Readers discuss what is lost when AI does the initial drafting, animal-assisted services in Hong Kong, and the anti-drug video fiasco
Feel strongly about these letters, or any other aspects of the news? Share your views by emailing us your Letter to the Editor at
[email protected]or filling in this Google form. Submissions should not exceed 400 words.
slopthat floods the internet, such as the AI-generated books people dump on online stores to make a quick buck.
But editing an AI first draft is not the same as writing from scratch. When a top writer builds from the ground up, they make unique choices throughout a text that give it a brand-defining style. Those nuances are lost because large language models predict the most probable next word, pulling prose towards the mean. At best, an AI-generated text is mediocre. At worst, it reads like every other AI text. Because this problem lies at the most fundamental level, surface editing can’t elevate the work. And when companies demand “10x” efficiency, writers feel pressured to accept work that’s merely good enough.
As someone who edits English language learning materials, I know the temptation to put AI in the driver’s seat. The grammar seems flawless. The vocabulary doesn’t feel out of place. For non-native speakers, having AI do the initial drafting can feel like a huge upgrade. But content writers shouldn’t sacrifice their intrinsic voice for a polished first draft. They’ve experienced the second language learning process and can adjust tone, simplify concepts and scaffold language in ways a generic tool can’t replicate. By maintaining this voice, they help define the publisher’s identity.
The value of an authentic voice is exactly why the rush to automate is so risky. If AI makes every company’s writing the same, whether it’s about a product or a service, consumers have no reason to choose one brand over another. Competition shifts from quality to price, creating a race to the bottom.
The solution is simple: let humans write the first draft and use AI to brainstorm, test outlines and polish. It may not mean “10x” productivity, but it preserves originality while still letting companies leverage AI.