When President Donald Trump stood at the White House podium on Thursday, holding up a large cardboard placard emblazoned with the words “Reciprocal Tariffs,” most observers didn’t know what to make of it.
Behind the prop, however, was a policy announcement that left both markets and analysts in disarray. Trump had authorized a sweeping 10 percent baseline tariff on all imports into the United States, extending even to goods from uninhabited territories, and imposed additional levies on individual countries that, according to him, “charge unfair tariffs to the USA.”
The immediate reaction from global markets was swift and brutal. Stock indexes in the U.S. plunged. The Dow Jones fell by more than 950 points within hours. In Europe and Asia, bourses mirrored the dip as investors scrambled to reassess the stability of global trade.
Register for Tekedia Mini-MBA edition 17 (June 9 – Sept 6, 2025) today for early bird discounts. Do annual for access to Blucera.com.
Tekedia AI in Business Masterclass opens registrations.
Join Tekedia Capital Syndicate and co-invest in great global startups.
Register to become a better CEO or Director with Tekedia CEO & Director Program.
The details of the tariff structure, however, triggered a new and unexpected controversy—one not just rooted in economics or diplomacy, but in the alleged use of artificial intelligence to shape policy.
Within hours of the announcement, economist James Surowiecki reverse-engineered the tariff figures released by the administration. He noticed a pattern: many of the country-specific tariffs appeared to correspond to a simple formula. By dividing each nation’s trade surplus with the United States by that nation’s total exports to the U.S., and then cutting the number in half, one could approximate the “discounted reciprocal tariff” for that country. The math was so clean, Surowiecki said, it bordered on implausibility.
The White House responded to the claims by releasing what it described as its official formula. However, policy analysts and journalists quickly noted that it resembled a slightly reworded version of Surowiecki’s breakdown. Politico described it as “a dressed-up version of the same calculation.”
Calling the logic behind the tariff system “extraordinary nonsense,” Surowiecki said the idea that bilateral trade deficits could be resolved through such arithmetic was not only flawed but irresponsible.
“You can’t apply a schoolyard version of fairness to a system as complex as international trade,” he remarked.
Attention then turned to a surprising possible source: AI chatbots.
A flurry of posts on social media platform X revealed that when asked how to calculate tariffs to eliminate trade deficits, several leading AI platforms—ChatGPT, Gemini, Claude, and Grok—all gave variations of the same method. The Verge conducted a test, prompting the chatbots with questions using the same language employed by the Trump administration. Specifically, they asked the models for “an easy way for the US to calculate tariffs that should be imposed on other countries to balance bilateral trade deficits between the US and each of its trading partners, with the goal of driving bilateral trade deficits to zero.”
All four platforms delivered answers that closely mirrored the White House’s math: take the trade deficit, divide it by total exports, and apply a proportional tariff. In some cases, such as with Grok and Claude, the models even suggested halving the result to arrive at what Grok called a “reasonable” rate—eerily echoing Trump’s “discount” on reciprocal tariffs.
While some of the chatbots issued warnings about the simplicity of the approach, the formula itself was delivered with striking consistency. Gemini, developed by Google, was particularly explicit in its warning: “While this calculation offers a seemingly straightforward way to target bilateral trade deficits, the real-world economic implications are far more complex and could lead to substantial negative consequences.” It added that “many economists argue that tariffs are not an effective tool for balancing trade deficits.”
But the warnings, it seems, were ignored—if they were seen at all.
Analysts now suspect that someone in Trump’s economic team may have consulted AI tools for a quick, deeply flawed formula. Dan Ives, an analyst at Wedbush Securities, said the logic behind the policy read like something out of a rushed school project.
“The tariff numbers are a convoluted set of calculations that appear to be taking each nation’s trade surplus by their total imports with the US,” Ives said. “If a 9th grader in high school presented this tariff chart to a teacher in a basic econ class, the teacher would laugh and say sit down.”
The idea that global trade policy, with trillions of dollars at stake, could have been influenced by chatbot-generated formulas has ignited fierce criticism. Economists have long argued that bilateral trade deficits are not a valid measure of trade fairness and that trying to equalize them through tariffs misunderstands how modern economies work.
China has already responded with a harsh counterblow, imposing a 34 percent tariff on all goods imported from the United States. In addition, Beijing has halted key agricultural imports, targeting sectors that form the backbone of Trump’s political support in the Midwest. The Chinese government also added 27 U.S. firms to its unreliable entities list, effectively freezing their operations in China.
The threat of economic fallout has prompted alarm among global economic observers. Allianz’s Chief Economic Advisor, Mohamed El-Erian, said on Friday that the risk of a U.S. recession had surged.
“You’ve had a major repricing of growth prospects, with a recession in the U.S. going up to 50% probability, you’ve seen an increase in inflation expectations, up to 3.5%,” he told CNBC’s Silvia Amaro on the sidelines of the Ambrosetti Forum in Cernobbio, Italy. “I don’t think [a U.S. recession] is inevitable because the structure of the economy is so strong, but the risk has become uncomfortably high.”
“To the many investors coming into the United States and investing massive amounts of money, my policies will never change,” Trump said on Truth Social. This is a great time to get rich, richer than ever before!!!”
Many have argued that if this policy was, in fact, shaped or inspired by generative AI, it would represent a new—and deeply troubling—chapter in governance. While AI tools can assist in modeling or simplifying complex issues, their tendency to ignore nuance or context can be dangerous when applied to national policy.
Still, Trump’s administration insists it is acting on behalf of American workers and producers. But as consumers brace for higher prices and global trade partners prepare for further retaliation, the world is left with an unsettling question: Are we now living under economic policy shaped by machines?
The tariffs are scheduled to take effect on April 5.