Grammarly clone your writing style to rewrite text
Grammarly just announced a new feature called "Personalized voice detection and application" rolling out to business subscribers by year's end. The AI analyzes your day-to-day writing to build a custom voice profile, letting you rewrite any draft in your own distinct tone. Why it matters: it makes AI output feel authentic, but opens up tricky questions about identity and style cloning at work.

Grammarly wants to make sure your AI-generated emails actually sound like you.
The cloud typing assistant announced a new feature called "Personalized voice detection and application," slated to arrive for Grammarly Business subscribers by the end of the year. The tool passively analyzes how you write across apps and generates a personal "voice profile" that can rewrite any draft in your signature tone.
Here is how it works and why writers are raising eyebrows.
Learning your habits in the background
You do not need to upload writing samples to train it. Because Grammarly runs across numerous client apps, it builds your profile passively as you communicate, according to Tal Oppenheimer, head of product for Grammarly's client apps.
Each profile comes with an AI description of your style, tagging traits like "positive" or "encouraging." If you disagree with certain grammar quirks it highlights, like a tendency toward compound sentences or active voice, you can manually prune those characteristics.
Grammarly already sees over 12 million pieces of content generated weekly through its AI suite across 30 million active users and 70,000 teams. This feature aims to make that massive volume of text sound less robotic.
The ghostwriting dilemma
The ability to clone a writing voice immediately raises thorny ownership questions.
Critics worry that companies could easily use an employee's voice profile while they are on medical leave, or after they have been laid off, churning out blog posts under their byline without compensation or consent. Phishing impersonation is another clear risk.
These concerns follow real-world fallout: author Jane Friedman recently caught AI-generated books sold under her name on Amazon, while thousands of authors signed open letters and filed copyright lawsuits against AI firms like OpenAI over style replication.
Grammarly is building safeguards for now. Oppenheimer noted that profiles cannot be exported or accessed by anyone else on the team, emphasizing that the feature is meant to help writers understand their tone rather than replace them. Still, skeptics wonder whether Grammarly will face corporate pressure to loosen those boundaries later.
Why it matters
For knowledge workers buried in draft emails, this is genuinely handy. Standard AI output usually requires heavy line-editing to strip out stiff corporate clichés, so a passive voice-matching system could save serious time on routine communication.
We think Grammarly's approach is clever from a product standpoint. By gathering data passively across desktop workflows, it removes the friction of manual prompt tuning that platforms like ChatGPT require.
However, it wades right into the messy debate over who owns a writer's style. Music platforms are negotiating vocal likeness rights and artists are pushing for training royalties, but writers have little legal protection when software clones their phrasing.
Locking profiles to individuals is the right move for launch, but as enterprise customers demand productivity shortcuts, Grammarly will face a delicate balancing act between protecting individual authorship and serving the corporate buyers paying the bills.
Sources (1)
- techcrunch.com(opens in a new tab)Originaltechcrunch.com