Grammarly, the writing tool that once relied solely on proofreading for correct grammar and spelling, has added several generative AI features over the years. In October, CEO Shishir Mehrotra announced that the overall company was rebranding as Superhuman to reflect a new suite of AI-powered products. However, the AI writing “partner” is called Grammarly. “When technology works everywhere, it starts to seem normal,” Mehrotra wrote in his press release. “And that usually means there’s something extraordinary happening under the hood.”
The expanded Grammarly platform now offers an AI solution for every imaginable need – and some you probably never had before. There’s an AI chatbot that will answer specific questions while creating a draft, there’s a “Paraphraser” feature that suggests style changes, there’s a “Humanizer” that modifies according to the selected voice, there’s an AI grader that predicts how your document will score as college coursework, and there’s even tools to flag and tweak phrases typically produced by larger language models. (Sure, you’re using AI to do everything here, but you don’t want that sound Like that.)
Perhaps most disingenuously, however, Grammarly now has an “Expert Review” option, which, instead of looking like a generic critique from an anonymous LLM, lists a number of actual academics and writers available to weigh in on your text. To be clear: those people have nothing to do with this process. As a disclaimer makes clear: “References to experts in this product are for informational purposes only and do not indicate any affiliation with Grammarly or endorsement by those individuals or entities.”
As advertised on a support page, Grammarly users can ask for suggestions from living authors and scholars such as Stephen King and Neil deGrasse Tyson (neither of whom responded to requests for comment), as well as virtual versions of the dead such as editor William Zinsser and astronomer Carl Sagan. Presumably, these various AI agents are trained on the actions of the people they seek to mimic, although the legality of this content-gathering still remains questionable, and is the subject of many, many Copyright lawsuits.
“Our expert review agent examines the writing the user is working on, whether it’s a marketing brief or a student project on biodiversity, and leverages our built-in LLM to surface expert content that can help the document writer shape their work,” says Jane Dakin, senior communications manager at Superhuman. “Suggested experts depend on the substance of the writing being evaluated. The Expert Review Agent does not claim endorsement or direct involvement with those experts; it provides suggestions inspired by the experts’ works and points users toward influential voices whose scholarship they can explore in more depth.”
Someone like King might consider AI progress unstoppable, and there would be no one left to defend Zinsser’s 1976 handbook. on writing well From the big tech vultures, but what about the countless other giants who still want to protect their content from being compressed into an algorithm? Vanessa Heggie, associate professor of the history of science and medicine at the University of Birmingham, recently shared on LinkedIn a particularly egregious example of how the feature works, accusing superhumans of “creating little LLMs” based on the “cuts” of people living and dead and trading on “their name and reputation.” Screenshots he posted showed the availability of an AI agent’s analysis based on David Abulafia, an English historian of the medieval and Renaissance periods who died in January. “Obscene,” Heggie wrote.
<a href