Grammarly sued over AI feature that passed off advice as coming from scholars without their consent

Grammarly has shut down its Expert Review feature after a class-action lawsuit was filed. The feature presented AI-generated edits as though they came from well-known authors and scholars — without their knowledge or consent. It's the latest case of an AI product crossing a line where a digital persona runs into the rights of living people.

Grammarly is one of the most widely used writing-improvement tools. According to the company, more than 30 million people use its services every day: students, journalists, managers, copywriters. The product started out as a simple spell checker, but in recent years it has grown into a full-fledged AI platform with style, tone, and personalization features.

Expert Review was one of those features. The pitch sounded appealing: instead of getting feedback from a generic algorithm, users would get it in the style of specific recognized authors or scholars. In practice, according to the plaintiffs, the system didn't just draw inspiration from someone's style — it directly attributed the advice to real people, using their names as a mark of quality. The authors and academics whose names were involved didn't know about this use, never gave consent, and received no compensation.

The core legal issue is the so-called right of publicity: a person's right to control the commercial use of their name, likeness, and reputation. U.S. law protects this right well, especially in states like California and New York. Because the suit is a class action, the affected parties have joined forces to press a joint claim — and if the court certifies it as a single case, the potential payouts could be substantial.

Grammarly moved fast: it disabled the feature the same day the lawsuit became public. That limits the immediate reputational damage and signals the company is willing to engage, but it doesn't erase liability for what has already happened. Lawyers note that courts typically weigh a prompt shutdown in the defendant's favor when setting damages, but it does not eliminate liability itself.

The Grammarly case is part of a broader wave. As AI products shift from generic advice to personalized content, they increasingly lean on "real people" as an anchor of trust. Celebrity deepfakes in ads, AI voices of dead musicians, chatbots imitating living influencers — the legal landscape around all of this is still taking shape. Grammarly landed in it unexpectedly: a text-editing feature seems far removed from the world of deepfakes. But the principle is the same — someone else's name is used as an asset without the owner's permission.

For the market, this is a signal: even features that look like harmless UX improvements carry legal risk if they rely on attribution to real people. Before using the name or likeness of any living person — an author, scholar, or expert — as a "label" for AI advice, companies must obtain explicit written consent. Otherwise, the next lawsuit may cost more than whatever they saved by skipping legal review.