector.
Ironically, it was US President Joe Biden who really brought this vision of limiting access to the AI supply chain to life by issuing the AI Bill of Rights and two EOs on AI, and by working with other countries to reduce AI risks and govern AI collectively. He took many steps to secure American AI leadership, investing in chip factories, STEM (science, technology, engineering and mathematics), education and research. Yet simultaneously, his administration challenged the market power and business practices of the tech giants. Such actions alienated many in the venture capital and tech sectors. Although many tech leaders were quiet, and several other prominent CEOs actively supported Vice President Kamala Harris, a significant number of tech leaders and venture capitalists supported Trump in the 2024 election. Under Trump 2.0, I foresee deep divisions among the tech companies that are courting the new administration’s officials. Here are a few predictions as to what
Trump 2.0 might do.
End Efforts to Advance Trustworthy AI
“Trustworthy AI” refers to the development, acquisition and use of AI in a manner that fosters public trust and confidence while protecting privacy, civil rights, civil liberties and American values, consistent with applicable laws.
During his first term, Donald Trump signed an EO that defined the building blocks of trustworthy AI as efforts to ensure the validity and reliability of AI models, their security and resilience, model accountability and transparency, model explainability and interpretability, privacy and fairness by developing strategies to mitigate bias. Many of these building blocks cannot be achieved without some form of AI or business regulation.
However, Elon Musk, Trump’s second-biggest donor, is determined to limit AI regulation. Moreover, the Republican platform promised to repeal Biden’s EOs that attempted to further trustworthy AI at home and abroad. The party sees these EOs as undermining innovation. Leaders of other tech companies may disagree, as the EOs have provided clarity on the use of AI and AI development and governance.
Maintain the Biden Administration’s International Efforts to Ensure that AI Is Trustworthy, Safe and Resilient
Earlier in his career, Musk expressed support for preventing many AI risks, including the establishment of the AI Safety Institute, located at the Department of Commerce. Musk’s main stated concern is existential risk, whereas the leaders of many other companies are concerned about more immediate risks such as discrimination or hallucinations.
However, the AI Safety Institute was established in 2023 by Biden’s EO. It may not survive without Congress approving legislation to create such a body. My gut feeling is that most tech CEOs like the AI Safety approach because it is built on testing and standards, rather than on mandates.
End the Biden Administration Efforts to Challenge the Business Practices of Data Giants
Under Trump, companies may be given freer rein to merge and invest in smaller, more innovative companies at home and abroad. Most of the giant companies will be grateful, but emerging companies such as Reddit or smaller AI companies such as Hugging Face may be frustrated.
Under Trump 2.0, restrictions on data flows are likely to increase. However, the leaders of giant tech companies are divided as to the short- and long-term implications of these supply chain restrictions.
Expand Efforts to Limit the Ability of Adversaries to Acquire US Data
Trump signed the first EO on a key element of the AI supply chain related to data flows. In October 2024 alone, the United States issued several rules or asked for public comment on restrictions on data broker sales of US data sets to US adversaries, as well as restrictions on investment and expertise.
Under Trump 2.0, restrictions on data flows are likely to increase. However, leaders of the giant tech companies are divided as to the short- and long-term implications of these supply chain restrictions. While the United States has so far limited these EOs to adversaries, this also means reduced access to data and less accurate, complete and representative data sets. Moreover, they undermine America’s commitment to digital solidarity, which the US Department of State defines as “a willingness to work together on shared goals, to stand together, to help partners build capacity, and to provide mutual support.” It seems unlikely that Trump 2.0 will maintain efforts to foster digital solidarity.
Update the Language of America’s Digital Trade Agreements and Draft and Sign New Ones
The signatories of the United States-Mexico-Canada Agreement (USMCA) agreed to review the agreement six years after its entry into force and decide whether to extend the agreement another 16 years. The review will take place in 2026, and the Trump administration may view bolstering it as a useful demonstration that the United States is not always protectionist or isolationist. It could even push for renegotiating various chapters.
The USMCA included a digital trade chapter that is out of date and, if renegotiated, could be used as a model for future trade agreements with other partners. Should the Trump administration as well as Mexico and Canada decide to renegotiate the chapter, it could engender significant support among not only the tech giants but also the many American farmers and small and medium enterprises who also benefit from cross-border data flows. The Biden administration’s unwillingness to negotiate was widely criticized by business, as well as by Republican and Democratic members of Congress. Despite their protectionist rhetoric, Trump administration trade officials may not arouse protectionist opposition and could reap significant business community support by negotiating other digital trade agreements with key markets such as the United Kingdom and European Union.
Rethink America’s Commitment to Open as Well as Proprietary Foundation Models
In 2023, US government agencies investigated the risks of open versus closed models. Both the US Federal Trade Commission and the US Department of Commerce found these risks to be manageable, and also asserted that openness provided great benefits to society and the companies.
However, firms such as Open AI that rely on proprietary models may challenge these findings, presenting another possible divide within America’s tech sector.
Further Divide the Tech Community on Data and AI Sovereignty
US government agencies have quarrelled with countries and trade blocs such as India and the European Union that insist that data must be sovereign, which means it can only be stored and analyzed on trusted domestic cloud platforms. Now the United States is becoming more like these nations when it asserts sovereignty over data and restricts its flows. Some American companies are even marketing AI and data sovereignty to policy makers in many nations to capture market share. For example, executives at high-speed chipmaker NVIDIA have encouraged governmental clients to adopt sovereign AI through its “AI nations” initiative. Microsoft and Amazon Web Services, which provide cloud services as well as AI foundation models, also benefit from domestic efforts to promote AI. Meanwhile, other companies see the danger (and protectionism) in such approaches. Taken in sum, these policies may lead to a further clash among the tech bros.
These predictions may or may not come true, but one thing is certain: whatever the Trump administration does to nurture and govern the tech sectors, its actions will likely divide tech companies, Americans and the world.
This article first appeared in Fortune.
