Grammarly has been accused of dabbling in “necromancy” after academics noticed that the platform has resurrected recently deceased scholars.
A post from German historian Dr. Verena Krebs suggests that Grammarly has appropriated the work of dead historians and academics to help users refine their writing.
While Grammarly is a good tool for non-native speakers who write primarily in English, Krebs said in a post, the historian was alarmed when she uncovered the platform’s new “Expert Review” feature.
Grammarly “now offers to summon colleagues, both living and dead, to “expert review” the piece,” Krebs wrote via Bluesky.
One of the experts suggested to Krebs was British historian David Abulafia, who died in January of this year.
It’s unclear as to whether Abulafia gave his consent to be immortalized in this way, as the feature was introduced in August 2025, just months before his death.
However, a disclaimer from the software company suggests that none of these experts have been given the opportunity to opt in or out.
“References to experts in Expert Review are for informational purposes only and do not indicate any affiliation with Grammarly or endorsement by those individuals or entities,” Grammarly said via its support page.
Cybernews has explored the new feature, which offers tips from living experts such as journalists, economists, and writers.
Although the feature appears useful, various academics have voiced concerns surrounding the ethics of resurrecting dead scholars and using AI to synthesize their life's works.
In a LinkedIn post found by Futurism, associate professor of the history of science and medicine Vanessa Heggie said that dead academics are reviewing users' work “without anyone’s explicit permission.”
Grammarly is “creating little LLMs based on their scraped work and using their names and reputations. Obscene.”
The experts referenced in Grammarly’s new feature are AI agents supposedly trained on publicly available works by the scholars they imitate.
To achieve this, the software company would need to employ bots to scrape the internet for works related to the expert.
The legality of this practice is hazy, and many other companies have faced lawsuits surrounding the use of copyrighted works to train AI models.
Big tech companies like Apple have been sued for using copyrighted content to train their AI.
The Cybernews community is talking about this. Be a part of the conversation.
Two neuroscientists claimed that the tech giant used shadow libraries to illegally access thousands of pirated books to train its Apple Intelligence model, including two books written by the academics.
The ethical implications of using scholars’, writers’, and academics’ work to train AI are severe and have sparked outrage within the community.
“I have seen a lot of cursed stuff in my time in academia, but this is among the *most* cursed,” historian Claire E. Aubin said via Bluesky.
“Grammarly is generating miniature LLMs based on academic work so that users can have their writing ‘reviewed’ by experts like David Abulafia, who died less than two months ago.”
Under the post, the account Litbowl claims that instead of creating “miniature LLMs,” Grammarly is providing the AI with a “persona prompt” which allows the AI to give the user feedback based on the expert's characteristics and their published work.
Persona prompting is a technique used to generate tailored, specific answers to prompts on highly specialized topics.
This not only violates those who did not consent to their personalities being appropriated for profit, but also arguably erases our own personalities as a result of incessant AI use.
Persona prompts and Grammarly’s use of experts, while useful, aren’t adding much value outside of refining the writing in the style of that persona.
While Litbowl said that this practice is “no less gross,” it’s different from creating little LLMs.
Instead, Grammarly’s feature is “both dumber and even weirder” as it seems to be using public summaries and descriptions of the scholars’ work as the foundation for user feedback, Litbowl claims.
Through this, the user alleges an abuse of name, image, and likeness (NIL) on Grammarly’s part.
So, the issue raised by fellow academics still stands, that this is most likely a violation of copyright and is also morbidly unethical when it comes to appropriating the dead to remain relevant.
Unlock more exclusive Cybernews content on YouTube.
Your email address will not be published. Required fields are markedmarked