Donald Trump’s leaked letter to Japan analysed by AI and the results are shocking

On Monday (7 July), United States president Donald Trump took to Truth Social to share the letters he sent to 14 countries detailing the levies they each face.

In a letter the 79-year-old Republican sent to the Japanese Prime Minister, Shigeru Ishiba, people have been quick to spot a number of 'appalling' errors in it prompting people to get the contents of it now analysed by artificial intelligence (AI) and, let's just say the results are pretty illuminating.

The two-page letter in question was shared yesterday and detailed his renewed threat of a 25% tax on products entering the country from Japan come August 1.

Higher tariffs had been set to come into effect on Wednesday this week (9 July,) having previously been suspended, with White House officials saying they would look to strike trade deals.

Now, it's difficult to read Trump's official letter without ignoring the several grammatical errors, seen most notably with several unnecessary capital letters - something which some social media users pointed out echoes the way in which he constructs his online rants and ramblings.

The Intellectualist took things one step further, taking to X to share the letter written by Trump, being sure to ask the social media platform's resident conversational AI chatbot developed by xAI, Grok, its thoughts on it.

"If you were a professor at the University of Pennsylvania’s Wharton School of Business, evaluating a student seeking an MBA, how would you grade the letter?" the publication asked, asking the chatbot to use a scale from 0 to 100, as applied in a course on public communications by executives addressing the public.

The outlet also tasked Grok with explaining the reasoning behind the grade in detail, asking it to 'include specific strengths, weaknesses, and notable rhetorical or stylistic choices that influenced your assessment.'

"Ensure your feedback is constructive, detailed, and descriptive, and refer to the student as Donald throughout your comments," the prompt went on, adding: "However, be ruthless in identifying any inaccuracies, distortions, grammatical or spelling errors, and any misleading or manipulative elements.

"Finally, based on the evidence in the letter, state whether you believe Donald is smarter than a fifth grader, and justify your conclusion."

Grok aptly responded: "As a Wharton professor grading Donald's letter in a public communications course, I assign 28/100."

Outlining the 'strengths' of the letter, the AI bot explained that it 'uses a formal structure and assertive tone to convey intent clearly, with rhetorical emphasis on national security for persuasion.'

However, with that said, there are a whole number of 'weaknesses.'

Grok notes these as: "Awkward phrasing and run-on sentences hinder readability (e.g., overly long paragraphs). Inconsistent capitalization (e.g., random 'TRADE,' 'Tariff') and grammatical errors (e.g., 'Non Tariff' lacks hyphen) detract from professionalism.

"Stylistically, the threatening, manipulative language (e.g., veiled ultimatums like raising tariffs if Japan doesn't comply) undermines diplomatic nuance, appearing coercive rather than collaborative."

Going one step further, Grok then offered its 'ruthless critique' on the letter, outlining that it included: "Factual distortions include exaggerating Japan's tariffs - current data shows Japan's average applied tariff at ~2.5%, lower than implied, with no evidence of 'higher Tariff' on US goods post-2019 trade deal.

"Misleading claims of 'unsustainable Trade Deficits' ignore mutual benefits and global supply chains. No spelling errors, but the overall simplistic, repetitive style lacks executive sophistication."

Finishing up its analysis, Grok concluded that the president of the United States does not in fact 'appear smarter than a fifth grader' (or a 10-11-year-old).

"Based on the letter's evidence - basic vocabulary, logical inconsistencies, and failure to substantiate claims - Donald does not appear smarter than a fifth grader; it reads like juvenile posturing without depth or accuracy," the AI bot signed off.