Consider these lines from John F. Kennedy's 1961 inaugural address:

"Ask not what your country can do for you—ask what you can do for your country."

"Let us never negotiate out of fear. But let us never fear to negotiate."

"United, there is little we cannot do… Divided, there is little we can do."

"If a free society cannot help the many who are poor, it cannot save the few who are rich."

And from the 1962 moon speech at Rice University:

"We choose to do these things not because they are easy but because they are hard."

And from Lyndon Johnson's inaugural speech:

"John Kennedy's death commands what his life conveyed—that America must move forward."

"On the 20th day of January, in 1961, John F. Kennedy told his countrymen that our national work would not be finished 'in the first thousand days, nor in the life of this administration, nor even perhaps in our lifetime on this planet.' But, he said, 'let us begin.' Today, in this moment of new resolve, I would say to all my fellow Americans, let us continue."

Sound familiar?

Were John F. Kennedy's speeches written by an AI? Probably not. But were AIs trained on the speechwriter Ted Sorensen? It would appear so. Or more accurately, AIs have absorbed the Sorensen toolkit: symmetry, parallelism, antithesis, repetition, and the neatly turned reversal.

Sorensen, who served as Kennedy's primary speechwriter and special counsel, is widely regarded as one of the most influential political prose stylists of the twentieth century. These are iconic lines, and they have been anthologized, quoted, and reprinted across decades of textbooks, essay collections, and online archives—the same corpora that feed large language model training data. In that sense, the Sorensen style entered modern AI not through deliberate emulation but through sheer ubiquity in the written record.

The problem with AI voice is not the techniques themselves but their overuse—the deployment of grand rhetorical flourishes to describe something as mundane as a trip to 7-Eleven with the same language once used to send men to the moon. This recognizable pattern—soaring antithesis applied to trivial subjects—has become one of the most reliable tells that a passage was machine-generated, to the point that style guides and prompt-engineering advice now routinely instruct users to dial back balanced clauses and dramatic reversals.

That, however, is likely to be an easy problem to fix. Labs can adjust system prompts, fine-tune on more varied prose, and offer style controls that flatten the default register. Don't expect AI voice to continue in its current form.

Hat tip: murgh